THE PSYCHOLOGY

OF SECRETS
THE PLENUM SERIES IN
SOCIAL/CLINICAL PSYCHOLOGY
Series Editor: C. R. Snyder
University of Kansas
Lawrence, Kansas

Current Volumes in the Series:

ADVANCED PERSONALITY
Edited by David F. Barone, Michel Hersen, and Vincent B. Van Hasselt
ANATOMY OF MASOCHISM
June Rathbone
CULTURES, COMMUNITIES, COMPETENCE, AND CHANGE
Forrest B. Tyler
HANDBOOK OF SOCIAL COMPARISON
Theory and Research
Edited by Jerry Suls and Ladd Wheeler
HUMOR
The Psychology of Living Buoyantly
Herbert M. Lefcourt
PERSONAL CONTROL IN ACTION
Cognitive and Motivational Mechanisms
Edited by Miroslaw Kofta, Gifford Weary, and Grzegorz Sedek
PHYSICAL ILLNESS AND DEPRESSION IN OLDER ADULTS
A Handbook of Theory, Research, and Practice
Edited by Gail M. Williamson, David R. Shaffer, and Patricia A. Parmelee
THE PSYCHOLOGY OF SECRETS
Anita E. Kelly
THE REVISED NEO PERSONALITY INVENTORY
Clinical and Research Applications
Ralph 1. Piedmont
SOCIAL COGNITIVE PSYCHOLOGY
History and Current Domains
David F. Barone, James E. Maddux, and C. R. Snyder
SOURCEBOOK OF SOCIAL SUPPORT AND PERSONALITY
Edited by Gregory R. Pierce, Brian Lakey, Irwin G. Sarason,
and Barbara R. Sarason

A Continuation Order Plan is available for this series. A continuation order will bring
delivery of each new volume immediately upon publication. Volumes are billed only
upon actual shipment. For further information please contact the publisher.
THE PSYCHOLOGY
OFSECRETS

ANITA E. KELLY
University of Notre Dame
South Bend, Indiana

SPRINGER-SCIENCE+ BUSINESS MEDIA, LLC

ISBN 978-1-4613-5193-1 ISBN 978-1-4615-0683-6 (eBook)
DOI 10.1007/978-1-4615-0683-6
©2002 Springer Science+Business Media New York
Originally published by Kluwer Academic/Plenum Publishers in 2002
Spftcover reprint of the hardcover 1st edition 2002

AII rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted
in any form or by any means, electronic, mechanical, photocopying, microfilming,
recording, or otherwise, without written permission from the Publisher.
PREFACE

On an MTV special aired in 2000, young interviewees were asked to

confess the worse thing they were ever told during a romantic breakup.
One person tearfully responded "that I suck in bed." More recently, an
acquaintance of mine admitted to his new girlfriend that he "has a mean
streak." She decided not to date him after that. Another memorable and
painful example of openness occurred years ago when I served as a
member of a suicide intervention team. I was called to a very disturbing
scene in an upscale neighborhood to console a woman who was threaten-
ing to take her life on the lawn in front of her children. Her husband had
just confessed his long-term affair to her that morning and she felt that her
world was coming apart. Fortunately, she did not take her life but was left
with the humiliation of haVing her neighbors know about her private
troubles. The question these examples bring to mind is, "Why do people so
often reveal potentially stigmatizing personal information to others?"
The reader probably has an intuitive answer to this question already.
It can seem like such a burden-even torture-to keep secrets from other
people. Hiding such things as feelings of discontent from a boyfriend or
girlfriend, violations of the law from close friends, and indiscretions from
employers can be alienating. People want others to know them; therefore
they often end up disclosing self-incriminating information. At the same
time, however, people do not want to be reminded of the things that make
them feel bad or ashamed. What can be done to alleviate this burden of
secrecy? Will this anguish pass in time? What are the factors that should
enter into making a wise decision to reveal a secret? These are three of the
central questions addressed in this volume.
Psychotherapists dating back to Freud (1958) have promoted the reve-
lation of therapy patients' secrets in treatment, no matter how humiliating
v
vi PREFACE

or excruciating their secrets might be. In fact, such openness was the
essence of Freud's Fundamental Rule of Psychoanalysis. Leaders of the
encounter group movement in the United States in the 1960s, too, advo-
cated getting one's secrets-as well as angry and painful feelings-out in
the open when confronting interpersonal difficulties (see Bok, 1983). Such
developments were precursors to the present-day emphasis that thera-
pists, talk show hosts, and televangelists place on the cathartic benefits of
revealing secrets. Talk show hosts like Jenny Jones and Jerry Springer
almost daily elicit detailed accounts of sexual molestations, marital infi-
delities, and other painful private events on television. Revealing what
were once considered extremely private matters to the media is becoming
a new social norm (Krestan & Bepko, 1993). It is estimated that over 21
million viewers follow television talk shows that regularly encourage
participants to reveal their deepest family indiscretions (Mitchell, 1993).
One of the purposes of this volume is to question whether this movement
toward openness has gone too far and to examine the negative conse-
quences of revealing personal secrets. Another purpose is to look at when
and why revealing secrets can be extremely helpful, leading to health
benefits such as improved immunological functioning (e.g., Petrie, Booth,
Pennebaker, Davison, & Thomas, 1995).
My interest in secrets began informally about 14 years ago when I
observed the lack of discretion that some of my acquaintances and fellow
psychology trainees seemed to have. They revealed others' secrets-even
their therapy clients' secrets-at a rate that made me question whether I
should reveal any private information to them myself. I knew several
people who participated in mental health groups such as Al-Anon and
Recovery. They would describe the details of meetings to me, including the
names of fellow members, even though they acknowledged that the
groups were supposed to be anonymous and had taken an oath of confi-
dentiality as part of their membership. Moreover, as part of my doctoral
training in psychology, I participated in a practicum support group that
was supposed to be confidential. I happily unburdened my secrets to my
trusted fellow group members only to discover later at least one of the
other group members had revealed my secrets to people outside the
group!
A little later in my training, I took great interest in social psychological
research and was especially influenced by the work of Barry Schlenker,
Roy Baumeister, and Dianne Tice on how a person's public self-portrayals
can influence his or her private self-conceptions. Along these lines, it
disturbed me that I had allowed others to know personal things that I did
not want to keep in mind about myself. After all, once people have formed
certain opinions of their peers-as my friends may have formed of me-
PREFACE vii

they tend to give the peers feedback that constrains them to being consis-
tent with those impressions (e.g., Swann, 1996). It follows that people
should not reveal to close others what they do not want to remember about
themselves.
Another key, more formal observation occurred when I was conduct-
ing a study on secret keeping in outpatient therapy (Kelly, 1998). I asked
42 clients (who had received an average of 11 therapy sessions) to indicate
what, if any, relevant secrets they were keeping from their therapists. As it
turned out, just over 40% of them said that they were keeping such a secret.
What was surprising was that when I asked them to indicate (on a 9-point
Likert-type scale) how stressful it was to keep those secrets, many of them
indicated that it was not at all stressful to keep the secrets from their
therapists. This observation contradicted much of the published theoreti-
cal work on the stressful nature of secret keeping and made me wonder
whether secrets may be burdensome at first but may become less so after
people have kept them for some time. This idea is developed further in the
last few chapters of this volume.
My perspective as a theorist and researcher is that informal observa-
tions such as those I have just described can serve as an informative
starting point for theory development and subsequent research. Then, as
the theory is formally tested using the research methods available to
psychologists and as new data support or fail to support the theory, it must
be revised accordingly. In essence, what began as my informal observa-
tions about the lack of discretion of my peers has led to my making more
formal observations and theorizing about when people should and should
not reveal their secrets. My theoretical model hinges on the features of the
confidant, which are described in the last two chapters of this volume.
In contrast to this philosophical approach to conducting research,
prominent psychology researchers have argued that because psychology
in general (see Kagan, 1994) and counseling psychology in particular (see
Hill, 1982) are so young, researchers should avoid overtheorizing and
should attend only to the emerging facts from formal research. Kagan
(1994) wrote, "Although I respect theory, I share Percy Bridgman's suspi-
cion of a priori ideas, especially in young sciences like psychology. Too
many investigators persuade themselves of the correctness of an idea
through thought alone" (p. 117). However, my view is that researchers
virtually always have preconceived ideas about how their studies will tum
out, and I think it is better to articulate what these ideas are and then
scrupulously revise these ideas based on the new evidence that emerges
(see Strong, 1991, for a similar argument). Even though psychology is a
young science and the formal study of secrecy is especially recent, there
are many facts already available to help people predict, understand, and
viii PREFACE

control the outcomes of their decisions to reveal their secrets. My purpose

in this volume is to provide these facts in a manner that is held coherently
together by theory. At the same time, however, I have been careful
throughout this volume to point out the holes in the available data (partly
the result of researchers' having paid more attention to studying self-
disclosure than to studying secrecy).
I aim to identify the processes that may underlie the benefits or
drawbacks to revealing personal secrets across many settings involving
many different kinds of people. As such, I often rely on experiments
conducted in the laboratory, even though they may be criticized for their
artificial nature. Because experiments allow cause-and-effect inferences to
be drawn, they lend more confidence to identifying the processes underly-
ing the phenomena being studied; in this case, the effects of revealing
secrets. Experiments are emphasized in this volume because as Mook
(1983) so aptly put it, "experiments give us an understanding of real world
phenomena ... because the processes we dissect in the laboratory also
operate in the real world" (p. 156).
This is the first authored volume to provide a comprehensive descrip-
tion of the psychological benefits and drawbacks to revealing personal and
potentially stigmatizing secrets. There also is an edited volume available,
The Secrets of Private Disclosures (edited by Sandra Petronio) published in
2000, that has valuable chapters on the implications of disclosing HIV
status and on gender and cultural differences in disclosure presented from
a communications perspective. My volume differs greatly from that one in
the sense that I have offered an integration of ideas from one consistent
theoretical perspective of this growing literature on secrets, starting with
redefining the nature of secrecy. I draw from clinical and social psychologi-
cal research, including findings from my own research program, to explain
why people keep secrets, what it is about revealing secrets that leads to
health benefits, and when revealing secrets can backfire, even in psycho-
therapy sessions. I also present a theoretical model for making a wise
decision to reveal a secret: This decision hinges on whether one's confidant
is likely to react positively to the revelation.
The interest in empirically studying secret keeping is a very recent
phenomenon. In fact, many of the studies on secrecy that I cite in this
volume have been conducted only in the last few years. This interest has
taken off with James Pennebaker's (e.g., 1989) work on the theoretical
dangers associated with inhibiting the expression of traumatic events or
emotional upheavals and with Daniel Wegner's (e.g., 1992) work on the
paradoxical effects of the thought suppression that goes into keeping
information secret. Also, the development of the Self-Concealment Scale to
measure one's tendency to keep negative or distressing information from
PREFACE ix

others (Larson & Chastain, 1990) has spawned a series of studies assessing
how high self-concealers differ from low self-concealers on various mea-
sures of well-being. Because the interest in studying secret keeping is so
recent, there is a growing (but still not huge) literature on the effects of
secret keeper per se. Thus, there are times when I draw from research on
related constructs, such as research on the inhibition of emotional expres-
sion, repressive coping styles, and inhibited temperal types, to fill in the
gaps concerning how secrecy might affect our interpersonal relationships
and physical and mental health. I also discuss the research on the benefits
of writing about traumas, as opposed to writing about secrets per se,
because often traumas are kept secret from at least some other people.
In Chapter 1 I define secrecy and its related constructs and describe
the types of secrets people keep and why they keep them. In Chapter 2 I
examine the evidence that people who do keep secrets-as compared with
those who do not-tend to be less physically and psychologically healthy.
In Chapter 3 I explain why secrecy has been linked to such problems and
suggest that secret keeping per se may not be problematic. In Chapter 4 I
look at the very compelling evidence that revealing secrets in anonymous
and confidential settings leads to health benefits. In Chapter 5 I offer an
explanation for why revealing secrets leads to such health benefits. In
Chapters 6-7 I discuss the effects of revealing secrets to one's therapist in
particular and offer a new self-presentational theory of psychotherapy to
explain why high levels of clients' revelation in therapy are not associated
with positive outcomes. In Chapter 8 I describe the dilemmas to revealing
secrets to other people and illustrate how introducing different types of
confidant feedback into the revealing process can dramatically alter the
outcomes of revealing. In Chapter 9, the final chapter, I offer recommenda-
tions for when one should and should not reveal secrets. It is my hope that
this final chapter will help secret keepers sort out the important factors in
making a wise decision to reveal their secrets. Perhaps they can live better
with decisions that are made consciously and deliberately, as opposed to
unconsciously or passively.

ACKNOWLEDGMENTS

I thank the National Institutes of Health and the Graduate School of

the Univeristy of Notre Dame for supporting my research between 1997
and 1999. I also thank Mercedes Kelly and Pascal Lavallee for their in-
sights, my graduate and undergraduate research assistants for their hard
work, and C. R. Snyder for encouraging me to write this volume.
CONTENTS

CHAPTER 1
The Nature of Secrecy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Defining Secrecy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
How Common Is Secrecy? ............................... 9
Types of Secrets People Keep ............................ 10
VVhyPeopleKeep Secrets ....... ...... .......... ......... 15
Rules of Revealing Secrets ............................... 16
VVhy People Reveal Secrets .............................. 17
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

CHAPTER 2
Individual Differences in Secret Keeping . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Correlates of Unconscious Inhibition and the Repressive
Coping Style ....................................... 24
Correlates of Conscious Inhibition ........................ 31
Profile of a Secret Keeper ................................ 38

CHAPTER 3
Why Secrecy Is Linked to Problems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Inhibition Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Preoccupation Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
xi
xii CONTENTS

Self-Perception Model ................................... 54

Diminishes Social Support ............................... 56
A Predispositional Explanation. . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

CHAPTER 4
Health Benefits of Revealing 67
Revealing Traumatic versus Trivial Events ................. 69
Revealing to a Psychotherapist versus Revealing Privately . . . 72
Writing versus Talking .................................. 74
Revealing Previously Disclosed versus Undisclosed Events .. 75
Revealing Real versus Imagined Traumas.. . .... . . . ... . .... 76
Healthy Undergraduates versus Clinical Samples . . . . . . . . . . . 77
Meta-analyses .......................................... 78
Summary.............................................. 78
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

CHAPTER 5
What Is It about Revealing Secrets That Is Beneficial? 81
New Insights ........ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Catharsis .............................................. 85
New Insights versus Catharsis ........................... 88
Study 1 ................................................ 89
Study 2 ................................................ 91
Discussion ............................................. 96
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

CHAPTER 6

Secrecy and Openness in Psychotherapy . . . . . . . . . . . . . . . . . . . . . . . . .. 101

Theoretical Perspectives on the Role of Clients' Openness ... 103
Empirical Findings on the Role of Clients' Openness in
Therapy ........................................... 117
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 126
CONTENTS xiii

CHAPTER 7
Why Openness May Not Be Therapeutic: A Self-Presentational View
of Psychotherapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 129
Self-Presentation Research ............................... 132
Self-Presentational View of Client Change ................. 141
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 157

CHAPTER 8
Dilemmas to Revealing Secrets and the Role of the Confidant 161
Negative Consequences of Revealing Secrets. . . . . . . . . . . . . .. 164
Positive Consequences of Revealing Secrets. . . . . . . . . . . . . . .. 174
Consequences of Revealing May Depend on the Confidant.. 178
Features of Helpful Confidants ........................... 181
Optimal Number of Confidants .......................... 185
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 185

CHAPTER 9
When to Reveal Personal Secrets in a Particular Relationship ........ 187
Proposed Model for When to Reveal. . ..... ... .. . ... . ... .. 191
Options to Revealing the Complete Truth. . . . . . . . . . . . . . . . .. 203
Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 209
Summary .............................................. 214
Testing the Model with Sample Secrets. . . . . . . . . . . . . . . . . . .. 214
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 216

References .................................................. 219

Index ...................................................... 253

CHAPTERl

THE NATURE OF SECRECY

"I once killed for love." It was this late-night confession in August 1996, to
two of her college roommates at the Naval Academy, that led Diane
Zamora to a sentence of life in prison with no possibility for parole for at
least 40 years. Zamora and her ex-fiance, David Graham, were convicted
of bludgeoning and shooting Adrianne Jones to death in 1995 (when all
three were in high school together). The two had lured Jones to a remote
area in Texas to kill her after Graham told Zamora he had slept with Jones.
Zamora later confessed to the police that when Graham admitted to hav-
ing had sex with Jones, Zamora had flown into hysterics, screaming, "Kill
her, kill her!" (People Magazine, March 4, 1998, p. 67).
Stories such as this one are both disturbing and perplexing. In addi-
tion to the horror of wondering why someone would calculatingly "kill
for love," they leave one wondering, Why did Graham tell his emotionally
unstable fiancee that he had had sex with an extremely attractive rival?
Perhaps more perplexing, why did Zamora tell her roommates and at least
two other people, including her best friend, that she was involved in a
murder? Later, at her trial, she tried to put a defensive spin on her pre-
vious confession to the police, saying that it was the domineering Graham
who forced her to unwillingly partake in the murder. But it was too late to
take back her confessions to so many people, including her own recorded
words to the police.
The first question of why Graham confessed his affair to Zamora is
perhaps a little easier to address. Marriage and family therapists have
long believed that it is essential for a marital partner who has had an af-
fair to confess to the other spouse (e.g., Brown, 1991; Pittman, 1989; Shlien,
1984) and their rationale also could apply to unmarried couples. Presum-

1
2 CHAPTER 1

ably, it is not the extrarelational sex that causes problems, but rather the
secrecy surrounding the sex (Pittman, 1989). Keeping such a secret requires
omitting truths or telling lies, and this deception may undermine a per-
son's sense of self-worth (Shlien, 1984). The guilt and shame of the secret
affair may become unbearable for the person who is having the affair
(Shlien, 1984). Some marital therapists have even asserted the secret extra-
relational affairs should be revealed even if they were brief and happened
a long time ago because "hiding it means holding back a piece of oneself"
(Brown, 1991, p. 138). This explanation could capture what Graham might
have been experiencing: To continue to feel close to Zamora, he may have
felt compelled to confess and clear his guilty conscience.
Zamora's motivations for confessing the murder to her roommates
seem a bit more puzzling, because she had so much to lose by confess-
ing and seemingly so little to gain. When the police had found Jones's
body 9 months earlier, they declared the murder unsolved. Zamora must
have known that her roommates were bound by both criminal and mili-
tary laws to reveal such a confession to the authorities. If Zamora had
just kept her secret, she would likely have been on her way to achieving
her dream of becoming an astronaut by now, instead of being imprisoned
for life.
Was her motivation based on feeling guilty about the murder? The
answer to that question seems to be "no." One of the roommates, Jennifer
McKearney, testified in Zamora's trial that when McKearney asked Za-
mora if she had any regrets about the murder, Zamora had said, "No, it
had to be done" (People Magazine, March 2, 1998, p. 68). McKearney said
that Zamora had claimed that anyone who got between herself and
Graham would have to die, and that Jones was "a tramp and a slut" who
deserved what she got. Perhaps Zamora felt that it was too much pressure
to keep a secret that big to herself. Or perhaps as a midshipwoman, she
took pride in her fierceness and daring in having taken the life of another.
She may have been boasting and enjoying the attention that she received
from such a bold, unique confession. It also is possible that even though
she did not feel guilty about the murder, she did feel guilty about not
letting her closest friends know such an important secret. People will
probably never fully grasp the answers to these questions in this case, but
researchers have begun to address the question of whether people have
some sort of primitive or compelling need to confess emotional experi-
ences to others (e.g., Rime, Mesquita, Philippot, & Boca, 1991a). In this
chapter, I describe this research and explain why people sometimes reveal
even very damaging personal secrets. But first, I define secrecy and de-
scribe the types of secrets people keep and why they keep them.
THE NATURE OF SECRECY 3

DEFINING SECRECY

Keeping secrets means to hide deliberately information from other

people (Margolis, 1974). Secrecy also has been called "active inhibition of
disclosure" (e.g., Pennebaker, 1989) and one's predisposition to engage in
secrecy has been labeled "self-concealment" (Larson & Chastain, 1990).
Secrecy can be experienced as burdensome and stressful (see Pennebaker,
1989,1990), and as such it is not merely the opposite of self-disclosure (i.e.,
revealing personal information to others) (see Collins & Miller, 1994).
Keeping secrets requires expending cognitive and emotional resources
(see Lane & Wegner, 1995; Wegner, 1989, 1992, 1994; Wegner, Lane, &
Dimitri, 1994), whereas simply not self-disclosing information does not
involve such efforts. Factor analyses of measures of both self-concealment
and self-disclosure have shown that self-concealment is empirically dis-
tinct from self-disclosure (Larson & Chastain, 1990).
As implied by the definition above, secrecy always involves at least
two people. Secret relationships can involve only one person's knowledge
of the relationship, such as in the case when someone has a secret sexual
desire for a married friend or coworker (Wegner et aI., 1994). But even in
such cases, the very existence of the secret depends on the secret keeper's
awareness of another person. People work to hide their secrets only if
other people who are not supposed to know the secrets are around. For
example, imagine a teenager who would feel comfortable telling his clos-
est male friends about his exotic pornography collection but would be
horrified if his mother found out about it. He would expend energy
keeping the pornography hidden when she is present but not when the
friends are present (unless he thinks they might tell her about the pornog-
raphy). Hence, when discussing the essence of secrets, it only makes sense
to think of them in terms of "keeping secrets from whom?"
It might seem obvious to many readers that the most crucial defining
characteristic of secrecy-and the biggest determinant of its effects-
should be the type of secret that a person is keeping. For example, keeping
secrets about having committed a murder presumably would be more
effortful and stressful, and thus more harmful, than keeping secrets about
having smoked marijuana. However, it is the process of expending energy
to keep information from other people that defines secrecy, not the type of
secret. This book is written with this definition in mind and is about any
personal information that people work to keep from others. For example,
some people might view their admitting to having tried cocaine as actually
enhancing their group's opinions of them (i.e., the group might see such
experimentation as a sign of daring). Of course, other people, such as
4 CHAPTER 1

George W. Bush, might see such an admission as an extremely risky act

that could have cost a presidential election. In this example, the type of
personal information is the same-trying cocaine-but in the latter case,
it is a secret; and in the former case, it is not.

SECRECY VERSUS PRIVACY

In understanding the nature of secrecy, it is important not only to

keep in mind that secrets always have a social context but also to distin-
guish secrecy from privacy. Whereas privacy connotes the expectation of
being free from unsanctioned intrusion, secrecy does not. Secrecy involves
active attempts to prevent such intrusion or leaks, and the secret keeper
exerts this energy in part because he or she perceives that other people
may have some claim to the hidden information. For example, the hygiene
rituals that people engage in on a daily basis are considered to be private in
American society, because people agree that these practices are not for
public display. However, having a hidden dangerous contagious disease
would be considered a secret because society has a right or obligation to
intervene in such cases. Also, the hidden rituals that members of various
cults or religious sects engage in may be considered secret rather than
private, because members of those sects are aware that the broader society
does not necessarily expect such rituals to be confidential.
The distinction between secrecy and privacy is not a trivial one. It
helps to explain why there has been so much controversy surrounding
whether people who work in hospitals and other medical facilities or
people who are about to marry should be tested routinely for HN infec-
tion. In both cases, the public or another person's safety may be threatened
by keeping such information secret, and thus people may feel that they are
entitled to have access to that information. At the same time, the individual
who is being tested for HN may feel that he has a right to keep his medical
records private, especially if he plans to use prophylactic devices at all
times when treating patients or having sex (see Cline & McKenzie, 2000;
Yep, 2000, for further discussion of the dilemmas to disclosing HIV-
positive status).
What may be private information in one relationship may be secret
information in another. For example, in some new romantic relationships,
both partners may explicitly or implicitly agree that each may have sex
with other partners as long as they use condoms. They may have an
understanding that those other sexual encounters are private matters that
do not need to be discussed. However, in relationships that are intended to
be monogamous, such as the one between Graham and Zamora, both
partners may expect full disclosure about any extrarelational activity. They
THE NATURE OF SECRECY 5

also might expect disclosure about any previous sexual relationships, and
any deliberate omission of such information would be considered a secret.
Even if one of the partners viewed such information as private be-
fore the relationship started, the fact that he is aware that she expects
revelation might cause him to exert energy in hiding that information from
her. This expenditure of energy occurs because in deciding that this infor-
mation should be kept a secret, he must constantly monitor information
that is consistent with the state of mind that he wishes to maintain as well
as monitor the information that he wishes to hide from others (Wegner et
al., 1994). It is not easy for people to engage in these dual processes: "The
secret must be remembered, or it might be told. And the secret cannot be
thought about, or it might be leaked" (Wegner et al., 1994, p. 288). In
a nutshell, whether information is merely private or is secret depends
on the expectations that the people in a particular relationship have
about what should be disclosed. This distinction between secrecy and
privacy is central to understanding the essence of secrecy and its contex-
tual nature.

SECRECY VERSUS REPRESSION

As indicated earlier, by definition secrecy is a social phenomenon.

But what about times when people are keeping secrets from themselves? Is
that a form of secrecy? The concept of keeping secrets from oneself and
not even realizing that one is engaging in such censorship has been called
repression. Whereas self-concealment is defined as the "predisposition to
actively conceal from others personal information that one perceives as
distressing or negative" (Larson & Chastain, 1990, p. 440), the term "re-
pression" has been used to mean keeping painful thoughts and impulses
out of conscious awareness (Schwartz, 1990). People who deny symptoms
and anxiety and who score high on measures of social desirability (i.e.,
they try to look like good people) have been referred to as having a
repressive coping style (Weinberger, Schwartz, & Davidson, 1979). Al-
though both self-concealment and repression are seen as types of defen-
sive coping strategies or ways of avoiding dealing with unpleasant mate-
rial (Spielberger et al., 1991), the main distinction between self-concealers
and repressors is that the former engage in conscious secret keeping, where-
as the latter engage in unconscious censorship of undesirable material.
Whereas the empirical study of self-concealment is relatively new
(e.g., Larson & Chastain, 1990), repression and the broader concept of
inhibition have received a good deal of attention since Freud's early theo-
rizing about these concepts (see, for example, King, Emmons, & Woodley,
1992; Polivy, 1998). It is interesting that repression has received more re-
6 CHAPTER 1

search attention than self-concealment, given that repression is more diffi-

cult to study because researchers cannot directly assess unconscious pro-
cesses. This greater emphasis on repression is perhaps a testament to the
contribution of Freud's work on the unconscious aspects of personality in
general and on repression in particular, which he depicted as the corner-
stone of all defensive mechanisms.
Repression is conceptually but not necessarily empirically related to
self-concealment. Spielberger et al. (1991) investigated the relationship
between these two defensive coping strategies by asking German college
students to complete the Self-Concealment Scale (SCS) (Larson & Chas-
tain, 1990), the Marlowe-Crowne Social Desirability Scale (SD5-CM), and
the Taylor Manifest Anxiety Scale (MAS) (Luck & Timaeus, 1969). (The
combination of the latter two scales is a measure of the repressive coping
style.) The scores on the SCS were positively correlated with scores on the
MAS for both men and women, demonstrating that self-concealers report
being high in anxiety. However, the repressors reported being low in
anxiety. Based on the results from these self-report measures, Spielberger
et al. (1991) suggested that self-concealment and repression tap different
aspects of defensive coping. Likewise, other researchers have demon-
strated that self-concealment is related to self-reported depression (e.g.,
Kelly & Achter, 1995; Larson & Chastain, 1990), whereas repression is not
(e.g., Zung & Gianturco, 1971). For example, Zung and Gianturco (1971)
observed a strong negative correlation between self-reported depression
and repression among 159 patients seen in an outpatient clinic, and thus
they concluded that the depressed patients did not repress or deny their
illnesses and that they were sensitive to their symptoms.
However, it does not really make sense to use self-report measures to
compare repressors with self-concealers, because by definition repressors
deny being anxious or distressed, even though they might actually be
distressed. Detecting differences between patients and controls with re-
spect to inhibition of expression must be done instead with overt behavior
and physiological assessments (Traue & Michael, 1993). For instance, when
Anderson (1981) exposed patients with psychosomatic disorders to physi-
cal and mental stress, the patients who had the strongest physiological
responses actually reported that they had experienced less stress than
those who had weaker physiological responses. Researchers have shown
that repressors, as compared with high-anxious or low-anxious partici-
pants, report fewer negative emotions in diary ratings (Shapiro, Jamner,
& Goldstein, 1993) and fewer negative childhood memories (Davis &
Schwartz, 1987). Repressors also report being older during their early
negative memories (Davis & Schwartz, 1987; Myers & Brewin, 1994). Even
though repressors seem to have trouble remembering early negative
THE NATURE OF SECRECY 7

events, this trouble does not generalize to their recall of positive events
(Myers & Brewin, 1994).
In sum, repression and self-concealment are conceptually related-
both are seen as forms of avoiding threatening material-but they are
not necessarily empirically related. Repression historically has received
more empirical attention than self-concealment, even though it is more
difficult to study. A way around the problem of trying to assess repression
directly has been to consider people to be high in repressiveness (or high
in repressive-defensive coping) when they score high on social desirability
measures and report having low levels of anger or anxiety (Weinberger, 1990).
These people may believe and act as though they have few problems, but
they may actually be engaging in some kind of unconscious self-deception
or denial (Weinberger, 1990). Thus, the major distinction between self-
concealment and repression seems to be that the former involves suppres-
sion, which is conscious, and latter involves an unconscious avoidance of
threatening information (King et al., 1992).

SECRECY VERSUS LYING

As implied earlier, secrecy typically carries the negative connotation

of being a form of deception (e.g., Lane & Wegner, 1995). Whereas lying is
deception by commission, secrecy has been described as deception by
omission (e.g., Wegner et al., 1994). For instance, imagine a 40-year-old
woman who cheated on her taxes when she was earning a lot of money at
the age of 17 waitressing at an upscale restaurant. If she tells her husband,
"I have never cheated on my income taxes," then she is simply lying.
However, if she says, "I have never in all my adult life cheated on my
income taxes," then she is keeping the adolescent cheating a secret from
him. Either way, her intent is to mislead him about her record of integrity.
But if she chooses to omit the truth rather than directly lie and if he ever
happens to find out about the cheating, then she always can claim that she
was merely protecting her privacy by not telling him about it.
An interesting note about secrecy is that concealing information from
others may seem less reprehensible than actively lying to them because it
is a passive process (Ekman, 1991). Gesell (1999) found that a sample of
undergraduates did indeed rate secrets as less immoral than lies. Perhaps
people feel that secrecy is not as bad as lying because when they have been
deceived solely by omission, they can still feel that if they put a direct
question to the deceptive person, the person could be relied on to give a
truthful answer (or to decline to answer, and then the truth can be inferred
from the silence). In contrast, once people recognize that someone has lied
to them, they may feel that they can no longer believe anything that the
8 CHAPTER 1

person says. Even admitting to one friend about having lied to another
friend can make the first friend wonder about the discloser's integrity.
However, there are likely to be cases when a person actually may feel
more angered by concealment than by lying (see Sweetser, 1987). This
anger stems from the fact that the target cannot complain about being lied
to and may feel as though the deceiver got by on a technicality (Sweetser,
1987). A now-familiar example of this type of reaction to concealment was
the Independent Counsel Ken Starr's outrage when former president Bill
Clinton refused to answer questions about the precise nature of his inti-
mate contacts with Monica Lewinsky. Clinton denied in the Paula Jones's
deposition that he and Ms. Lewinsky had had a "sexual relationship."
Clinton maintained that none of his sexual contacts with Ms. Lewinsky
constituted "sexual relations" within a specific definition used in the
Jones's deposition. Under that definition, a person engages in "sexual
relations" when the person knowingly engages in or causes (I) contact
with the genitalia, anus, groin, breast, inner thigh, or buttocks of any
person with an intent to arouse or gratify the sexual desire of any person.
"Contact" means intentional touching, either directly or through clothing.
According to what the President testified was his understanding, this
definition" covers contact by the person being deposed with the enumer-
ated areas, if the contact is done with an intent to arouse or gratify," but
it does not cover oral sex performed on the person being deposed. He
testified: "[I]f the deponent is the person who has oral sex performed on
him, then the contact is with-not with anything on that list, but with the
lips of another person. It seems to be self-evident that that's what it is ....
Let me remind you, sir, I read this carefully" (Starr Report, 1998).
The reason Clinton's testimony was so enraging to many Americans
was that it was obvious his intention was to mislead them into believing
that he had no sexual relationship with Monica Lewinsky when in fact he
did have one. Perhaps a testament to the notion that secrecy is less repre-
hensible than lying is that fact that Clinton's job approval ratings stayed
high throughout the impeachment hearings. Also, the US Senate did not
find Clinton guilty of perjury, and thus did not force him to step down
from the preSidency. Later, during his final days in office, in a plea bargain
to avoid further prosecution, Clinton finally admitted that his intent in-
deed was to mislead those putting him on trial as well as to mislead the
American public. Right after this confession, his approval ratings plum-
meted and he received a resurgence of scrutiny of his actions as president.
In particular, he fell under attack by prosecutors over his final-hour presi-
dential pardons, especially the pardoning of billionaire Marc Rich. Accord-
ing to Crystal Champion (2001), there may be a trade-off associated with
concealment and lying, such that concealment may make the recipient
THE NATURE OF SECRECY 9

angrier about the deception but lying may do more damage to the de-
ceiver's overall reputation as a moral, decent person. Clinton's decision
finally to confess about the Lewinsky affair may have made people less
angry about his sneaky omissions at trial, but it may have hurt him in
terms of his overall reputation.

HOW COMMON IS SECRECY?

Even though researchers have not yet charted the frequency of se-
crecy in people's everyday experiences, undergraduates participating in
a recent survey study perceived secrecy and lying to be equally common
(Gesell, 1999), and lying has been found to be a very common occurrence
(DePaulo, Kashy, Kirkendol, Wyer, & Epstein, 1996). In two separate diary
studies, 77 college students and 70 demographically diverse members of
the community kept track of how many times per day they told lies
(DePaulo et al., 1996). The students reported that they lied about two times
per day on the average and the community members reported telling one
lie per day (DePaulo et al., 1996). The participants in those studies tended
to tell relatively more other-oriented lies to women, such as "your hips
don't look big in that skirt" in an effort to spare the women's feelings. In
contrast, the participants tended to tell men more self-centered lies, such
as "I am not attracted to your old girlfriend" in an effort to protect the liar's
interests.
I contend that secrecy defined as deliberately keeping information
from others-but not necessarily all others-is even more common than
lying. It is such a common phenomenon that virtually all adults of normal
intellectual and psychological functioning do keep personal secrets at one
time or another. Anecdotal evidence for this suggestion comes from my
own laboratory research on secrecy. Nine years ago, when I first started
this line of research, I prescreened undergraduate participants for those
who reported that they were keeping a personal secret, thinking that I
would only ask those who had one to take part in my studies. However,
I soon learned that such prescreening was not necessary. Virtually all the
students who came into the laboratory without such prescreening could
generate a personal and private secret that they had told no one or very
few people. Likewise, in studies where large groups of people were asked
to recall an emotional experience that they have never shared the majority
said that they readily could recall such an experience (Rime, Philippot,
Boca, & Mesquita, 1992). Also, Vangelisti (1994) found that 99% of her
sample of undergraduates reported that they were keeping a secret from at
least one of their family members.
10 CHAPTER 1

Theorists have argued that secrecy may be considered a critical ele-

ment in the development of one's ego boundaries or sense of identity (see
Hoyt, 1978; Margolis, 1966; Taush, 1933). Learning about society's taboos
(e.g., not to blurt out embarrassing observations, such as "Mommy, that
man is missing a leg!") and learning to keep such information to oneself
are thought to be central aspects of healthy development (Szajnberg, 1988).
Developmental psychologists have noted that as children mature, they
learn to conceal information from others and to use such information to
influence or manipulate them (Peskin, 1992). In one study, 3-year-olds
were not yet capable of strategically concealing information from others
in order to obtain a desired object (a preferred sticker), whereas 5-year-
olds were able to conceal information to obtain the object (Peskin, 1992). In
sum, secrecy is a nearly universal phenomenon, and being able to keep
personal secrets may even be seen as a sign of maturation.

TYPES OF SECRETS PEOPLE KEEP

People tend to keep their most embarrassing, disturbing, or traumatic

personal experiences concealed. Secrets are likely to involve negative or
stigmatizing information that pertains to the secret keepers themselves
(Norton, Feldman, & Tafoya, 1974). These kinds of secrets are stigmatizing
personal secrets (Kelly & McKillop, 1996) and they are the focus of this
volume. There is a wide range of the types of personal secrets people keep,
which includes secrets about being gay (Cole, Kemeny, Taylor, & Visscher,
1996) or lesbian, being raped (Binder, 1981; Burgess & Holmstrom, 1974),
cheating on a test (Kelly, Klusas, van Weiss, & Kenny, 2001), and having
medical conditions, such as AIDS (Larson & Chastain, 1990). Across a
number of studies, one category of personal secrets emerges the most
commonly occurring: sexual secrets (e.g., Hill, Thompson, Cogar, & Den-
man, 1993; Kelly, 1998; Kelly et al., 2001; Norton et al., 1974). In the follow-
ing paragraphs, I provide more details on the kinds of secrets that people
in both nonclinical and clinical samples have reported keeping and the
kinds of topics couples and families consider taboo (i.e., off limits for
discussion).

UNDERGRADUATES' SECRETS

Norton et al. (1974) examined the frequency of themes for secrets

written anonymously by 359 undergraduates in an encounter group exer-
cise. Sex-related secrets occurred most often, followed by failure-related
secrets. The researchers then drew a random sample of 49 secrets from
THE NATURE OF SECRECY 11

the original 359 secrets and asked 190 other undergraduates to rate the
levels of risk associated with the potential discovery of those secrets.
Secrets ranged from the risky, "I have had incestuous relations with a
member of my family" to the nonrisky, "1 smoke dope." The secrets that
were perceived as the most risky were those relating to sex, mental health,
and violence or destruction.
In another study of secrets among undergraduates, my students and
I invited 85 undergraduates to our laboratory where, after completing a
stream-of-consciousness writing task, the undergraduates were asked to
write confidentially and anonymously about the most personal and pri-
vate secrets of their lives (Kelly et al., 2001). Specifically, they were asked to
think of a secret that they had told no one or very few people. Among the
85 participants, 28 listed sexual secrets (e.g., about having been raped,
experimenting with novel sexual acts, having had sex for money); 17
described secretly desiring or having a romantic relationship; 12 had fam-
ily secrets (e.g., having been neglected as a child, having an unemployed
father); 10 described interpersonal alienation; 7 listed secrets about death
or suicide (e.g., their own suicide attempts); 4 listed delinquencies (e.g.,
cheating on a test); 3 listed an addiction/bulimia; 3 indicated abortion/
pregnancy; and 1 described a health problem. Thus, once again it was
sexual secrets that were most frequently mentioned.
Instead of asking participants to generate their own secrets, Lane and
Wegner (1995) gave undergraduates a list of 50 topics and asked them to
rate to what extent they kept those topics secret. The researchers then
subjected the ratings to factor analysis and found that four distinct factors
emerged. They labeled these "offenses," "worries," "sorrows," and "sins."
Offenses included stealing things and masturbation; worries included
getting mugged and failing a test; sorrows included being lonely; and
sins included using marijuana and thoughts of the devil. The last two
categories could be distinguished from the other categories in that sins
referred to personal moral weaknesses, whereas offenses were socially
disapproved of acts and often punished by society. The extent to which
participants reported keeping the various events secret was found to be
greatest for sorrows, next for offenses, next for sins, and least for worries.
Finkenauer and Rime (1998a) tried to capture the defining characteris-
tics of secrecy by exploring the differences between secrets and nonsecrets
in two studies. In the first study, undergraduates recalled the most recent
important emotional event that they had shared with another person and
the most recent important emotional event that they had not shared. The
researchers focused their comparisons of the ratings of shared and non-
shared events only on the negatively valenced events (i.e., they dropped
the responses of participants who reported positive secrets). The second
12 CHAPTER 1

study was almost identical to the first, except that the participants in the
second study were asked to recall either a shared or a nonshared event, not
both. Also, participants in the second study were college students and
their acquaintances who ranged in age from 15 to 75 years. For both
studies, the researchers hypotheSized that the emotional intensity, type of
emotion (i.e., shame and guilt), and holding back from expressing emo-
tions during the events would all discriminate between the secret and
nonsecret events. Also, they predicted that keeping an emotional event a
secret would make the participants ruminate over (i.e., mentally hold
onto) the event more than if they had shared the event. The researchers
based their predictions in part on the findings from a pair of surveys in
which Wegner et al. (1994) showed that (1) past relationships that partici-
pants presently thought about were more likely to have been kept secret
than ones that they did not think about and (2) participants who reported
that a past relationship had been kept secret indicated that they were still
preoccupied with thoughts of the relationship.
Contrary to Finkenauer and Rime's (1998a) predictions, in both
studies the participants' ratings of the overall intensity of their emotional
experiences failed to differentiate between shared and nonshared events.
In Study I, the participants indicated that the nonshared events were
actually less negative than shared ones. Moreover, participants did not
report that they ruminated more over nonshared events than over shared
ones. Likewise, there were no differences in participants' ratings of how
traumatic the events were and no differences in the extent to which they
felt recovered from the event and its consequences. However, the type of
emotion surrounding the events did distinguish nonshared events from
shared ones. For nonshared events, participants reported greater shame,
disgust (in Study 2 only), and guilt; increased appraisal of personal re-
sponsibility; and more holding back from emotional display during the
event. They also reported that they had engaged in a greater search for
meaning and efforts to understand what happened in the cases of non-
shared events (Finkenauer & Rime, 1998a). Thus, the studies uncovered
some expected differences in participants' past attempts to understand the
two types of events but failed to uncover differences in how much partici-
pants currently ruminated over or were bothered by the events.

CLIENTS' SECRETS

Hill et al. (1993) asked 26 psychotherapy clients who had received an

average of 86 therapy sessions to indicate what if any secrets they had not
told their therapists. Secrets were defined as major life experiences, facts,
or feelings that clients do not share with their therapists that occur over a
THE NATURE OF SECRECY 13

relatively long time frame and do not necessarily stem from events within
the therapy (Hill et al., 1993). Surprisingly, even though the clients were
selected by their therapists for participation in the study, and even though
the clients had paid a good deal of money for the therapy, nearly half of
them (46%) reported keeping secrets from their therapists (Hill et al., 1993).
The themes of the clients' secrets encompassed sex, failure, and mental
health issues (Hill et al., 1993). For example, one of the clients reported that
he was keeping his gay attraction toward the therapist a secret.
The themes to clients' secrets discovered by Hill et al. (1993) were
similar to the themes described by encounter group and group therapy
members in previous research (Norton et al., 1974; Yalom, 1985). Yalom
(1985) reported that the three most common themes of secrets for group
therapy members were: (1) a deep conviction of personal inadequacy, (2) a
sense that they do not or cannot truly care for another person, and (3) some
type of sexual secret.
Whereas both Hill et al. (1993) and Yalom (1985) asked clients about
their secrets in general, I (Kelly, 1998) asked 42 short-term psychotherapy
outpatients to indicate whether they were keeping secrets from their thera-
pists that they believed were relevant to their treatment. The outpatients
in this sample had received an average of 11 therapy sessions (ranging
from 3 to 30) at a community mental health hospital. They completed
confidential surveys that would never be seen by their therapists. Just over
40% of the outpatients said that they were keeping relevant secrets from
their therapists. Consistent with Hill and colleagues' (1993) findings on
keeping general secrets from the therapists, whether they were keeping
relevant secrets was not related to how many sessions they had received.
The types of relevant secrets that the 17 clients reported were as follows
(they each listed only one secret, as it turned out): 7 said that they secretly
were desiring the wrong person/had secret relationship difficulties; 4
indicated sexual secrets; 2 listed a health problem; 2 listed drugs / alcohol; 1
indicated lying/ delinquency; and 1 said that he would take his secret to
the grave. Hence, the relevant secrets from this sample of clients were
similar to the secrets from other clinical and nonclinical samples: they most
frequently were about sex and sexual desires.

TABOO TOPICS IN CLOSE RELATIONSHIPS

Related to the concept of secrecy are taboo topics-topics that are

considered off limits for discussion-in relationships. Baxter and Wilmot
(1985) interviewed 40 male and 50 female undergraduates and asked them
to comment on taboo topics in their cross-sex relationships. Twelve of
the relationships were platonic, 25 were romantic potential (i.e., they were
14 CHAPTER 1

not yet romantic but could become so), and 53 were romantic. The most
frequently mentioned taboo topic, representing 34.4% of the total number
of taboo topics mentioned, was the state of the relationship. The state of
the relationship referred to how serious the romantic relationship was or to
whether the platonic relationship would become romantic. Next was ex-
trarelationship activity (15.7%), followed by relationship norms (12.8%),
prior relationships (12.8%), conflict-inducing topics (11.0%), and negative
self-disclosures (4.7%), such as having been arrested for shoplifting in
the past.

FAMILY SECRETS

Secrets have been of particular interest to family systems researchers,

largely because when families come in for therapy, secrecy is thought to
be the source of the symptoms displayed by one or more family members
(e.g., Saffer, Sansone, & Gentry, 1979). Vangelisti and Caughlin (1997) asked
two large samples of undergraduates to recall and describe a secret that
their family, as a group, kept from nonfamily members. The results
showed that the types of family secrets they described could be cate-
gorized as taboo topics, activities that are stigmatized or condemned by
the family and the larger society; rule violations, which break rules common
to many families; and conventional topics, which represent information that
is not necessarily wrong but often is considered inappropriate for discus-
sion (Vangelisti & Caughlin, 1997). Examples of taboo secrets were the lack
of a previous marriage (when there was supposed to be one), substance
abuse, finances, sexual preferences, mental health of family members, and
physical abuse. Examples of rule violations were premarital pregnancy
and cohabitation. Finally, examples of conventional secrets were physical
health problems of family members, circumstances surrounding a family
member's death, and personality conflicts among family members. Taboo
secrets were the most common, with 75.4% and 74.1% of the participants
in the two samples mentioning those; rule violations were mentioned by
14.7% and 9.5% of the participants; and conventional secrets were men-
tioned by 9.9% and 16.4% of the two samples, respectively.
Even though it is interesting to read what kinds of secrets family
members keep, predicting whether a member will keep a secret does not
seem to be enhanced by knowing what type of secret it is (Vangelisti &
Caughlin, 1997). Rather, such predictions are enhanced by looking at the
functions of the secrets and at the relationships between the secret keepers
and the people to whom the secrets pertain (Vangelisti & Caughlin, 1997).
In particular, Vangelisti and Caughlin (1997) found that neither the valence
of the secrets nor the extent to which family members identified with the
THE NATURE OF SECRECY 15

secrets predicted whether they were likely to reveal it to outsiders. How-

ever, if the secrets served evaluation, maintenance, privacy, or defense
functions, family members were unlikely to reveal them (Vangelisti &
Caughlin, 1997).

SUMMARY

A number of typologies of secrets have emerged from research on

undergraduates, psychotherapy clients, and families. Cutting across the
different typologies, one can see that people most frequently report having
sexual secrets and keeping secret their desires for (or involvement in) a
romantic relationship. People also report keeping secrets that could make
them look maladjusted: the mental health category from the Hill et al.
(1993) study, the deep conviction of personal inadequacy sentiment ex-
pressed by group therapy members (Yalom, 1985), the sorrows category
from the Lane and Wegner (1995) study, and the failure-related events
category presented by Norton and colleagues (1974) all suggest that type of
secret. As far as what seems to distinguish secrets from nonsecrets, Finke-
nauer and Rime (1998a) found support for the idea that people are inclined
to keep secret events that they feel guilty about or ashamed of or that they
feel are disgusting. However, they also found that people do not neces-
sarily ruminate more over secrets as compared with nonsecret events.

WHY PEOPLE KEEP SECRETS

Having just identified the types of secrets people keep, I now pinpoint
the reasons people give for keeping secrets. The 17 clients who reported
keeping a relevant secret in psychotherapy were asked, "What has pre-
vented you from sharing your relevant secret(s) with the therapist?"
(Kelly, 1998). Five clients said that they were afraid to express feelings,
three stated that they were too ashamed or embarrassed, three were
concerned that revealing the secrets would show the therapist how little
progress had been made, three stated that there was no time, two said that
they would not tell anyone, two said that they were not motivated to
address the secret, one indicated loyalty to another, and one said "noth-
ing." In Hill and colleagues' (1993) study of the long-term therapy sample
of clients, the most frequently listed reason for keeping secrets was that the
clients felt too ashamed or embarrassed to share their secrets with their
therapists. Regarding taboo topics, the primary reason respondents gave
for not wanting to discuss the most frequently mentioned topic-the
nature of the relationship-was "relationship destruction" (Baxter &
16 CHAPTERl

Wilmot, 1985). Specifically, both the more- and the less-committed partners
feared that articulation of the unequal commitment levels would force the
partners to recognize the inequity, and thus break up.
Consistent with these findings, a number of scholars have argued that
people keep secrets because of their concerns about receiving disapproval
from others (e.g., Bok, 1982; Larson & Chastain, 1990; Simmel, 1950; Stiles,
1987; Wegner & Erber, 1992). A potential outgrowth of such disapproval is
being abandoned by confidants who have a negative reaction to a given
revelation (see Kelly & McKillop, 1996). It would seem that underlying
any decision to keep (or to reveal) a secret may be what Baumeister and
Leary (1995) identified as a fundamental human motivation to belong to a
community or group. Baumeister and Leary (1995) hypothesized that
people need to form and maintain strong, stable interpersonal relation-
ships and postulated that this need is for frequent, nonaversive inter-
actions within an ongoing relational bond. They offered substantial sup-
port for their hypothesis by showing that people form social attachments
readily under most conditions and resist the dissolution of existing bonds.
Moreover, once people do adjust to the loss of a partner, they typically do
quite well with a substitute romantic partner who fulfills their belonging
needs. Baumeister and Leary (1995) also showed that a lack of attachments
is linked to a variety of ill effects on health, adjustment, and well-being.
They concluded that existing evidence supports the hypothesis that the
need to belong is a powerful, fundamental, and extremely pervasive mo-
tivation that appears to have multiple and strong effects on emotional
patterns and cognitive processes. It follows from their conclusions that
people would go to great lengths to avoid revealing secrets that might
leave them ostracized from their support network. Consistent with this
idea are Norton and colleagues' (1974) findings that secrets relating to sex,
mental health, and violence or destruction were perceived by a group of
undergraduates as the most risky statements-those secrets might cost
them precious relationships.

RULES OF REVEALING SECRETS

Now I address the flip side of concealing and describe the rules of
revealing secrets, along with people's motives for revealing them. Based
on their qualitative analysis of 24 secret revelations, Rodriguez and Ryave
(1992) concluded that the revealing of secrets follows an organized se-
quence of events. The first step is the secret frame, wherein the revealer
indicates that a secret is coming and indicates who should not be told. The
second step involves the confidant's accepting the secret nature of the
THE NATURE OF SECRECY 17

revelation and indicating that he or she can be trusted not to reveal it. The
third step is the disclosure of the contents of the secret. The fourth and
final step is the confidant's acknowledgment of having received and un-
derstood the secret. The first step in this sequence is similar to what
Bellman (1981) termed the preface, which is the set of instructions a revealer
gives to the confidant just prior to revealing the secret. Such instructions
serve to inform the confidant that the secret must not be repeated. They
also assure the confidant that the circumstances are appropriate for the
revealing, and they unite the revealer and confidant. Last, they assure the
confidant that the revealer can be trusted even though he or she is sharing
a secret. Bellman (1981) notes that a contradiction arises when the secret
that is being revealed pertains to someone else. Specifically, the confidant
is being asked to keep a secret when the revealer himself or herself did not.
This seeming contradiction can be resolved by considering another
rule to the transmission of secrets. People may expect to be told secrets that
their closest friends know about individuals who are outside their imme-
diate social circle. At the very least, they are likely to view less negatively
revelations about acquaintances than about close friends. Yovetich and
Drigotas (1999) found that college students were more likely to communi-
cate private information gathered from a lower-level intimate to a higher-
level intimate (upward transmission) rather than from a higher-level inti-
mate to a lower-level intimate (downward transmission) in both imagined
(Study 1) and actual (Study 2) instances of secret transmission. In addition,
observers (Study 3) evaluated the revelation of a secret more negatively
when an individual passed the secret from a higher-level intimate to a
lower-level intimate. As these findings imply, people may excuse their
closest friends for telling them other people's secrets.

WHY PEOPLE REVEAL SECRETS

Given that revealing personal secrets may make people look bad, as
described earlier, why do people so often end up revealing their secrets to
others? Derlega and Grzelak (1979) described five motives for revealing
secrets and suggested that some combination usually accompanies the
revealing. These are self-clarification, social validation, relationship devel-
opment, social control, and expression. Self-clarification refers to the re-
vealer's desire to acknowledge his or her position. For example, a woman
might say, "Even though I am Republican, I am clearly pro-choice and
had to have an abortion myself when I was 19." Revealing secrets based on
wanting social validation is an effort to confirm one's sense of self-esteem.
Relationship development motives often are spawned by hearing some-
18 CHAPTER 1

one else's revelations and wanting to reciprocate to enhance an equitable,

intimate bond. People who reveal secrets to establish social control are
attempting or hoping to control the behavior of another by expressing dis-
approval of their actions. For instance, a man might tell his girlfriend who
is considering a breast augmentation procedure that he left his previous
girlfriend after she had that operation. Finally, the motive for expression is
experienced as a compelling desire to talk about feelings or thoughts.
The motive for expression recently has received a good deal of atten-
tion in the clinical and social psychological literatures, with some re-
searchers arguing that disclosing personal emotional experiences to others
is a compelling need (e.g., Tait & Silver, 1989). In a number of studies in-
volving both positive and negative emotional experiences, people have
reported sharing the experiences with others, most of the time sharing the
experiences repetitively and a short time after the experience (for reviews,
see Rime et al., 1992; Rime, 1995a,b). Rime et al. (1992) found that for about
90 to 96% of the emotional experiences people had, they reported having
disclosed the experiences to others. Rime and colleagues (Rime et al.,
1991a; Rime, Noel, & Philippot, 1991b) also found that participants shared
their emotionally intense experiences, as compared with less emotionally
intense experiences, with others more frequently and over an extended
period of time, a pattern that was replicated in diary studies in which
participants kept track of their most emotional events of each day (Rime
et al., 1995). The findings also were replicated in situations where people
were contacted after some important emotional experience, such as the
birth of a child or a car accident, and then again weeks later (Rime et al.,
1995). In addition, when participants were induced to feel emotions in the
laboratory, they shared the emotional experiences with others when they
were given the opportunity to do so (Luminet, Bouts, Delie, Manstead, &
Rime, 2000). These findings on expression of emotion pertain directly to
expression of personal secrets to the extent that the secrets are charged
emotionally.
Beyond describing the different motives for revealing, researchers
have offered deeper explanations concerning why people reveal their emo-
tional experiences to others. One explanation stems from the cognitive-
motor view of expression (Rime & Schiaratura, 1991). According to this
view, critical parts of one's experiences are encoded or retained at a non-
verbal level in the form of mental images, bodily movements, and affect-
related visceral changes (such as a churning stomach or racing heart).
These nonverbal forms remain the focus of attention until they can be
assimilated and put into words, particularly when the experiences are
more emotionally intense. Another idea is that people experience emotion
when their anticipations of how the world should operate are disrupted.
THE NATURE OF SECRECY 19

If the disruption is intense enough, it may challenge a person's basic

assumptions about the self and the world (Janoff-Bulman, 1992). Such a
person should be motivated to interact with others as a means of helping
to confirm or disconfirm beliefs about the self and reconstructing assump-
tions about the world.
In both of these theoretical explanations, revealing is depicted as a
way of making meaning out of emotional events, which is consistent with
the commonly recognized Zeigarnik effect. Zeigarnik (1927) showed that
people continue to think about and remember interrupted tasks more than
finished ones, suggesting that they may have a need for completion or
resolution of the events. Thus, people may reveal their secrets to others in
order to try to get a sense of resolution on the secret.
These explanations for revealing make sense regarding emotional
secrets, but they cannot account for the times when people choose to reveal
a secret that is likely to cost them a relationship even though they do not
necessarily need resolution on that secret. For example, imagine the fol-
lOWing situation. A woman says to a potential dating partner, "I would not
date a man who has ever had sex with a prostitute .. , but other than that
I don't care how many sexual partners he has had." Imagine, too, that he
once did have sex with a prostitute but had never considered that informa-
tion to be relevant to any new relationship. Paradoxically, now, just when
the revelation might be most damaging to him, he may feel the most
compelled to confess it. Thus, on his third date with the woman, he blurts
out, "Actually, I have had sex with a prostitute, but it happened when I
was just a teen." This perplexing confession stems from the fact that, as
discussed earlier, private information can become secret information in the
context of particular relationships. Just knowing that the woman would
reject him for a revelation about his experience with a prostitute might
make him focus on that information and cause him to suppress it. Ironi-
cally, at least initially, suppression makes information actually come to
awareness more readily, particularly when the person is under stress
(Wegner & Erber, 1992). If the woman said that she did not care about what
he did before, then probably none of his specific sexual acts would have
come to mind to be suppressed, and therefore he would not have become
preoccupied with those thoughts and then blurted out the damaging
confession (see Wegner, 1994; Lane & Wegner, 1995).
Certainly, he could have kept his experience with the prostitute a
secret and dismissed his date's comments as too intrusive to be addressed.
However, since the humanistic, encounter group movement of the 1960s,
many Americans have come to expect openness about things such as
previous sexual and assault histories when building new relationships
(see Bok, 1982). Because of the cultural context, he may have felt that he
20 CHAPTERl

should "come clean" with her to start things off and may have felt obli-
gated to tell her about the prostitution. Otherwise, the belief is that what
closeness he develops with her will be limited if she cannot completely
accept him and what he has done.
In addition to the expectations for disclosure of one's relational part-
ner, I contend that two other important elements go into whether one feels
the need to disclose secret information. The first of these issues is the extent
to which the secret keeper feels that the secret is central to his or her
identity. If the man who had sex with a prostitute felt that this particular
behavior is part of who he really is (i.e., it is part of his self-concept), then
he might feel that it is important to reveal such information in an effort to
be truthful with his new girlfriend. However, if he feels that the behavior
was a rare exception for him, say, for example, that he was pressured to do
it at a birthday party that happened a long time ago, then he may actually
feel pleased to get away with keeping that information a secret. The
interesting paradox here is that people may feel that their self-descriptions
better represent who they really are when they lie about some things as
opposed to when they are completely disclosing about their shortcomings.
People generally view more desirable self-portrayals as compared with
less desirable self-portrayals as more truthful or representative of them-
selves (Schlenker, 1980, 1986). For example, my colleagues and I (Kelly,
Kahn, & Coulter, 1996) observed that in both samples of undergraduate
and client participants, those who had rated themselves as being de-
pressed as compared with those who had rated themselves as non-
depressed viewed their self-descriptions as less representative of them-
selves. Perhaps this notion can explain why Bill Clinton said in his
infamous quote about marijuana use, "I tried it, didn't like it, and didn't
inhale." That statement seems absurd on its face, because people would
agree that one has either tried pot or not; there is no half-trying! But if
Bill Clinton felt that being a pot smoker was foreign to his self-concept,
then he might have felt that it would be a more accurate reflection of
himself to say that he did not really try it. (Of course, it is always pOSSible
that he wanted to make sure that he would win the election and that the
veracity or representativeness of his statement was irrelevant.)
The second issue is that a secret keeper may fear that the secret will be
discovered later and that then the relationship will be destroyed. This fear
should be debilitating, given that, as described earlier, there is a good deal
of empirical support for the notion that a fundamental human motivation
is the need to belong. Fearing that the relational partner may ultimately
abandon him or her because of a revelation is indeed a great threat, and it
may help explain why people so often reveal their damaging secrets.
Getting back to the Graham-Zamora example of secrecy, I suggest
THE NATURE OF SECRECY 21

that Diane Zamora may have revealed her secret to her friends because she
may have feared that any relationships that she could develop with them
would be impaired by her own knowledge of the secret and awareness of
the possibility that they could later discover it. Now, anyone who gets to
know and like Zamora does so with the knowledge of the murder. Even
though it should be harder to get people to like her, at least when they do,
she can trust that such liking will not be destroyed by a future revelation.
Of course, I am merely speculating here; she just may have been bragging
and not thinking about the grave consequences of her confessions.

CONCLUSION

Secrecy is an extremely common phenomenon and the ability to

conceal information from others even may be seen as a sign of maturity or
normal adult functioning. Secrets are those things that people hide from at
least one other person; thus, it only makes sense to think of secrets in the
context of relationships. The most frequently reported types of personal
secrets that people keep are sexual secrets and secrets that make them
look maladjusted, either like a failure or as though they are mentally ill.
These themes refer to information that might make the people look like
they are abnormal or are not living up to others' expectations and that
might cause them to be rejected or ostracized from their groups. Being
rejected by one's social group is no small event. A lack of attachments has
been linked to a variety of ill effects on health, adjustment, and well-being
(see Baumeister & Leary, 1995). This fundamental motivation to belong
may explain both why people sometimes choose to reveal and sometimes
choose to keep their secrets. They may choose to reveal some secrets in an
effort to forge relational bonds with particular individuals and may choose
to conceal those same secrets from other people in an effort to avoid being
rejected by them. Another intriguing possibility for why people sometimes
reveal damaging secrets to others is that people have a compelling need to
disclose their emotional experiences with others (e.g., Rime, 1995a). How-
ever, it is not yet clear that such a need does exist, given that the primary
evidence to support this notion is simply that people very frequently do
share their emotional experiences, especially those more emotionally in-
tense experiences, with others. It may be that because people feel an
expectation from others to disclose their emotional experiences, they do
disclose. It is possible that if these cultural norms changed, people would
not disclose as often. Another intriguing explanation for why people
reveal certain secrets but not others is that they are willing to reveal secrets
that they feel depict who they really are. For example, if they felt that they
22 CHAPTER 1

did something that was unrepresentative of who they normally are, even if
they were responsible for their actions, they would be less likely to reveal
those events than if they felt that those secrets did capture who they really
are. Given that people generally see more favorable events and descrip-
tions of themselves as representative of who they really are, those secrets
that they reveal are likely to be ones that still allow them to see themselves
in a favorable light.
CHAPTER 2

INDIVIDUAL DIFFERENCES
IN SECRET KEEPING

/II don't even know if I have the capacity for normal emotions or not
because I haven't cried for a long time. You just stifle them for so long that
maybe you lose them, partially at least. I don't know./I (Jeffrey Dahmer
on his confessions. http://serial-killers.virtualave.net/dahmer5.htm. Re-
trieved 5/10/00.)
Over the years, I have worked closely with a number of students in
my laboratory. Some of them have told me a great deal about themselves
and their families, others have been so secretive that I did not even know
that they were seriously dating or engaged to another student in the
psychology department. Most people probably have had similar experi-
ences, in which they were shocked to discover hidden information about
people whom they presumably knew well. Let me be clear right from the
start of this chapter: these people who keep secrets and are hard to get
to know are sicker, both psychologically and phYSically, than people who
do not keep secrets.
An extreme example of this secretive kind of person was the late serial
killer, Jeffrey Dahmer. His father, Lionel, told television reporters how
just prior to his son's arrest in 1991, he had engaged in very mundane
conversations with Jeffrey about Jeffrey's weight and job, as though noth-
ing were wrong. He was pleased to see that Jeffrey seemed to be living a
relatively normal existence, only to be stunned to discover later that Jeffrey
was capable of keeping tremendous secrets from him. These secrets sur-
rounded Jeffrey's gruesome killing of more than a dozen people. The
aspect of this case that is most relevant to this chapter is throughout much
of his life, the highly secretive Jeffrey experienced depression, anxiety,
23
24 CHAPTER 2

loneliness, and extreme shyness, just as people who conceal negative

personal information in general are shy and vulnerable to depression,
anxiety, cancers, infectious diseases, and other physical and psychological
problems (e.g., Larson & Chastain, 1990).
In this chapter, I review the evidence that people who keep stigmatiz-
ing, negative, or distressing personal secrets tend to be sicker than those
who do not, and I address the question, "Who are these secret keepers?"
Part of understanding them involves contrasting self-concealment (i.e.,
the predisposition to keep secrets) with another, more frequently studied
form of psychological inhibition: repression. Although as indicated in the
previous chapter, secrecy necessarily involves at least two people, repres-
sion might be thought of loosely as keeping secrets from oneself. I explore
how repression and self-concealment each relate to physical and psycho-
logical problems and then conclude by providing a profile of the secret
keeper. Because of the difficulty in using self-report measures to assess the
correlates of repression (Le., repressors do not know when they are re-
pressing unpleasant material) (Wegner, 1989), in my review of this re-
search, I focus on studies that assessed physiological differences between
repressors and nonrepressors. In contrast, in my review of the research on
the correlates of self-concealment, I have included studies that used partic-
ipants' self-reports, because self-concealment is a conscious process and
self-concealers do not seem to deny their symptoms (see Kahn & Hessling,
2001; King et aI., 1992).

CORRELATES OF UNCONSCIOUS INHIBITION

AND THE REPRESSIVE COPING STYLE

At the age of 64, Jacqueline Bouvier Kennedy Onassis succumbed

to non-Hodgkin's lymphoma. Her death saddened millions of people
around the world, who remembered her as someone who was very mod-
est about her achievements and who remained extremely composed dur-
ing very trying times. She faced the emotional torment of the birth of a
stillborn child in 1956 and the assassination of her husband, President John
F. Kennedy in 1963 with quiet courage. Images still haunt Americans of
her brave and contained behavior at her husband's widely publicized
funeral. Because of her outward composure in such trying times, she
frequently has been held up as an example of a life lived with courage,
respect, and dignity. Yet one wonders whether such composure may have
taken a toll on her health.
The inhibition of emotional expression has been seen as having mixed
consequences. On the one hand, the inhibition of such expression helps
INDIVIDUAL DIFFERENCES 25

avoid the escalation of angry feelings and potentially deadly aggressive

acts (Suls & Fletcher, 1985). Avoidant strategies, as compared with non-
avoidant ones, generally are associated with more positive adaptation in the
short run (Suls & Fletcher, 1985). Moreover, the flip side of repression-
being overly vigilant and disclosing of problems-has been linked to
cardiovascular disease (Taylor, 1990). On the other hand, inhibited expres-
siveness has been identified as a risk factor for psychosomatic problems,
such as back pain and headaches (Traue & Kraus, 1988), and a repressive
coping style has been linked to problems as severe as cancer (e.g., Cox &
McCay, 1982; Dattore, Shontz, & Coyne, 1980; Jensen, 1987; Kissen, 1966;
Temoshok, 1987). For example, Garron and Leavitt (1979) found a small
negative relation between self-reports of hostility and low-back pain levels
among a large sample of medical patients. Garron and Leavitt (1979)
interpreted this finding as supporting theories relating low-back pain to
the inhibition of anger. Cochrane and Neilson (1977) found that psychiatric
patients who were classified as being chronically or endogenously de-
pressed tended to inhibit expression of aggression more than did patients
who were not depressed. Moreover, engaging in inhibition is associated
with the increased muscular activity that accompanies stress (Traue, 1995).
It has been demonstrated in a number of studies that there is an inverse
relation between emotional expression and autonomic reactivity (Buck,
1984; Hokanson & Burgess, 1962; Hokanson & Shetler, 1961). Finally, inhib-
ited expressiveness may be a sign to other people that the person has
deficits in emotions and communication (Traue, 1995). In the following
paragraphs, I describe in more detail the findings on the harmful correlates
of unconscious inhibition and the repressive coping style.

CANCER

Cox and McCay (1982) concluded that among a set of psychosocial

variables, such as perceived social support, the strongest predictor of
developing cancer is an antiemotional attitude, especially an inability to
express negative emotions. Likewise, in a review of the research on the
psychological antecedents of cancer, Greer (1983) found evidence that the
development of cancer may be related to the predisposition to engage in
emotional inhibition.
Further supporting the notion that unconscious inhibition is linked
to cancer are the findings from a study of 52 women with a history of breast
cancer and 34 healthy women (Jensen, 1987). The women were studied
over the course of an average of 624 days to see what personality charac-
teristics of the women predicted the recurrence of their cancer (Jensen,
1987). The spread and recurrence of cancer was greater on average in the
26 CHAPTER 2

women who had a repressive personality style and who tended to avoid
expressing negative emotions. Jensen (1987) was able to show that these
relations existed even when he statistically controlled for the stage of the
disease at the time the women were diagnosed, age, total length of disease
course, hematologic factors, and blood chemistries at the time of onset.
Even more compelling evidence that the repressive coping style may
be linked to cancer comes from a prospective study of 400 women present-
ing with an abnormal lump and/or an abnormal mammogram (Flowers
et al., 1995). The majority of the 36 women who were ultimately diagnosed
with a malignant tumor were repressive copers, whereas repressive copers
were not overrepresented among the woman who did not require a biopsy
or who had a negative one.
Kneier and Temoshok (1984) found that repressive styles of coping with
stressful emotions were associated with high cancer incidence and poor
prognosis. They compared the repressive coping reactions of three matched
groups of people ranging in age from 40 to 65 years: 20 malignant mela-
noma patients, 20 cardiovascular disease patients, and 20 disease-free
controls. Repressive coping reactions were defined as reactions in which
there was a discrepancy between self-reported anxiety and physiological
response to anxiety-provoking statements. (The validity of this procedure
was established through demonstrating correlations with other indices of
repressive tendencies such as the Taylor Manifest Anxiety Scale.) The
researchers found the melanoma group was significantly more repressed,
whereas the cardiovascular disease participants were the least repressed
or most sensitized. These differences in defensive posture were indepen-
dent of disease severity; in other words, the differences were not merely
the result of differences in disease-related anxiety. The researchers sug-
gested that coronary-prone and cancer-prone individuals may be at oppo-
site poles on the distributions of several coping and personality variables,
with cancer-prone individuals being more likely to engage in repressive
coping (Kneier & Temoshok, 1984).
Patients in another study who had malignant melanoma (i.e., a viru-
lent form of skin cancer) and who reported little distress in the early
phases of their cancer, despite their facing the same difficult situation as
the other patients, tended to experience greater recurrences and had
higher death rates than those who reported more distress (Fawzy et al.,
1993). The researchers noted that this relationship was likely due to the
fact that those who minimized their illness tended to cope less effectively
with the cancer, such as not seeking medical treatment or complying with
treatment.
In yet another study, Goldstein and Antoni (1989) investigated the
relationship of three repressive coping styles to breast carcinoma incidence
INDIVIDUAL DIFFERENCES 27

and metastasis. These three discrete repressive coping styles were labeled
as Introversive, Cooperative, and Respectful, as measured by the Millon
Behavioral Health Inventory (MBHI). Upon admission to a cancer treat-
ment unit, 44 female patients recently diagnosed with nonmetastatic or
metastatic breast carcinoma completed the MBHI. All patients received
mastectomy, chemotherapy, radiation, and/or endocrine therapy. These
patients were compared with 34 controls on measures of coping style and
psychological distress. The cancer patients, as compared with the non-
cancer patients, were more likely to employ a repressive coping style.
Moreover, the only group to attain a clinically significant mean score on the
Respectful scale was the group of cancer patients with metastasis. Gold-
stein and Antoni (1989) concluded that the incidence of repressive coping
styles may be disproportionately high among breast cancer patients.
To obtain a better understanding of the inhibited emotional expres-
sion evident in breast cancer patients, Servaes, Vingerhoets, Vreugdenhil,
Keuning, and Broekhuijsen (1999) compared 48 breast cancer patients and
49 healthy women on measures of disturbed emotional processes (i.e.,
alexithymia, or having a limited capacity to put feelings into words), emo-
tional disclosure, emotional expression, assertiveness, repression, and dis-
tress. The patient group showed significantly more ambivalence over
emotional expression, more restraint, and more anxiety than the healthy
controls. No differences were found between the two groups in alexi-
thymia, expressing emotions in general, or willingness to talk with others
about emotions. The image of the breast cancer patient that emerged in the
study was that of a person who has conflicting feelings with regard to
expressing emotions, is reserved and anxious, is self-effacing, and re-
presses aggression and impulSiveness. The researchers interpreted their
findings to mean that cancer patients' inhibited behavior is a reaction to
the disease rather than a reflection of a personality characteristic predis-
posing an individual to (breast) cancer.
In sum, the researchers who have examined the link between having
a repressive coping style and cancer typically have looked at people who
already have cancer and then measured their recovery/death rates. Al-
though Servaes et al. (1999) suggested that repressive coping is actually a
response, not a precursor, to the disease, what does seem clear is that once
a person has cancer, this repressive coping response is linked to poorer
prognosis and reduced recovery from the disease.

PHYSIOLOGICAL AROUSAL

Weinberger et al. (1979) studied the autonomic nervous system corre-

lates of repression. They asked repressors (Le., who were high in social
28 CHAPTER 2

desirability and low in self-reported anxiety) and nonrepressors to com-

plete sentences that were aggressive, sexual, or neutral in nature. The
repressors obtained significantly higher galvanic skin response, electro-
myogram, and heart rate scores, which suggests that they experienced
more arousal and stress in response to the sentence completion tasks than
did the nonrepressors.
Weinberger and co-workers' (1979) findings intensified researchers'
interest in investigating the physiological consequences of repression,
particularly concerning heart rate and arousal. For example, Siegman,
Anderson, and Boyle (1991) subsequently replicated the link between
the repressive coping style and elevated heart rate. However, they did not
find a Significant relationship between the repressive coping style and
heightened blood pressure. The findings from other studies suggest that
repression may be linked to heightened skin conductance levels, but not
with heightened cardiovascular reactivity (Barger, Kircher, & Croyle, 1997;
Pennebaker, Hughes, & O'Heeron, 1987).

HEART DISEASE

Denollet (1993) postulated that distressed individuals (characterized

by elevated levels of type A behavior, anger, hostility, and life stress) and
inhibited individuals (characterized by the nonexpression of anger) may
be particularly coronary prone. Supporting this idea was an examination
of 60 42- to 64-year-old males at high risk for coronary heart disease who
were evaluated in terms of their expressive style, specific nonverbal cues,
personality, and health (Friedman, Hall, & Harris, 1985). Judges watched
videotapes of the participants and rated their appearance, the actual
audio and video nonverbal cues that the participants emitted, and what
they said (words in the transcript). As it turned out, the men who were
rated as more repressed and tense, as opposed to more talkative and
charismatic, tended to be more illness prone, thus supporting the notion
that repression may be linked to heart disease (Friedman et al., 1985).
However, another study with male cardiac patients did not support
this link (Denollet, 1991). Denollet (1991) investigated the potential influ-
ence of negative affectivity (NA) and repressive coping on cardiovascular
fitness among 178 male cardiac patients (aged 31-76 years) who were
undergoing rehabilitation. The patients were categorized as high in NA,
low in NA, or as repressive copers. No association was found between
coping style and cardiovascular fitness as measured by exercise stress
testing.
Because both the inhibition of emotions, such as hostility, and the
expression of such emotions have been linked to heart disease, Siegman
(1993, 1994) focused his research efforts on clarifying the connection be-
INDMDUAL DIFFERENCES 29

tween hostility or anger and various negative physical side effects. He

(1993, 1994a,b) challenged the widely held belief that the expression of
anger is beneficial to mental and physical health. In particular, he reviewed
experimental and correlational studies from several different laboratories,
including his own, that demonstrated that the full expression of anger,
with its vocal manifestations (loud and fast-paced), actually was associ-
ated with significant cardiovascular hyperactivity (e.g., Engebretson, Mat-
thews, & Scheier, 1989; Siegman, Anderson, Herbst, Boyle, & Wilkinson,
1992; Suarez & Williams, 1990). Furthermore, epidemiological studies indi-
cate that such expressions of anger also are related to coronary heart
disease and to some physiological and hormonal changes that have been
implicated in the pathophysiology of coronary heart disease. Siegman
(1994b) concluded that neither the mere experience of anger nor its re-
pression has any of the above negative cardiovascular consequences and
that the expression of anger is most clearly related and consistently to
coronary heart disease and its risk factors. Siegman (1994b) suggested that
when dealing with negative emotions such as anger, people may benefit
from speaking about these emotions in a calm, slow voice, as opposed to
either repressing or venting the anger.

CORTISOL SECRETION

Roger and Najarian (1998) investigated the relationship between emo-

tional inhibition and cortisol secretion (i.e., the release of a hormone in-
volved in maintaining blood pressure) during stress among 51 student
nurses (aged 18-42 years) undergoing a written examination as part of
their training. The participants completed the Emotion Control Question-
naire, a measure of their tendencies to engage in mental rumination (i.e.,
the tendency to mull over mental events) and emotional inhibition, one
month before the written exam. Cortisol secretion was selected as a mea-
sure of experiencing stress because it previously was demonstrated that
parents who were less able to cope with the loss of fatally ill children
tended to secrete more cortisol than parents who coped better (Wolff,
Friedman, Hofer, & Mason, 1964) and that phobic participants tended to
secrete more cortisol in response to provocation as opposed to neutral
slides (Frederikson & Engel, 1985). The written examination was consid-
ered by the nurses to be a stressful event, given that the nurses' scores
would have a direct bearing on their continued training. Urinary-free
cortisol was assayed from samples taken immediately after the examina-
tion and again 3 weeks later (i.e., 2 days after their scores on the written
examination were announced) as the control period. As it turned out,
differences in cortisol secretions between the stressful event and control
period were significantly positively associated with emotional inhibition
30 CHAPTER 2

and the tendency to engage in mental rumination. These findings are yet
more evidence that repression is linked to harmful physiological conse-
quences in the face of stress.

IMMUNOLOGIC FUNCTIONING

The link between repression and immunologic functioning recently

has received much attention. For example, in one study, undergraduates
were asked to complete self-report measures of repression and to write a
letter to a friend about a highly stressful event that they had not widely
discussed with others (Esterling, Antoni, Kumar, & Schneiderman, 1990).
Blood was collected between 45 minutes to 4 hours after the writing and
later tested for antibody titers to the latent viral pathogen that is very
prevalent in the general population, the Epstein-Barr virus (EBV). This
immunologic test was used because, among 20 immunologic variables
studied, antibody titers against EBV have been found to be the most
consistent and significant correlate of psychosocial stressors (Van Rood,
Bogaards, Goulmy, & van Houwelingen, 1993). Increases in EBV are inter-
preted as a sign of experiencing stress. Judges coded the degree to which
participants used emotional words out of total words in their writing. The
participants who tended to use fewer emotional words actually had
greater levels of EBV antibody titers after the writing than did participants
who used more emotion words. Also, the participants who scored higher
on the self-report measure of a repressive interpersonal style had greater
levels of EBV antibody titers than did those who scored lower. The re-
searchers interpreted their findings to mean that both having a personality
that is inclined to avoid expressing negative emotions and engaging in
such inhibition are associated with poorer cellular immune control over
the latent virus. However, it is important to note it was not clear whether
participants who used fewer emotional words were engaging in greater
inhibition or simply were being less expressive. (As mentioned in Chapter
1, self-concealment and self-disclosure are empirically distinct and are
not mere opposites.)
In another study, the link between having a repressive coping style
and immunologic functioning was examined once again (Esterling, An-
toni, Kumar, & Schneiderman, 1993). Specifically, Esterling et a1. (1993)
measured the production of antibody titers in response to an EBV viral
capsid antigen in a normal, healthy college population made up of people
previously exposed to EBV. These undergraduates completed a battery
of physical status questions and items pertaining to potential behavioral
immunomodulatory confounds, along with the Taylor Manifest Anxiety
Scale and the Marlowe-Crowne Social Desirability Scale (i.e., low scores
on the former and high scores on the latter indicate a repressive coping
INDIVIDUAL DIFFERENCES 31

style). Participants reporting high and middle levels of anxiety had higher
antibody titers to EBY, suggesting poorer immune control over the latent
virus, as compared with the low-anxious group. Similarly, participants
with a repressive coping style had higher antibody titers than their low-
defensive counterparts. Thus, being too anxious or too repressed was
associated with poorer immune response.
Researchers have suggested that the associations between repressive/
defensive coping styles, enhanced stress responsivity, and reduced immu-
nocompetence may be mediated by the hyperactivity of endogenous
opioid systems Gammer & Leigh, 1999). In other words, it is through this
neurohormonal mechanism-the release of excess endogenous opioids-
that a person could simultaneously experience reduced sensitivity to pain
and distress and increased sympathetic nervous system and endocrine
activity. Jammer and Leigh (1999) conducted a series of experiments and
indeed did find some support for their hypothesis that endorphinergic
dysregulation is associated with repressive/ defensive coping styles. How-
ever, more research is needed to establish this underlying mechanism
more definitively.

SUMMARY
Overall, research on the repressive coping style has supported the
idea that repression is associated with increased cancer rates. Some studies
also have shown that repression is linked to poorer immunologiC function-
ing, increased skin conductance, and increased cortisol secretion in re-
sponse to stress. However, the findings are less clear regarding the link
between repression and elevated blood pressure (Siegman et al., 1991) or
heart disease (Taylor, 1990). In fact, people who are overly vigilant and
disclosing of problems have a greater likelihood of. developing cardio-
vascular disease (Bonnano & Singer, 1990; Taylor, 1990). For example, with
anger it is actually the expression, not the repression, of this negative
emotion that seems to be linked to heart problems (see Siegman, 1994b).
Apparently, either extreme of expressiveness or inhibition is problematic.
Sigman's (1994b) solution is that when dealing with negative emotions
such as anger, people should speak about these emotions in a calm, slow
voice, as opposed to either repressing or venting the anger.

CORRELATES OF CONSCIOUS INHIBmON

In this section, I first identify the correlates of self-concealment, or

the predisposition to keep secrets, and then identify the correlates of
keeping a particular secret. I keep these reviews separate because the two
32 CHAPTER 2

have been studied separately; however, as the reader will see, both are
related to a variety of problems.

SELF-CONCEALMENT

Since the introduction of the 10-item Self-Concealment Scale (SCS)

(Larson & Chastain, 1990), this personality variable has received increasing
attention among psychology researchers (e.g., Cepeda-Benito & Short,
1998; Kelly, 1998; Kelly & Achter, 1995; Larson, 1993a,b). Cramer and Barry
(1999) evaluated the psychometric properties of the SCS and found that in
a large sample of university students, it had both good internal consistency
and reliability over time.
Researchers have shown that high scorers on the SCS-high self-
concealers-have more physical and psychological problems than do low
self-concealers (Cepeda-Benito & Short, 1998; Larson & Chastain, 1990;
Ichiyama et al., 1993; Kelly, 1998; King et al., 1992). For example, among a
sample of human service workers, self-concealment was found to be pos-
itively related to anxiety, depression, and bodily symptoms, such as back
pain and headaches (Larson & Chastain, 1990). These relations held up
after the researchers statistically controlled for participants' traumatic
experiences, trauma distress, disclosure of the trauma, social support,
social network, and self-disclosure (Larson & Chastain, 1990).
Similar relations between self-concealment and measures of psycho-
logical distress have been found in studies using samples of college stu-
dents (Cepeda-Benito & Short, 1998; Ichiyama et al., 1993; Kahn & Hess-
ling, 2001; King et al., 1992) and outpatient therapy clients (Kelly, 1998). For
example, I found that the correlation between symptom (as measured by
the Brief Symptom Inventory; Derogatis, 1993) and self-concealment
scores was moderately strong (r = .37) among a sample of 42 psycho-
therapy clients (Kelly, 1998). In a separate study, John Achter and I found
that the correlation between depression and self-concealment was quite
high (r = .50) among a sample of 375 undergraduates (Kelly & Achter, 1995,
Study 1). Interestingly enough, even though the women in our study had
higher depression scores than the men, the women and men did not differ
in their mean self-concealment scores.
In that same study, we asked the participants to imagine that they
were experiencing a series of problems that college students commonly
present to counselors and to report how likely they would be to seek
counseling for these problems (Kelly & Achter, 1995, Study 1). We also
asked them to indicate their general attitudes toward seeking counseling,
as well as to rate how strong their social support networks were. As it
turned out, high self-concealers, as compared with low self-concealers,
INDIVIDUAL DIFFERENCES 33

reported a higher likelihood of seeking counseling, despite their more

negative attitudes toward counseling. Moreover, although perceived so-
cial support was strongly negatively correlated with self-concealment
(r = - .45), it was not a significant predictor of likelihood to seek counsel-
ing. In Study 2, we found that more high self-concealers (57%) reported
having seen a counselor than did low self-concealers (37%). Thus, we
concluded that high self-concealers are an enigmatic group in that they
simultaneously have negative attitudes toward counseling and a high
likelihood of seeking counseling.
Cepeda-Benito and Short (1998) attempted to replicate our findings by
asking a large sample of undergraduates to complete questionnaires about
their perceived likelihood of seeking professional psychological help, atti-
tudes toward psychotherapy, fears of psychotherapy, psychological dis-
tress, social support, and self-concealment. In their study, as in ours, self-
concealment was positively associated with self-reported distress (r = .44)
and negatively related to perceived social support (r = - .28). However,
unlike our results, they found that self-concealment was not significantly
associated with reported likelihood of seeking counseling (r = .06). Also,
in their sample, 21% of the high self-concealers versus 19% of the low self-
concealers reported having seen a professional counselor, which was not a
statistically significant difference. Thus, our earlier finding that high self-
concealers, as compared with low self-concealers, are more likely to seek
counseling may be the case only in groups of people who overall have a
high rate of seeking counseling.
Recently, Macdonald and Morley (2001) asked 34 people referred to
psychotherapy to keep a diary for one week of disclosed and nondisclosed
emotions. They found that 68% of their emotions were not disclosed, as
compared with much lower rates of nondisclosure observed in studies
with nonclinical samples (e.g., Rime et al., 1991a). They suggested that the
people in this clinical sample were habitual nondisclosers of emotional
and personal experiences, which is consistent with our earlier observation
that high self-concealers may be more likely to end up in therapy (Kelly &
Achter, 1995).
In addition to being related to depression and anxiety, self-concealment
has been linked strongly to social anxiety (Gesell, 1999), which refers to
feeling tense when interacting with people at parties or other social set-
tings, and to shyness and low self-esteem among college students (Ichi-
yama etal., 1993). Given their tendency to feel anxious in social settings, it
is not surprising that high self-concealers report having a relatively strong
need to be alone (Cramer & Lake, 1998). In two investigations using large
samples of college students, the link between scores on the Preference for
Solitude Scale (PSS) and Self-Concealment Scale was explored (Cramer &
34 CHAPTER 2

Lake, 1998). The three subscales of the Preference for Solitude Scale identi-
fied by Cramer and Lake are: Need for Solitude (i.e., the desire to be alone),
Enjoyment of Solitude, and Productivity during Solitude. Interestingly
enough, of those three factors, only the Need for Solitude factor (r = .38)
was correlated significantly with self-concealment, and that factor may be
seen as the most negative of the three factors because it involves trying to
get away from people as opposed to taking pleasure in being alone. Also,
as in previous research, self-concealment was found to be significantly
correlated with loneliness (r = .54), low self-esteem (r = .54), and social
anxiety (r = ,29) (Cramer & Lake, 1998).
King et al. (1992) explored the relations between self-concealment (as
measured by the Self-Concealment Scale) and a variety of measures of
constructs pertaining to the broader concept of inhibition. One such
construct-conflict over emotional expression--:-involves wanting to ex-
press emotions and not 'being able to, as well as expressing emotions and
wishing one had not (King & Emmons, 1990). In two previous studies,
King and Emmons (1990, 1991) demonstrated that such conflict was asso-
ciated with depression. In therr study of 155 adults from the community
and undergraduates, King et al. (1992) found that self-concealment scores
were positively correlated with scores on measures of ambivalence over
emotional expression (r = .67), emotional inhibition (r = .39), alexithymia
(i.e., lack of access to one's emotions) (r = .19), emotional control rehearsal
(i.e., a measure of rumination) (r = .30), and obsessional thinking (rs = .39
and .50, with scores on two separate measures of obsessional thinking).
Obsessional thinking refers to the tendency to ruminate over details of
events. Note that the correlations were particularly high for ambivalence
over emotional expression and obsessional thinking. In contrast, self-
concealment scores were negatively related to scores on measures of emo-
tional expressiveness (r= -.44), self-control (rs = -.27 and -.34, with
scores on two separate measures of the ability to delay gratification and
control one's impulses), restraint (i.e., impulse control, consideration of
others, suppression of aggression, and responsibility) (r =- .29), and self-
deception (r = - .33). Self-concealment scores were not statistically signifi-
cantly correlated with social desirability scores (r = - .14).
King and co-workers' (1992) next step in this study was to tty to assess
underlying factors to inhibition. They conceptualized the broader con-
struct of inhibition as composed of the following three factors: rumination,
behavioral control, and emotional constriction. Factor analyses of these
three elements showed that rumination and emotional constriction were
positively correlated with each other, whereas rumination correlated nega-
tively with behavioral control. In factor analyses of scores on all their
measures of inhibition, self-concealment loaded most highly on rumina-
INDMDUAL DIFFERENCES 35

tion. It also loaded on emotional constriction and loaded negatively on

behavioral control. Thus, the picture of the self-concealer that emerges
from the King et al. (1992) study is of a person who ruminates over prob-
lems, is ambivalent about emotional expression, does not engage in self-
deception, inhibits expression of emotions, but does not necessarily inhibit
behaviors.
Supporting the idea that self-concealers do not engage in self-deception
were the findings from Kahn and Hessling's (2001) study of 279 under-
graduates. Specifically, the undergraduates' self-concealment scores were
significantly negatively correlated with their social desirability scores (r =
- .39). Kahn and Hessling (2001) also found that self-concealment was
negatively related to both extraversion (r = - .24) and social support (r =
- .33), which is consistent with the earlier findings that self-concealers
reported being shy (Ichiyama et al., 1993) and having relatively weak social
support networks (e.g., Kelly & Achter, 1995).

KEEPING A PARTICULAR SECRET

Instead of using the Self-Concealment Scale to sort people into high

and low secret keepers, some researchers have opted to compare people
who have a particular secret with those who do not. For example, Finke-
nauer and Rime (1998b) asked 377 undergraduates and their relatives and
acquaintances to complete questionnaires in which they were asked if they
could recall an important emotional life event that they had kept secret.
Approximately 42% reported having such a secret memory, whereas ap-
proximately 56% indicated that they had no such memory. Finkenauer and
Rime compared these two groups on illness and life satisfaction self-
reports. As it turned out, the respondents with a secret memory had
greater total illness scores and were less satisfied with their lives than
respondents without such a memory. The researchers showed that having
a secret memory predicted illness scores, even when they statistically
controlled for participants' levels of negative affect (i.e., the tendency to
experience negative emotions and dissatisfaction). Moreover, based on
their findings from path analyses on the links among having a secret
memory, illness scores, and life satisfaction scores, Finkenauer and Rime
(1998b) concluded that it is through making people ill that having a secret
memory indirectly affects life satisfaction.
In a similar vein, researchers also have shown that gay men who tend
to conceal their sexual orientation from others, as compared with those
who do not, are at a greater risk for developing ailments such as cancer
(Cole, Kemeny, Taylor, & Visscher,1996a; Cole, Kemeny, Taylor, Visscher, &
Fahey, 1996b). Specifically, one study involved 80 gay men who were HlV-
36 CHAPTER 2

seropositive but otherwise healthy at the beginning of the study (Cole et

al., 1996b). These men were assessed a 6-month intervals for 9 years for
signs of lllV progression. The measures included how long it took for the
men to develop a critically low T-Iymphocyte level, to receive an AIDS
diagnosis, and to die from AIDS. As it turned out, the men who tended to
conceal their homosexual identity experienced a more rapid progression
to all these negative outcomes than did the men who were more" out of the
closet." The researchers attributed these differences to the idea that it is
stressful to inhibit the expression of one's sexual identity and that such
stress can lead to negative health effects. They were able to rule out
competing explanations for these remarkable findings by statistically con-
trolling for various demographic characteristics, health practices, sexual
behaviors, antiretroviral therapy, depression levels, anxiety levels, and
degree of social support among the men in the sample.
In another study, these same researchers examined the incidence of
cancer and infectious diseases among 222 HIV-seronegative gay men
(Cole et al., 1996a). Those men who concealed the expression of their
homosexual identity, as compared with those who did not, experienced a
significantly higher incidence of cancer and several infectious diseases
(i.e., pneumonia, bronchitis, sinusitis, and tuberculosis) over as-year
follow-up period. These effects could not be attributed to differences in
age, ethnicity, socioeconomic status, health-relevant behavioral patterns
(e.g., drug use, exercise), anxiety, depression, or reporting biases (e.g.,
negative affectivity, social desirability) between the men who were "in the
closet" and those who were "out of the closet." The researchers speculated
once again that the findings support the notion that inhibiting expression
of something as important as one's sexual orientation is stressful and can
take a toll on one's health.
Likewise, Pennebaker and Susman (1988) discovered that survivors of
childhood traumas who did not discuss the traumas with others, as com-
pared with those who did discuss them, tended to develop more problems
such as hypertension, cancer, and influenza. In addition, Pennebaker and
Harber (1993) observed that among victims of the 1989 Lorna Prieta Earth-
quake, the greatest amount of physical and psychological distress oc-
curred during the time period 2 to 6 weeks after the earthquake that
coincided with when the victims continued to think about the earthquake
but squelched their desire to talk about it with others.
In yet another very important study, patients diagnosed with breast
cancer who less openly shared with others their angry and depressed
feelings about their cancer, as compared with those who were more open
in their expression of these feelings, actually died sooner (Derogatis, Abe-
loff, & Melisaratos, 1979). Also, recent widows and widowers who talked
INDIVIDUAL DIFFERENCES 37

less (e.g., with friends and family) about their spouse's death tended to
have more health problems and tended to ruminate more about the death
in the year following the tragedy (Pennebaker & O'Heeron, 1984). These
correlations remained even when the researchers statistically adjusted for
number of friends these individuals had before and after the loss of the
spouse.

SUMMARY

There is mounting evidence that self-concealment is moderately

strongly to very strongly linked to a variety of problems, ranging from
depression and anxiety, to rumination, back pain, and headaches. More-
over, keeping a particular secret, such as being gay, has been linked to
increased vulnerability to cancer, infectious diseases, and acceleration to
AIDS symptoms.
At this point, one might ask the question, "Could this link between
self-concealment and problems be explained by the possibility that secret
keepers simply have had more negative life events than nonsecret keepers,
and therefore they have both more things to hide and more problems?"
Larson and Chastain (1990) addressed this possibility by statistically con-
trolling for participants' reports of traumatic life experiences (e.g., having
been raped or molested) before the age of 17, and they still found a
moderately strong link between self-concealment and problems. How-
ever, one could argue that such a control was inadequate because although
self-concealers may have been willing to admit to having the tendency to
keep secrets, they did not want to admit to the specific traumatic experi-
ences they had in their youth. Another problem is that Larson and Chas-
tain's (1990) sample was composed of people substantially older than 17.
The high self-concealers within the sample may have experienced more
traumas since the age of 17 than the low self-concealers.
Another pOSSible explanation for the link between secret keeping and
problems is that it is really neuroticism or negative affectivity (i.e., the
tendency to experience negative emotions and dissatisfaction) that under-
lies the link. Put another way, it could be that people who tend to complain
and to admit to negative things are Simply endorsing both high symp-
tomatology and the negative quality of keeping secrets. However, as men-
tioned earlier, Finkenauer and Rime (1998b) showed that having a secret
memory predicted illness scores, even when they statistically controlled
for participants' levels of negative affect.
Although these two alternative explanations already have been ad-
dressed to some extent by researchers, they draw attention to the fact that
the findings on the link between self-concealment and problems are cor-
38 CHAPTER 2

relational. As such, it is not possible to say that secrecy causes these

problems. It is possible, however, at this point to describe what the secret
keeper is like.

PROFILE OF A SECRET KEEPER

Even though the findings from the literature on conscious forms of

inhibition have more direct bearing on profiling the secret keeper, who is
defined as someone who keep secrets from others, the findings on both
repression and self-concealment shed light on who these secret keepers
are. People who keep secrets from themselves-repressors..,-do not report
experiencing distress and problems, but they do show signs of physical
weakness. Repressors have weaker immune responses to stress and tend
to get more physiologically aroused by stress than do nonrepressors,
which may II:lake them vulnerable to cancers in the long run. There is some
evidence that repression may have short-term benefits, such as aVOiding
the escalation of aggressive outbursts. Also, repressors are less likely to
complain about problems, SO they may be easier to be around in the face
of an immediate problem or crisis.
People who tend to keep personal secrets from others-
self-concealers-tend to be lonely, shy, and somewhat introverted. They
are conflicted over the expression of emotion, and they tend not to be
emotionally expressive. They also are socially anxious and have low self-
esteem. To a lesser extent, self-concealers have an impoverished emotional
life in the sense that they have a relatively low access to their emotions.
They also report having a relatively strong need to be alone, perhaps
because of their shyness and anxiety around others. They mull over their
troubling thoughts and problems and are depressed and anxious gener-
ally. A troubling finding that has emerged from King and co-workers'
(1992) study is that self-concealers report being somewhat impulsive and
unable to inhibit or control their behaviors. Thus, they are in the unpleas-
ant position of not being emotionally expressive, while behaving impul-
sively, which may explain their high levels of rumination and obsessional
thinking. Specifically, these individuals are vulnerable to engaging in acts
that they later will ruminate over and regret. Self-concealers report having
more physical problems, too, like stomach pains and nausea. These people
tend to have low levels of social support. Compared with low self-
concealers, they are probably less well-liked and less able to form strong
interpersonal bonds with others, given that the exchange of secrets is
typically a part of forging relational bonds (see Derlega, Metts, Peronio, &
Margulis, 1993), although this idea remains to be tested. Overall, high self-
INDIVIDUAL DIFFERENCES 39

concealers are more likely to be unhappy, troubled individuals than are

low self-concealers. Without much doubt, being a keeper of personal
secrets seems to be negative. To date, there is no evidence that self-
concealment is linked to any positive qualities or signs of healthy psycho-
logical functioning, with the exception that these individuals do not tend
to engage in self-deception. But even with that quality, there is evidence
that some self-deception is a positive or healthy quality that is associated
with both well-being and the ability to care about others (Taylor & Brown,
1988, 1994).
Thus, self-concealers are enigmatic on three fronts: First, they are
lonely, yet they report having a high need to be alone. Second, they report
that they lack behavioral control (i.e., they are relatively impulSive in terms
of their actions), but they also report that they inhibit their expression of
emotions. Along these same lines, they tend to ruminate over negative
events or question their behaviors after they have performed them. Third,
they have negative attitudes toward counseling, yet they have more prob-
lems, and thus are more likely to seek counseling than are low self-
concealers (Kelly & Achter, 1995).
I conclude by returning to the example of Jeffrey Dahmer, because he
so epitomizes the extreme levels of self-concealment and the negative
characteristics associated with being a high self-concealer. Until his arrest
in 1991, he had concealed his problems from virtually everyone, including
committing murders in his grandmother's basement and then in his own
apartment without detection. Even in the face of a crisis when one of his
victims almost escaped and the police entered his apartment, Dahmer
calmly was able to conceal his murders from the police. Like other self-
concealers, Dahmer was lonely and shy. Even as a young boy, his father
recalls Jeffrey's having spent extensive periods of time alone in the woods.
At the same time, the reason Dahmer gave police for why he killed his
victims was that he was lonely and did not want them to leave him. Like
other self-concealers, he had at one point sought professional psychologi-
cal services. He had checked himself into an alcoholism rehabilitation
center, but his treatment did not work. Finally, Dahmer was both impul-
sive in his actions-at one point he decided that he would no longer try
to control his impulses to kill-and emotionally inexpressive. His behav-
ior at his trial was characterized as emotionless. However, like other self-
concealers, he did not deny his troubles or try to present himself in a
favorable light. When his crimes were finally discovered by the police, he
admitted to feeling "thoroughly evil."
CHAPTER 3

WHY SECRECY IS LINKED

TO PROBLEMS

Just as Jeffrey Dahmer represents an extreme version of the high self-

concealer (not to mention a sociopath), Katherine Power seems to epito-
mize the problems associated with keeping an important, troubling secret
(see Burger, 1997; Polivy, 1998). She was a Brandeis University student who
drove the getaway car in 1970 for a revolutionary group who performed a
bank robbery in Boston that went terribly wrong. A police officer was
killed. After the robbery, Power went underground, eventually settling
near Corvallis, Oregon. She married and bore a son and was teaching and
working as a chef under the alias Alice Metzinger. However, in September
1993, Power finally turned herself in to Massachusetts police, pleaded
guilty to manslaughter, and was sentenced to 8 to 12 years in prison. What
is intriguing about this case is that her name was removed from the FBI's
most wanted list after 14 years, and it seemed as though she had gotten
away with the crime. Yet in 1992, she had become anxious and depressed
and sought therapy, telling her psychotherapist about the crime in the
second session (Burger, 1997). "Power was so desperate to talk about her
crime that she had become seriously depressed" (Polivy, 1998, p. 188).
As described in the previous chapter, there is evidence that people
who keep secrets have poorer mental and physical health than those who
do not. But does secret keeping actually cause problems as it seems in
Katherine Power's case? More broadly, why do secret keepers have such
problems? Answering these questions is the focus of this chapter.
As I alluded to in the previous chapter, researchers have attempted to
explain the link between secrecy and symptoms by arguing that keeping
secrets or actively inhibiting self-disclosure is stressful and that such stress
41
42 CHAPTER 3

makes one sick (e.g., Pennebaker, 1985, 1989, 1990). Although this inhibi-
tion idea has received the most attention from psychologists, other re-
searchers from both clinical and social psychology (e.g., Derlega, 1993;
Temoshok, 1983) have proposed their own explanations for the link be-
tween secret keeping and health problems. In this chapter, I describe these
explanations and the evidence that has been offered to support them. I
conclude this chapter by proposing a new explanation for the link between
secrecy and problems that is based on the possibility that people who are
born with the predisposition to inhibit social expression also may be born
with the predisposition for various kinds of illnesses.

INHIBmON MODEL

Because the notion that the inhibition of feelings can lead to illness
(Pennebaker, 1985) has received the most attention, I begin by evaluating
the inhibition model as a means of explaining the link between secrecy and
problems. Roughly 15 years ago, Pennebaker and Chew (1985) noted that
the act of inhibiting ongoing behavior requires physiological work, and
they demonstrated that undergraduates who were induced to inhibit their
expression of the truth (i.e., to lie) to experimenters experienced elevated
skin conductance levels relative to their baseline skin conductance levels.
Shortly thereafter, Pennebaker (1985,1989,1992) proposed a model of the
relationship between traumatic experience and psychosomatic disease
that included the following three propositions: (1) To inhibit actively one's
behavior is stressful and disease-related; (2) when individuals do not or
cannot express thoughts and feelings concerning a traumatic event (i.e.,
when they engage in behavioral inhibition), there is an increased proba-
bility of having obsessive thoughts about the event and of having illnesses
in the long run; and (3) conversely, the act of confiding or otherwise
translating the event into language reduces autonomic activity (in the
short run) and leads to long-term reductions in disease rates. According to
Pennebaker (1997a,b), how the event is discussed, the possibility of ever
coming to terms with the event, and the ultimate consequences of discuss-
ing the experience are all variables that may influence the outcomes of
confiding, inhibition, and, down the road, health. Temoshok (1983) earlier
had expressed a similar idea in her multidimensional model of illness, in
which she depicted a coping style that involves an insufficient expression
of thoughts or emotion as a part of the development of psychosomatic
disturbances.
Pennebaker (e.g., see 1997a,b) cited a good deal of experimental re-
search from his own laboratory to support his inhibition model. However,
WHY SECRECY IS LINKED TO PROBLEMS 43

for obvious ethical reasons, these experiments were not designed to test
the idea that secret keeping causes negative health effects. What Penne-
baker's experiments were designed to show-and did show-is that
revealing previously undisclosed traumatic experiences in a confidential,
anonymous setting leads to health benefits. As such, these experiments
will be described in the next chapter. Because of the ethical constraints
involved in inducing people to keep secrets to see whether they become
sick, as yet no published experiments have tested directly the long-term
health effects of secret keeping. In several experiments, however, investi-
gators have assessed the effects of inhibiting the behavioral expression of
emotions, and the results of these are described next.
Cioffi and Holloway (1993) asked undergraduates to endure a cold
pressor pain induction to their hands. The participants were randomly
assigned to conditions in which they were instructed either to (1) concen-
trate on their room at horne (distraction), (2) pay close attention to their
hand sensations (monitoring), or (3) remove awareness of those sensations
from mind (suppression). During the 2 minutes immediately following the
withdrawal of the painful cold induction, the participants in the monitor-
ing condition reported the most rapid recovery from the pain. Participants
in the suppression condition reported the slowest recovery. What is most
interesting about Cioffi and Holloway'S findings is that later during the
experimental hour, the participants in the suppression condition as com-
pared with the other participants interpreted an innocuous vibration against
their necks as more unpleasant. The researchers suggested that the sup-
pression task had drained participants' capacities to cope with pain, and
that, therefore, the suppression of pain is not an effective coping strategy.
However, Colby, Lanzetta, Kleck (1977) found that inhibiting expres-
sions of pain reduced the distress associated with the pain. Ten male
undergraduates were asked to pose three levels of pain expression while
they were receiving electrical shocks, which they terminated at their toler-
ance level. It was found that even though pain tolerance levels were not
related to their level of expression, participants' skin conductance levels
were. Specifically, when participants inhibited their expressive behavior,
they experienced decreased skin conductance responses to the shock,
suggesting that they were less distressed by the shock.
Lanzetta, Cartwright-Smith, and Kleck (1976) uncovered a similar
pattern in three experiments in which they asked participants either to
conceal, freely express, or exaggerate their facial displays associated with
the anticipation and reception of painful shocks. Participants in the conceal
conditions, as compared to those in the free-expression or exaggeration
conditions, experienced lower skin conductance levels and reported feel-
ing less pain.
44 CHAPTER 3

In another experiment, women were instructed to imagine three posi-

tive and three negative emotional scenes (McCanne & Anderson, 1987).
During the initial imagination of each scene, the participants were told
simply to imagine the situation. They next were instructed to imagine the
situation again and enhance the muscle tension in the respective muscle
groups in their faces that corresponded with the emotions in the positive
and negative scenes. The participants then were instructed to imagine the
scene a third time and suppress the muscle tension in the same muscle
groups. They were successful in altering their muscle tension in accord
with the experimental instructions, with no overt signs of changes in their
faces during imagination of the scenes. The participants reported that they
experienced less enjoyment and more distress during positive affective
trials in which they suppressed the muscle activity in their faces. Thus,
inhibiting the expression of positive emotion reduced the subjective expe-
rience of it.
To investigate further the effects of inhibiting emotions, researchers
have conducted a series of experiments using the following paradigm:
Participants are presented emotional films and asked to inhibit or express
their emotions while watching the films, and then their physiological and
self-reported changes in emotion are assessed. In one such experiment,
undergraduates were exposed to six negative videotaped scenes and were
randomly assigned to conditions in which they produced facial expres-
sions of suppression, spontaneous behavior, or exaggeration while watch-
ing the scenes (Zuckerman, Klorman, Larrance, & Spiegel, 1981). As it
turned out, higher levels of facial expressiveness were accompanied by
higher levels of autonomic activity and higher levels of self-reported
emotion. These relationships were obtained in comparisons across experi-
mental conditions as well as in correlational analyses within conditions,
once again supporting the idea that inhibition of the behavioral expression
of negative emotions decreases the experience of those emotions.
In a similar experiment, Bush, Barr, McHugo, and Lanzetta (1989)
examined the impact of facial control on undergraduates' self-reported
responses to dubbed and undubbed comedy video clips. The dubbing
involved inserts of smiling people. The participants were randomly as-
signed to conditions in which they either freely expressed their emotional
responses to the video clips (spontaneous condition) or controlled their
facial reactions (inhibition condition). The presence of dubs increased
electromyogram (EMG) activity over cheek and eye muscle regions for the
participants in the spontaneous condition but not for participants in the
inhibition condition. Participants in the spontaneous condition increased
smiling at dub points and reported significantly greater amusement to
dubbed compared with undubbed routines. Most important for the pre-
WHY SECRECY IS LINKED TO PROBLEMS 45

sent discussion, those participants in the spontaneous condition, as com-

pared with participants in the inhibition condition, reported significantly
more amusement to dubbed routines. Thus, inhibition led to less enjoy-
ment of the amusing films.
Likewise, Labott, Ahleman, Wolever, and Martin (1990) found that
inducing the inhibition of expression of negative emotions can offset
negative changes in both immunologic functioning and mood. Labott and
coworkers recruited a sample of undergraduate women who considered
themselves to be emotionally expressive in response to movies, and they
showed these women both sad and humorous movies. Before watching
the movies, the women were randomly assigned to conditions in which
they were told either to inhibit totally their expression of emotions or to
express overtly their emotions as much as possible. Unfortunately, the
instructions to inhibit or express emotions did not necessarily correspond
with actual behavior of the women. Seven women were dropped from
each experimental group because they either did not inhibit their laughter
or did not express sadness through crying, leaving eight women in each
group. As it turned out, regardless of the overt laughter that the women
expressed, the humorous stimulus resulted in improved immunity. More-
over, the expression of sadness through overt crying was associated with a
reduction in immunologic functioning, whereas the inhibition of crying in
the context of the same sad stimulus was not. Also, following the sad
stimulus, participants who were in the expression condition as compared
with those in the inhibition condition had more negative moods. Although
these results seem to support the idea that inhibition of negative moods
has positive effects, one must keep in mind that roughly half the women
were dropped from the study. The women who were retained because
they were able to cry in the expression condition may have been more
vulnerable to experiencing negative shifts in their immunologic function-
ing that the women who were dropped from the analyses.
In another experiment that involved the suppression of negative emo-
tions, 43 men and 42 women were asked to watch a short disgust-eliciting
film while their behavioral, physiological, and subjective responses were
being recorded (Gross & Levenson, 1993). The participants were randomly
assigned to conditions in which they were either told simply to watch the
film (no-suppression conditions) or to watch the film while behaving "in
such a way that a person watching you would not know you were feeling
anything" (suppression condition). Participants in the suppression condi-
tion did in fact reduce their expressive behaviors. At the same time, they
experienced a mixed physiological state: On the one hand, they showed
decreased somatic activity and decreased heart rate. On the other hand,
they exhibited increased blinking and indications of increased electroder-
46 CHAPTER 3

mal responding (i.e., elevated skin conductance levels)-both signs of

stress. Despite these physiological differences between the participants in
the suppression and no-suppression conditions, there were no differences
between the two groups in their subjective experiences of emotion. Thus,
it seems that the participants in the suppression condition were showing
some physiological signs of experiencing stress from their suppression
efforts but not other physiological signs of stress; and they were not feeling
any more or less disgust than the other participants.
More recently, Gross and Levenson (1997) assessed the physiological
effects of inhibiting negative and positive emotions in 180 female under-
graduates who watched amusing, sad, and neutral films. Half these partic-
ipants were randomly assigned to a condition in which they were in-
structed to inhibit the expression of their emotions while they were
watching the films, and the other half were instructed simply to watch the
films (i.e., control group). Participants in the inhibition condition, as com-
pared with those in the control group, reported that they felt less amuse-
ment during the amusing film; but there were no differences between the
groups in reported level of sadness during the sad film. The participants
in the inhibition condition experienced a greater increase in sympathetic
activation of the cardiovascular system than did the participants in the
control group when they watched either the amusing or sad film. In con-
trast, when they watched neutral films, there was no difference in sympa-
thetic activation between the two groups. Given that an increase in sympa-
thetic activation is a sign of experiencing stress, Gross and Levenson (1997)
suggested that their physiological findings support the notion that hiding
negative emotions is unlikely to make one feel better. The researchers
concluded that whereas the suppression of a neutral stimulus (i.e., one that
does not produce the impulse to express emotions) seems to have little
phYSiological impact, the "suppression of both positive and negative emo-
tions exacts a palpable phYSiolOgical cost" (Gross & Levenson, 1997, p.101).

SUMMARY

Overall, the findings on the physiological effects of inhibiting the

behavioral expression of emotions are quite complicated and they do not
seem to show consistent patterns. In one study, suppressing pain led to
reduced recovery from the pain (Cioffi & Holloway, 1993). However, in
other studies, inhibiting the expression of pain led to reduced skin conduc-
tance levels (Colby et aI., 1977; Lanzetta et aI., 1976) and reports of feeling
less pain (Lanzetta et aI., 1976). Regarding the inhibition of emotions,
suppressing amusement and sadness led to increased sympathetic activa-
tion of the cardiovascular system (Gross & Levenson, 1997). Yet in other
WHY SECRECY IS LINKED TO PROBLEMS 47

research, inhibiting the expression of the emotions of pain, pride, and

amusement was found to decrease one's self-reported experience of these
emotions (Bush et al., 1989; McCanne & Anderson, 1987; Zuckerman et al.,
1981). Also, inhibiting the expression of disgust did not lead to increases
in self-reported experiences of that emotion (Gross & Levenson, 1993).
Likewise, as mentioned in the previous chapter, discussing anger in a calm
voice seems to be better at decreasing the experience of anger than ex-
pressing the anger with agitated behavior (see Siegman, 1994b). Despite
other researchers' conclusions to the contrary (see Gross & Levenson, 1997;
Polivy, 1998), thus far there seems to be more evidence that supports as
opposed to contradicts the notion that inhibition of both positive and
negative emotions reduces the experience of those emotions at both physi-
ological and subjective levels. Moreover, none of these studies involved an
investigation of the long-term consequences of inhibiting the behavioral
expression of emotions. Therefore, it remains to be seen whether the
foundation of the inhibition model-that is, the notion that actively inhib-
iting one's behaviors is stressful and disease-related-will receive clear or
consistent empirical support.

PREOCCUPATION MODEL

Another popular explanation for the link between secrecy and prob-
lems is what Lane and Wegner (1995) called the preoccupation model of
secrecy. They used this model to explain how attempts to keep information
secret can cause intrapsychic problems, especially obsessive preoccupa-
tion with the secret. According to Lane and Wegner, obsessive preoccu-
pation develops in the follOWing way: First, to keep a thought secret,
people engage in thought suppression, which is the conscious avoidance
or active inhibition of a thought. Second, thought suppression causes
intrusive thoughts. Third, intrusive thoughts cause individuals to renew
their efforts at suppression in attempts to keep the thought secret. The
second and third steps of thought suppression and intrusion cycle back
and forth, causing the secret keeper to experience "continuing mental
unrest" (Lane & Wegner,1995, p. 239). According to Lane and Wegner, the
secret keeper is unable to break the cycle and reach resolution of the secret
without revealing it. They suggested that obsessive preoccupation de-
velops, continues (and possibly even escalates) over time, and can develop
into psychopathology such as having a full-blown obsession (see also
Wegner, 1989, 1992, 1994; Wegner et al., 1994). Most relevant for this vol-
ume, they proposed that obsessive preoccupation does not cease unless
the secret is revealed, and even after the secrecy has been lifted, preoccu-
48 CHAPTER 3

pations may persist. The bottom line to the preoccupation model is that
there are cognitive and emotional consequences of secret keeping, making
secrecy a threat to one's mental health.
The process through which thought suppression ironically may make
the suppressed information more accessible to awareness, the second step
of the model, has been termed the hyperaccessibility of suppressed informa-
tion (Wegner & Erber, 1992). In keeping a secret, one must simultaneously
monitor information consistent with the state of mind one wishes to
maintain and keep track of the information one wishes to hide from others
(Wegner, 1994; Wegner & Wenzlaff, 1996). For example, if an untenured
professor has had a sexual affair with a student, she must continually
watch against revealing that information when she is in the presence of
other faculty members. These others may actually come to serve as nega-
tive cues or reminders of the suppressed information, and their mere
presence may make it very difficult for her to keep the secret out of mind
(see Wegner, 1989, 1992, 1994).
Prior to the publication of the preoccupation model, Wegner and
colleagues gathered evidence to support the idea that suppressing infor-
mation makes that information hyperaccessible (Wegner & Erber, 1992;
Wegner, Schneider, Carter, & White, 1987; Wegner, Shortt, Blake, & Page,
1990; Wenzlaff, Wegner, & Klein, 1991). For example, Wegner and Erber
(1992) asked undergraduates to complete Stroop (1935), or color-naming,
tasks under time pressure. Those participants who were instructed to
suppress particular words as compared with participants who were not
told to suppress the words actually had those words come to mind more
readily. Likewise, in another two experiments, undergraduates who were
asked to suppress thoughts of a no-Ionger-desired past relationship expe-
rienced increased expressions of the thoughts after they suppressed them
(Wegner & Gold, 1995).
Lane and Wegner (1995; Studies 1 and 2) also demonstrated that
keeping a secret enhanced the cognitive accessibility of the secret and that
secrecy and thought suppression were both associated with increased
intrusiveness of the suppressed thoughts (Studies 3 and 4). Specifically,
they tested the idea that induced secret keeping (of a word) would be
associated with increased intrusiveness of that secret word (Study 3). They
presented undergraduates with four target words (e.g., car, child, house,
or mountain) one at a time and asked them to write about their stream of
consciousness following different instructions pertaining to each word for
a series of four 5-minute periods. The instructions were to (1) try to think
about the target word, (2) try not to think of the target word, (3) try to keep
the word a secret (from a group of experts who ostensibly would review
the writing and try to guess what the word was), and (4) simply to write
WHY SECRECY IS LINKED TO PROBLEMS 49

about their stream of consciousness with no special instructions regarding

the word. As it turned out, during the secrecy condition, participants'
ratings of their efforts to keep the word a secret were positively associated
with both the extent to which they intentionally and unintentionally
thought of the target word. Also, across all four conditions, attempts to
keep the word a secret were associated with efforts to suppress the word.
However, comparisons of the means across the conditions showed that the
mean for participants' unintentional thinking of the target was signifi-
cantly lower in the suppression and secrecy conditions than in the think
condition. Moreover, the means for unintentional thinking in the suppres-
sion and secrecy conditions did not differ from the mean for the control
condition. Despite these observations, Lane and Wegner (1995) concluded
that overall their findings support the idea that when people are keeping
a secret, their efforts to engage in thought suppression and their intrusive
thinking become linked.
In the fourth study of that series, which was described in Chapter 1,
Lane and Wegner gave undergraduates a list of 50 preselected topics that
included such items as masturbation, lying, dying, mother, father, and
being in love. They asked the participants to rate how intrusive their
thoughts were about each topic, how often they tried to suppress thoughts
of these topics, and how much they kept each of these topics secret from
others. There was a significant positive correlation between thought sup-
pression and secrecy (r = .31), between the intrusiveness of the thoughts
and secrecy (r = .32), and between the intrusiveness of the thoughts and
thought suppression (r = .23). Lane and Wegner (1995; Study 4) concluded
that they demonstrated that secrecy and thought suppression were both
associated with increased intrusiveness of the suppressed thoughts. One
must keep in mind, however, that this study was limited by the possibility
that participants may have rated as more secret and more intrusive the
topics that were relevant to them, thus making it seem as though secrecy
and intrusiveness are related. Also, as the researchers themselves noted,
the causal direction of these factors could not be established in this fourth
study, given its correlational design.
In addition to the limitations of those studies, there are limitations to
other studies in which the researchers concluded that people become more
preoccupied with their self-generated disturbing thoughts after suppress-
ing them. In the Wegner and Gold (1995) studies, for example, participants
who suppressed thoughts of a still-desired past relationship did not show
subsequent increases in expressions of the thoughts (although their skin
conductance levels, or indications of their emotional arousal, were ele-
vated following suppression). In another study (Trinder & Salkovskis,
1994), undergraduates who were asked to suppress their negative, person-
50 CHAPTER 3

ally intrusive thoughts for 4 days, as compared with participants who

were asked either to try to focus on the thoughts (i.e., the focus group) or
simply to keep track of the thoughts (i.e., the control group), reported more
intrusions of the thoughts. However, the control group and the focus
group received added reassuring instructions (i.e., "It doesn't matter if
the thought comes often or not; I'd like you to think about it for as long as
possible without changing it, just stay with it") that were not provided to
the suppression group. These added instructions may have led to the
group differences by suggesting only to the participants in the control and
focus groups that it was acceptable to have the intrusive thoughts. A
similar laboratory experiment showed that suppressing a negative, per-
sonally intrusive thought for 5 minutes led to greater intrusions of the
thought than did merely keeping track of the thought (Salkovskis &
Campbell, 1994). But this study too included the reassuring instructions
that were read only to the control group.
Despite these limitations, there have been recent attempts to apply
the preoccupation model to real-world health issues that often are kept
secret, such as abortion and eating disorders. Major and Gramzow (1999)
noted that having an abortion often is kept secret for good reason, given
that one in five adult women in the United States has had an abortion and
yet roughly half the adult population views abortion as the (im)moral
equivalent to murdering a child. They investigated the stigma of abortion
and the psychological implications of concealing the abortion among 442
women followed for 2 years from the dates of their abortions. As expected,
the women who felt stigmatized by abortion were more likely to report
that they felt the need to keep it a secret from family and friends. Keeping
the abortion a secret was associated with suppressing thoughts of the
abortion and with disclosing abortion-related emotions to others less fre-
quently. Increased thought suppression was associated with experiencing
more intrusive thoughts of the abortion. Both suppression and intrusive
thoughts surrounding the abortion, in tum, were positively related to
increases in psychological distress over the 2-year period. Emotional dis-
closure moderated the association between intrusive thoughts and distress
in the following way: Disclosure was associated with decreases in distress
among women experiencing intrusive thoughts of their abortion, but it
was unrelated to distress among women not experiencing intrusive
thoughts. Major and Gramzow (1999) concluded that secret keeping put
women at risk for experiencing increased intrusiveness of thoughts of the
abortion. However, it is important to note that the secrecy, thought-
suppression, and intrusive-thoughts measures were all administered at
the same time (i.e., at 2-year follow up), leaving open the possibility that
intrusive thoughts led to thought suppression. Moreover, contrary to Major
WHY SECRECY IS LINKED TO PROBLEMS 51

and Gramzow's predictions, intrusive thoughts were positively associated

with disclosing emotions surrounding the abortion. Also, after the re-
searchers statistically controlled for the women's personal conflict over the
abortion and their positive and negative affectivity in general, secrecy
surrounding the abortion was not associated with greater distress.
Smart and Wegner (1999) examined the psychological effects of con-
cealing the stigma of having an eating disorder during an interaction with
another person. They recruited undergraduate female participants who
either did or did not have eating-disordered characteristics. The partici-
pants were asked to play the role of someone who did or did not have an
eating disorder in an interview in which they would be answering ques-
tions that were related to the eating disorder. In Study 1, participants
with an eating disorder who role-played not having an eating disorder
exhibited more secrecy, suppression, and intrusive thoughts of their eating
disorder and more projection of eating disorder-related thoughts onto the
interviewer than did those participants with an eating disorder who role-
played someone with an eating disorder or those participants without an
eating disorder who role-played someone without an eating disorder. In
Study 2, the researchers found both increasing cognitive accessibility of
eating disorder-related words during the interview and after the interview
among those participants with concealed stigmas. The researchers con-
cluded that the participants who concealed their stigmas became preoccu-
pied with the control of stigma-relevant thoughts (Smart & Wegner, 1999).
Closer scrutiny of their data, however, shows that in Study 2 among
the participants with eating disorders there was no difference in intrusions
of the thoughts between those who played the role of someone without a
stigma and those who played the role of someone with a stigma. Also,
although scores on the cognitive accessibility measure increased more
sharply during the interview for those with a concealed stigma, the scores
were actually higher overall for the participants with an eating disorder
who played the role of someone with an eating disorder. The only clear
and consistent finding that supported the researchers' hypotheses sur-
rounding the preoccupation model across Studies 1 and 2 was that secrecy
was linked to greater thought suppression. In fact, in both Studies 1 and
2, judges rated the participants who played the role of someone without an
eating disorder as less neurotic and more comfortable during the interview
than those who played the role of someone with an eating disorder. The
data in Study 2 on cognitive accessibility via a word completion task
showed that at the end of the interview participants with eating disorders
had high accessibility of eating disorder-related words relative to partici-
pants without eating disorders, whether they were in the no-stigma or
stigma role playing conditions. Taken together, these findings do not
52 CHAPTER 3

provide support for the preoccupation model. More important for this
volume, interpersonal impressions were enhanced by the concealment.
Also, in Study 2 participants with an eating disorder who played the role
of someone without an eating disorder did not show impairments in their
cognitive functioning. The researchers suggested that a possible explana-
tion for this finding that contradicted their model was "that the women
with actual EDs [eating disorders] in this study were so practiced at
secrecy that their cognitive functioning was not influenced by their high
level of preoccupation with the stigma" (Smart & Wegner, 1999, p. 484).
Jeffrey Kahn and I earlier had made a similar claim that practice at
suppression can actually reduce its negative cognitive effects (Kelly &
Kahn, 1994). We demonstrated that undergraduate participants who sup-
pressed their own intrusive thoughts experienced a reduction in intrusions
of the thoughts relative to participants who expressed their own intrusive
thoughts (Exp. 1). Thus, we challenged the idea that over time suppressing
can lead to preoccupation with a secret. We suggested that perhaps with
experience suppressing their private thoughts, people develop strategies
and techniques that become virtually automatic and require little thought
or effort (Kelly & Kahn, 1994). Such experience controlling their thoughts
could make keeping secrets from others less difficult, so that people could
avoid sharing secrets and the rejecting feedback that is associated with
sharing altogether.
Additional findings that were contrary to Lane and Wegner's (1995)
and Major and Gramzow's (1999) conclusions about the link between
secrecy and thought intrusions (i.e., contrary to the preoccupational
model) were the findings described in Chapter 1 from Finkenhauer and
Rime's (1998a) studies, in which participants' ratings of events that they
kept secret and events that they had shared were compared. As mentioned
in Chapter 1, participants in both Studies 1 and 2 indicated that they did not
ruminate more over an event that they kept secret as compared with one
that they had shared with others. However, Finkenhauer and Rime's
(1998a) studies were limited by the fact that participants were asked to
generate only one emotional event in either the shared- or non-shared-
event categories. The participants might have picked a shared (Le., non-
secret) emotional event that was not representative of their shared emo-
tional events in general. The fact that it came to mind when they were
asked to pick an event during the study suggests that they were thinking
and maybe even ruminating about that event more than their other shared
events.
Fish and Scott (1999) obtained findings that more clearly contradicted
the preoccupation model. They investigated the relation between forget-
ting childhood abuse and the level of secrecy surrounding the abuse, and
WHY SECRECY IS LINKED TO PROBLEMS 53

they discovered that self-reported forgetting and secrecy were positively

related, not negatively related. Specifically, among a nonclinical sample of
423 adults (mean age 43.4 years) who completed a questionnaire regarding
childhood abuse history, 32% reported childhood abuse. Of those report-
ing abuse, 52% noted periods of forgetting some or all of the abuse, and
76% indicated there had been a time when no one but themselves and their
abuser knew about the abuse. Also, 47% of the respondents reporting
abuse indicated that an abuser tried to get them to keep the abuse a secret.
(That is, 47% said that there was a time when no one else but the abuser
knew the secret.) Most important for the present discussion, the respon-
dents who had periods of forgetting the abuse, as compared with those
who had continuous memories of the abuse, were more likely to report
that there was a time when the abuser tried to get them to keep the abuse
secret.
These findings make sense in light of Koutstaal, Schacter, Johnson,
Angell, and Gross's (1998) finding that mentally reviewing events actually
increases memory of those events. These researchers examined the degree
to which adults' memory for everyday events was improved by later
seeing photographs or reading brief verbal descriptions of those events.
Both older and younger adults watched everyday events on a videotape.
When the participants reviewed the events with either verbal description
or photographs, as compared with when they did not have any review,
they recalled more events and in greater detail. Thus, the researchers
demonstrated that recalling an event at one time often increases the likeli-
hood that it will be remembered at a sti1llater time. Verbal descriptions
enhanced later recall to the same degree as reviewing photographs. Kout-
staal et al. (1998) concluded that postevent review has clear potential
practical benefits for improving memory.

SUMMARY

What can the reader make of these conflicting findings on the associa-
tion between keeping an event a secret and ruminating over the secret
event, and thus of the preoccupation model of secrecy? There is only
mixed support for the idea that people tend to ruminate more over events
that they keep secret. In several correlational studies, thought suppression
has been associated with secrecy. Suppression also has been linked with
increased intrusiveness of the suppressed thoughts (e.g., Lane & Wegner,
1995; Major & Grarnzow, 1999). However, it may well be that people
suppress thoughts that are intrusive, rather than the thoughts that they
suppress become intrusive. Wegner addressed this possible concern by
demonstrating in several experiments that suppressing a thought, such as
54 CHAPTER 3

the thought of a white bear, led to increased preoccupation of that thought

(e.g., Wegner et al., 1987) and that suppressing thoughts made them hyper-
accessible (e.g., Wegner & Erber, 1992). However, Jeff Kahn and I (Kelly &
Kahn, 1994) found the opposite result with people's own unwanted intru-
sive thoughts. Moreover, to date, the long-term effects of thought suppres-
sion and secret keeping have not been assessed directly (i.e., experimen-
tally), only correlational assessments of the long-term link between these
two variables have been conducted. Major and Gramzow (1999) studied
the relation between keeping a secret about abortion and distress and
found no significant association when they controlled for key personality
variables. Given the conflicting findings, it is far from clear that keeping an
event secret is associated with becoming more troubled by that event.

SELF-PERCEPTION MODEL

In his influential self-perception theory, Bern (1967, 1972) postulated

that because people's internal states often are ambiguous (e.g., anger and
sexual arousal are quite similar emotions at a physiological level), people
look to their behaviors to determine these internal states. For example, if
a woman kisses a man whom she is not certain that she is attracted to,
she may say to herself, "Well, I kissed him, therefore I must be attracted
to him."
Some researchers have used self-perception theory to explain why
secret keeping is linked to problems. Specifically, people, such as those
who have been raped, feel shame about having experienced the stigma-
tizing traumatic event, and thus they choose to conceal the event from
others (Derlega, 1993; Pennebaker, 1985, 1989; Silver, Boon, & Stones, 1983).
Such individuals may then tell themselves that because of the fact that they
have hidden the experience from others, the event must indeed be very
negative or shameful (Derlega, 1993). It is through this self-perception
process (Bern, 1967, 1972) that people might develop lowered self-esteem
(Derlega, 1993). Ichiyama et al. (1993) used this idea to explain why low-
ered self-esteem was linked to self-concealment among a sample of college
students.
There is correlational evidence that secrecy is linked to shame. For
example, as mentioned in Chapter 1, Finkenhauer and Rime (1998a) found
that participants indicated higher ratings of guilt and shame over non-
shared events as compared to shared events. That finding is consistent
with the findings from Frable, Platt, and Hoey's (1998) study in which 86
Harvard undergraduates were asked to describe the who, what, and
where of their daily lives over 11 days and to rate their momentary self-
esteem and affect. Those participants with concealable stigmas (students
WHY SECRECY IS LINKED TO PROBLEMS 55

who indicated that they were gay, that they were bulimic, or that their
family earned less than $20,000 each year) reported lower self-esteem and
more negative affect than both those whose stigmas were visible and those
without stigmatizing characteristics.
To try to demonstrate whether concealment causes a negative evalua-
tion of the concealed information, Fishbein and Laird (1979) conducted a
clever experiment in which they induced participants either to reveal or
conceal an ambiguous piece of information (i.e., their score of +4.6 on an
intelligence test) from a confederate. The participants in the concealment
condition later rated that score as more negative than did participants in
the disclosure condition. Fishbein and Laird discussed the implications of
this finding in terms of self-perception theory (Bern, 1972). Specifically, the
participants looked to their actions (i.e., their revealing or concealing
information) to determine the meaning of the ambiguous information. The
researchers extended these findings to say that people might run the risk
of seeing themselves in unfavorable ways if they choose to hide informa-
tion about themselves from others. They speculated that such concealment
of information over time can lead to lowered self-perceptions of one's self-
worth. The reader should note that one problem with that study was that
the participants knew that the experimenter was aware of their scores
when the experimenter advised them either to reveal or conceal them. As
such, the participants in the concealment condition may have thought
that the experimenters were trying to protect them from embarrassment in
suggesting that they not mention their score to the" other participant" (i.e.,
confederate). Therefore, before feeling confident in the finding that secrecy
actually caused a negative evaluation of the secret information, re-
searchers will have to replicate these findings in a study without that
confound. Regardless, the findings offer the most direct support for nega-
tive self-perception processes involved in concealment thus far.
Khersonskaya (2001) conducted a similar experiment that assessed
what happened when participants were asked either to reveal or to con-
ceal ambiguous information. Like Fishbein and Laird (1979), she found
that participants in the conceal condition were less willing to have their
scores be made public. However, she also found no differences in their
satisfaction with their scores for the participants in the reveal versus
conceal conditions. Khersonskaya suggested that participants were less
willing to reveal simply to be consistent with their earlier instruction,
rather than because they viewed concealed material more negatively.

SUMMARY

Self-perception theory has been used to explain how people may

come to view negatively the information that they keep secret (e.g., Fish-
56 CHAPTER 3

bein & Laird, 1979; Ichiyama et al., 1993). Simply observing one's own
secret keeping behavior may make one assume that the hidden informa-
tion should be kept secret and may even lower one's self-esteem.

DIMINISHES SOCIAL SUPPORT

Larson (1993b) suggested that self-concealment is problematic on two

fronts: It in itself causes distress and it prevents people who are in need of
social support from receiving it. In fact, as mentioned in Chapter 2, self-
concealment has been found to be both positively related to depression
and negatively related to perceived social support (e.g., Kahn & Hessling,
2001). Larson (1993b) noted that people in stressful situations, such as
patients and staff in an oncology clinic, are faced with the dilemma of
whether to conceal or reveal the distress they are experiencing. According
to Larson (1993b), self-concealment as a coping response increases stress
and simultaneously diminishes the likelihood of helpful, empathic re-
sponses from others. Therefore, he concluded that oncology staff must find
safe contexts and confidants for sharing their inevitable fears, self-doubts,
and uncomfortable feelings and must help patients resolve the distress-
disclosure dilemma that attends the cancer experience.
Although Larson (1993) was not suggesting that the lower levels of
social support cause the distress surrounding secret keeping, he was sug-
gesting that the lack of social support accompanying the concealment
prevents one from recovering from distress. However, in a study that I
conducted with John Achter (Kelly & Achter, 1995), we found that whether
or not we statistically controlled for participants' social support levels,
self-concealment was strongly positively associated with depression. Like-
wise, even when Larson and Chastain (1990) statistically adjusted for
levels of social support, they still found that self-concealment was related
to physical and mental symptoms. The same was true in Cole and co-
workers' (1996b) study of HIV-positive gay men who were either in or out
of the closet. Even statistically controlling for levels of social support, Cole
et al. found that being in the closet predicted faster progression of HIV
infection. These findings do not support the notion that it is through
reduced social support that self-concealment leads to physical and psy-
chological symptoms.

SUMMARY

Several explanations have been offered in the psychological literature

for why secret keepers, as compared with nonsecret keepers, tend to have
WHY SECRECY IS LINKED TO PROBLEMS 57

more mental and physical problems. These explanations are the inhibition,
preoccupation, diminished social support, and the self-perception models
of secrecy. Regarding the inhibition model, Pennebaker has conducted a
series of experiments that have demonstrated that revealing secrets leads
to health benefits (see Chapter 4, this volume), but it does not follow that
concealing secrets necessarily leads to health problems. To date, there is
no direct, experimental evidence that secrecy causes health problems.
Researchers have explored the possibility that inhibiting behavioral ex-
pression of one's emotions is physiologically taxing, but the evidence is
quite mixed. The preoccupation model of secrecy also does not offer a
satisfactory explanation for the link between secrecy and problems. There
is evidence that with new thoughts, such as the thought of a white bear,
suppressing the thoughts leads to increased preoccupation with the
thoughts (Wegner et al., 1987). However, people do not seem to become
more preoccupied with their own intrusive thoughts as a result of sup-
pressing them (Kelly & Kahn, 1994). The self-perception and diminished
social support models have not yet received much testing. The little testing
the social support model has received does not point to the notion that it
is through lowered levels of social support that secret keeping is associated
with problems.

A PREDISPOSITIONAL EXPLANATION

Thus far, none of the proposed models has offered a compelling

explanation for the link between secrecy and problems, given the mixed or
limited findings for each. An element that has been missing from this
discussion is the notion that genetics may playa central role both in why
people inhibit social expression and why those same people are likely to
have more physical and psychological problems. Pennebaker (1993) noted
that inhibition can reflect genetic proclivities to be hypersensitive to novel
stimuli and that individuals predisposed to be hypersensitive tend to
engage in processes of inhibition that can impede the normal cognitive
changes that help them cope with upsetting experiences. He observed that
inhibition processes can completely disrupt normal social interaction, re-
sulting in further isolation by the traumatized individual. I speculate that
a more Simplified version of this explanation could fit the existing evi-
dence on the correlates of self-concealment: It is possible that the kind of
person who has the genetic predisposition to inhibit social expression also
may be the type of person who is more vulnerable to developing the kinds
of problems that have been linked with self-concealment. Along these
lines, it may be the case that keeping a particular secret per se is not
58 CHAPTER 3

necessarily problematic; keeping a secret may not cause the symptoms that
have been associated with self-concealment.

INHIBITED TEMPERAMENTAL TYPES

Indirect support for my suggestion comes from research by Kagan

(1994) and others (e.g., Kagan, Reznick, & Gibbons, 1989; Suomi, 1991), who
have provided evidence for a part genetic origin to what they call inhibited
and uninhibited temperamental types. Kagan (1994) identified two behav-
ioral profiles linked to these two temperamental types. Inhibited children
tend to be reluctant to initiate spontaneous comments with unfamiliar
people, show caution in situations requiring decisions, smile less, have
unusual fears and phobias, show large heart rate accelerations to stress,
and have atopic allergies. It is estimated that two out of every ten healthy
Caucasians inherit a physiology that leaves them inclined to be aroused
and disturbed by stimulation such as a colorful mobile early in the first
year and initially avoidant of novelty in the second and third years (Kagan,
1994). This physiology in inhibited children is thought to be a more reactive
circuit from the limbic area of the brain to the sympathetic nervous system.
Specifically, this more reactive circuit involves a low threshold of excit-
ability in the central and basolateral areas of the amygdala and in the
projections from these areas to cortical, motor, and autonomic targets. This
low threshold is believed to leave inhibited children especially vulnerable
to the state of fear of the unfamiliar. In contrast, about four out of every ten
children seem to inherit a physiology that biases them to be relaxed at 4
months and relatively fearless in early childhood (Kagan, 1994).
Kagan (1994) has demonstrated that infants who display a high-
reactive behavioral profile at 4 months (i.e., those who arch their backs,
pump their limbs, and fret or cry in response to novel stimuli) show higher
levels of early childhood fears of things such as unfamiliar people, un-
familiar facial expressions, and intrusions into their personal space than
do children who display a low-reactive profile. In tum, high fear at 9
months, which was observed in only 10% of the infants, was a good
predictor of high fear at 14 and 21 months. The relation between early
reactive profiles and later categorization of inhibited typologies is as fol-
lows: Half the low-reactive infants displayed uninhibited profiles at 3Yz
years, whereas 14% were inhibited; 40% of the high reactives were inhib-
ited at 3Yz years, whereas 25% displayed uninhibited profiles (Kagan,
1994).
Moreover, it has been demonstrated that there is predictive utility of
the typologies from age 7 to adolescence (Kagan et al., 1989). Using a
combination of observers' ratings of spontaneous conversation and smil-
WHY SECRECY IS LINKED TO PROBLEMS 59

ing to create indexes of an inhibited or uninhibited style, Kagan et al. (1989)

found that two thirds of the children who had remained in the inhibited
category from the second to the seventh year were quiet and serious
around age 13. Among the children who had remained uninhibited be-
tween the second and seventh year, 40% remained uninhibited from age
7 to age 13. He argues that such early demonstration of the existence of the
two typologies and relatively high consistency for the typologies across
one's childhood support the idea that the typologies may have stemmed in
part from genetic structure. Specifically, he argues that one's genetic
makeup affects whether one will have an overaroused limbic system.
Some children-those who are inhibited-show very early signs of hav-
ing an overresponsive amygdala. Kagan contends that this heightened
sensitivity leads them to avoid conflict and the stimulation of novel inputs.
Support for this argument can be found in studies of identical twins who
show similarity in inhibited or uninhibited behavioral patterns and found
in research that has demonstrated consistent associations between the two
typologies and their sympathetic reactivity and asymmetry of cerebral
activation. In particular, high-reactive infants showed increase sympa-
thetic activity in the cardiovascular system even before 4 months. Reac-
tivity to novel stimuli, as evidenced by limb thrashing, arching of the back,
and so on, at 4 months was a good predictor of later childhood fears (see
Kagan, 1994).
Given that there is evidence of a genetic component to the develop-
ment of an inhibited personality and that central elements of this tempera-
mental type are that the person inhibits social expression, has more fears,
and tends to be less cheerful and more serious than the uninhibited type,
it is reasonable to think that having symptoms such as anxiety, physical
pain, and depression may be linked to a partly biologically based tendency
to self-conceal.

POSSESSING A SECRET Is NOT NECESSARILY RELATED TO SYMPTOMS

Additional evidence for the predisposition argument comes from

research that suggests that keeping a secret is not related to symptoms
when researchers statistically account for other dimensions of conceal-
ment. In a series of studies, Jeff Kahn and I (Kahn & Kelly, 1998) found that
the Self-Concealment Scale was composed of three separate factors, which
we labeled the Possession of a Distressing Secret, Apprehension about
Disclosure, and Self-Concealment Tendency. The first factor, the Posses-
sion of a Distressing Secret, seems to reflect the extent to which an indi-
vidual holds some important, distressing secret. Within this factor, there
is no discrimination between holding one secret or more than one; items
60 CHAPTER 3

that assess both load on this factor. An example is, "I have an important
secret that I haven't shared with anyone." The second factor of the Self-
Concealment Scale, Apprehension about Disclosure, seems to reflect a
person's fears and concerns about the consequences of disclosing negative
information. Apprehension about Disclosure pertains specifically to the
fear a person reports experiencing when faced with revealing personal
information. An example is, "I'm often afraid I'll reveal something I don't
want to." Third, Self-Concealment Tendency reflects more of a chronic
behavior of withholding personal information. This factor does not reflect
the possession of a secret or fears about disclosing information; rather,
Self-Concealment Tendency seems to describe a behavior of concealing
potentially negative self-relevant information from others. An example is,
"When something bad happens to me, I tend to keep it to myself."
Consistent with Finkenhauer and Rime's (1998b) findings that people
who could recall a secret memory were sicker than those who could not,
we (Kahn & Kelly, 1998) found that the Possession of a Distressing Secret
was significantly correlated with symptomatology, as measured by the
Brief Symptom Inventory (r = .28). Apprehension about Disclosure and
Self-Concealment Tendency also were significantly correlated with symp-
tomatology scores (rs = .42 and .26, respectively). However, after conduct-
ing a multiple regression analysis in which all three factors of the Self-
Concealment Scale were entered into a model to predict symptomatology
scores, we discovered that although Apprehension about Disclosure still
was Significantly positively related to symptomatology (standardized B =
.38), Possession of a Distressing Secret and Self-Concealment Tendency
were not (standardized I3s = .04 and .07, respectively). This pattern of
findings suggests that holding a particular secret does not relate to psychi-
atric symptoms when other dimensions of self-concealment are accounted
for statistically. It seems that the link between self-concealment and symp-
tomatology observed by many researchers may be driven by this Appre-
hension to Disclose factor, not by possessing a particular secret. Thus, the
widely accepted idea that keeping particular secrets is stressful and may
lead to symptoms (see Pennebaker, 1989, 1990) begins to be challenged by
these data. However, the findings will need to be replicated before re-
searchers can feel confident about this pattern. I am especially cautious
because there were only two items in the Self-Concealment Tendency
factor, making that subscale less than optimally reliable.
In another study, mentioned in Chapter 1, I (Kelly, 1998) assessed
secret keeping and symptomatology levels in a sample of 42 therapy
outpatients. As could be expected, the participants' self-concealment
scores were positively related to their symptomatology scores (r = .37)
and to their reports of whether they were keeping a relevant secret from
WHY SECRECY IS LINKED TO PROBLEMS 61

the therapist (r = .42). However, when I statistically controlled for the

clients' self-concealment scores, I found that keeping a particular relevant
secret from the therapist was associated with a significantly greater reduc-
tion in symptoms across an average of 11 therapy sessions. This finding
supports the idea that although the type of person who keeps secrets has
greater symptomatology levels, keeping a secret per se is not necessarily
problematic (see Kelly, 2000a).
Kahn, Achter, and Shambaugh (2001) measured 45 counseling center
clients' predispositions to disclose (versus conceal) psychological distress
at intake and then measured their self-reported stress and symptomatol-
ogy at termination, which occurred on average after six or seven sessions.
As it turned out, the clients who scored higher on distress disclosure and
lower on concealment at intake experienced a greater reduction in self-
reported stress and symptomatology over the course of counseling. Kahn
et a1. (2001) interpreted their findings to mean that "clients who come to
their intake session with a tendency to share personally distressing infor-
mation with others possess important interpersonal qualities that would
likely facilitate the counseling process" (p. 200). However, I suggest that
their findings may be explained in a more parsimonious fashion as a
classic regression to the mean phenomenon. Specifically, the clients who
were higher in distress disclosure and lower on self-concealment may
have started out in a state of unusually high distress for them, and as might
be expected just with the passing of time, they became less distressed than
their high self-concealing counterparts. In essence, I am suggesting that
they regressed to their mean state of being healthy, whereas the more
chronically symptomatic high self-concealers stayed less healthy, indepen-
dent of any process of concealment of distress during their treatment.
The reader may be skeptical about my separating the process of
keeping a particular secret from general self-concealment in predicting the
outcomes of concealment. This skepticism is reasonable given the findings
described in the previous chapter that keeping a particular secret about
being gay (Cole et a1., 1996a,b) was predictive of illnesses years later and
that people who could recall a secret memory were sicker than those who
could not (Finkenhauer & Rime, 1998b). However, I suggest that these
previous patterns could be explained by the fact that the researchers did
not statistically control for what Kahn and I (Kahn & Kelly, 1998) labeled as
the Apprehension to Disclose, which may be a similar construct to the
general predisposition to be inhibited described by Kagan (1994). This
predisposition to be inhibited is likely to be linked with both keeping a
particular secret and symptomatology, and as such the relation between
keeping a particular secret and symptoms may be a spurious one. If what I
am suggesting-that the type of person who keeps secrets is vulnerable to
62 CHAPTER 3

illnesses and that secret keeping per se is not necessarily illness inducing-
is true, then other researchers who control for this factor in the future will
probably not find a link between keeping a particular secret and illness.

SocIAL ANXIETY Is RELATED TO SYMPTOMS

In another correlational study, undergraduates were asked to report

on how much they keep secrets and to complete the Self-Concealment
Scale (Gesell, 1999). These scores were combined to create a composite
secret keepers score. Then, a number of personality variables, including
social anxiety, manipulativeness, need for control, and sociability, were
entered into an analysis that was designed to predict secret-keeping ten-
dency. Social anxiety was most strongly associated with secret keeping
(standardized 13 = .66). In addition, a separate analysis entering social
anxiety and secret keeping together to predict symptomatology showed
that whereas secrecy in and of itself still predicted symptomatology scores,
the personality dimension of social anxiety accounted for much of the
predictive effect that secrecy has on symptomatology. Specifically, social
anxiety could explain 80% of the variance in symptomatology. In other
words, self-concealers tend to be socially anxious, and it is the social
anxiety that is more strongly related to symptomatology than secret keep-
ing. I see this finding as further (indirect) support for the notion that
inhibited personality types are more susceptible to symptoms, and that it
is not secret keeping per se that causes illness.

SHY OR INHIBITED PERSONALITY TYPES

In the previous chapter, I reviewed a number of findings on the

correlates of self-concealment and repression. Given that Kagan addressed
the inhibited temperament in his research, which involves shyness and the
inhibition of behavioral expression, I now provide evidence that shy or
inhibited personality types do tend to have greater illness rates than do
uninhibited type.
In one study, women with rheumatoid arthritis rated themselves as
definitely more shy and inhibited than their healthy sisters (Moos & Salomon,
1965). In another study, Biederman et al. (1990) examined psychopatho-
lOgical correlates of behavioral inhibition in 30 children (aged 4-7 years)
who were high-risk offspring of parents with panic disorder and agora-
phobia and 20 healthy controls (aged 4-20 years). The participants were
evaluated with the Diagnostic Interview for Children and Adolescents-
Parent Version. As it turned out, the inhibited participants, as compared
with the healthy controls, had higher rates of multiple anxiety disorders,
WHY SECRECY IS LINKED TO PROBLEMS 63

accounted for by high rates of overanxious disorders and phobic disor-

ders. The researchers concluded that behavioral inhibition to the unfamil-
iar may be one of multiple risk factors contributing to the development of
childhood anxiety disorders.
Schmidt and Fox (1995) also demonstrated the relation between ex-
treme shyness and emotional and psychosomatic problems. They asked
female undergraduates to rate themselves on measures of shyness, soci-
ability, personality variables, and the prevalence of various disorders and
then to interact with a female confederate. After the interaction, they were
rated by independent observers for their level of shyness. Extreme shy-
ness, as rated by the participants themselves and the observers, was found
to be a significant predictor of emotional and psychosomatic problems,
such as allergies and gastrointestinal functioning. It also was associated
with self-reported depression, loneliness, fearfulness, social anxiety, neu-
roticism, inhibition, and low self-esteem. Thus, consistent with Kagan's
(1994) findings with children, a picture of the shy, inhibited person with a
number of physical and psychological ailments emerged from this study.
Finally, Windle (1994) too found that behavioral inhibition was linked
to problems in a sample of 4462 male US military veterans. They measured
inhibition using an index of behavioral inhibition on the Minnesota Multi-
phasic Personality Inventory (MMPI). Higher behavioral inhibition was
associated with somewhat higher levels of cortisol, lower levels of social
support, and a higher prevalence of lifetime generalized anxiety and major
depressive disorders.

SUMMARY

I have offered a part-genetic predispositional explanation for why

secret keepers tend to have high levels of symptomatology. This explana-
tion is based largely on Kagan's findings that there are very early, rela-
tively stable signs of inhibited and uninhibited temperaments. Given that
the types show up so early, it seems that there may be a genetic element to
them. Kagan and other researchers have shown a link between shyness-
inhibited personality and symptoms such as allergies and fears. Based on
these findings and on preliminary findings that keeping a particular secret
is not linked to symptoms when the general apprehension to disclose is
accounted for statistically, I am suggesting that the kind of person who
keeps secrets may be vulnerable to illness, rather than that secret keeping
per se causes problems. This emphasis on a part-biological explanation to
account for the covariance of self-concealment and symptoms fits rather
neatly into a broader background of trait-based approaches to personality.
After observing the findings from the many family, adoption, and twin
64 CHAPTER 3

studies conducted by different researchers across the globe (e.g., Jang,

McCrae, Angleitner, Riemann, & Liveslez, 1988; Loehlin, 1992), Costa and
McCrae (1998) concluded that genetic influences account for 40 to 50% of
the variance in measured personality traits. However, McCrae and Costa
(1995) acknowledged that even if personality traits have a biological basis,
they still are psychological phenomena that must be understood in terms
of how they interact with people's actions and experiences.
Whether or not this predisposition to be inhibited is part-genetically
based as Kagan's findings suggest, my point is that the link between
keeping a secret and becoming ill may be a spurious one that perhaps
could be explained by the fact that each of these variables (Le., keeping
a secret and becoming ill) is linked to the predisposition to be inhibited. It
is important to note, however, that the evidence to support this idea is
correlational, and as such it is not possible to say with any certainty that
such a predisposition causes people both to keep secrets and to become ill.
This idea may best be viewed as another of a series of possible explana-
tions for the link between secrecy and problems that requires much more
empirical testing. Another limitation to my proposed idea is that I have
used the terms "apprehension to disclose," "social anxiety," "shyness,"
and "inhibited temperament" loosely to refer to a common underlying
personality construct, and future researchers will need to tighten the
definition and assessment of this dispositional factor before further exam-
ining the predispositional idea.

CONCLUSION

In the chapter, I reviewed several competing theories for why secret

keeping is linked to problems and presented the mixed evidence concern-
ing each of these theories. I also offered another possible explanation
based on the idea that secret keepers may have the biological predisposi-
tion to become sick. However, the reader may still be thinking, "Yes, but
I know that secret keeping is stressful and likely to cause problems."
Katherine Power's story of being on the lam for years seemed to illustrate
this point so well. It is difficult to pinpoint cause-and-effect links in such
cases, however, because it may have been her fear of getting caught rather
than the secrecy per se that drove her to become anxious and depressed.
Also, in her case, the secret involved something she owed to others: prison
time. She probably felt guilty about her crime and as though she deserved
to be punished for it. Her feelings of guilt alone could have made her
depressed. Imagine if the situation had been identical except that everyone
knew about her crime. Imagine too that prosecutors were spending years
WHY SECRECY IS LINKED TO PROBLEMS 65

deciding whether to seek to convict her. Under such circumstances-

awaiting the prospect of punishment without secrecy-she still may have
become depressed. For researchers to investigate the harmful effects of
secrecy, they will need to tease secrecy apart from feeling guilty or feeling
that something is owed. Also, it will be interesting to see how fears of
being caught playa role in the effects of secrecy. Perhaps if there had been
some statute of limitations on murder, such that she could not have been
prosecuted after her 23 years on the run, Katherine Powers might have
been able to keep her secret and not become depressed.
A very promising research direction is to see how one's level of self-
concealment interacts with a decision to reveal or conceal a secret in
influencing the outcomes of that decision. I predict that, on average, the
kind of person who normally conceals and then chooses to reveal a secret
will be worse off in terms of physical and mental health than a person who
normally reveals and then chooses to conceal a specific secret. I say this
partly because of the evidence that high self-concealers tend to ruminate
more than do low self-concealers; I suspect that they would worry about
the implications of their revelations long after the revealing. I also make
this prediction based on the evidence that high self-concealers would start
out sicker than low self-concealers before any decision to reveal or conceal
a secret would even be made.
CHAPTER 4

HEALTH BENEFITS
OF REVEALING

Psychologists and laypersons alike believe that" confession is good for the
soul," a theme that frequently has been depicted on talk shows and in
popular movies. For example, in the popular 1996 British film. "Secrets and
Lies," Hortense, a young black optometrist, sets out to find her birth
mother after her adoptive mother dies. Hortense is disturbed to find out
that her birth mother, Cynthia, is white. Cynthia is a factory worker who
lives in a run-down house with her whiny, moody daughter Roxanne.
Hortense calls Cynthia, who initially bursts into tears and refuses to see
her. Cynthia has difficulty acknowledging (first to herself, then to others)
that Hortense is her daughter. One of her first reactions to seeing Hortense
is a sincere denial that she has ever "been with a black man." But before
long, a suppressed memory jolts into her consciousness. Eventually, the
two develop a warm friendship, and Cynthia invites Hortense to a party
with her family and friends. One of the characters in the film notes that it is
"Best to tell the truth, isn't' it? That way nobody gets hurt." At the party,
Cynthia pretends that Hortense is her factory co-worker, which leads her
to ever-more convoluted lies. The film climaxes with all the secrets and lies
finally spilled out and all the people at the party hugging one another.
Cynthia says near the end, "This is the life, innit?" The intention of the
filmmakers seemed to be to demonstrate how difficult it can be to face up
to ugly truths, even while the evasions and unspoken grievances slowly
choke the secret keepers, but that in the end it is better to reveal these
truths. Finally embracing her daughter and coming forward with the truth
brings both mother and daughter new insights and joy.
In the December 1995 movie with a similar theme, "Dead Man Walk-
67
68 CHAPTER 4

ing," the male protagonist Matthew Poncelet brutally kills a teenage cou-
ple but denies responsibility for his crimes. In the end, he finally does
confess the truth to his spiritual advisor, Sister Helen Prejean, and the
audience gets the sense that despite his execution, this confession has
freed his soul.
Matthew Poncelet's character was largely based on the real-life Robert
Lee Willie, who was executed in Louisiana in 1980 for the kidnapping and
killing of a teenage girl. Sister Prejean had already learned many of the
details of his crimes by reading police reports, which included a confession
in which Willie provided a great deal of information to the deputy sheriff,
Mike Varnado, who investigated these crimes. Like movie makers, police
officers frequently appeal to people's beliefs in the benefits of open confes-
sion. This is especially evident when they coax confessions out of sus-
pected criminals under the pretext that the criminals will feel better after-
ward. The following are excerpts from a transcript of a Public Broadcasting
Service special on these crimes:

MIKE VARNADO: According to Willie, he took this knife and he cut her throat,
like this. And in his version he gave me in the confession, he said Joe was
between her legs, taking the knife and jugging her as deep as it would go in
her throat. I was outraged immediately that they would bring this girl up in
here-this is her home-and do these-do these vile things to her.. ,. And
the key to him confessing is-he asked me a question. He said, "I guess I'm
a big man" or "I'm making the headlines down there a lot" and things like
that. And I said, "Yeah." I said, "You are." I said, "You could be like Jesse
James," you know? And he said, "Yeah, I'll tell you about it. Yeah, I killed her."
ROBERT LEE WILLIE: [police audiotape] I asked her, I said, "Do you want a ride?"
She said yes. So she got in the middle of the seat, between me and Joe, and
we rode around and went up to Frickie's Cave and-
MIKE VARNADO: Willie showed absolutely no remorse through the whole
thing. None. He was proud of what he had done. He talked to me like this
was a Sunday afternoon football game we were diSCUSSing.
ROBERT LEE WILLIE: [police audiotape] He says, "You know where we can go
fuck this whore?"
MIKE VERNADO: He didn't have any-any problem telling me what they had
done, the brutal details. The problem he had was actually owning up to
being the one that actually cut the girl's throat. I guess he felt awkward
about doing that.
ROBERT LEE WILLIE: [police audiotape] Joe blindfolded her and we went down in
the bottom of the hill and Joe made her lay on the ground and he had this big
old knife and he just cut her throat and just started jugging her in the throat
with it, man-just jugging her and jugging her and- (http://www.pbs.
org/ wggh/ pages / frontline / angel! angelscript.html; retrieved May 14,
2000)
HEALTH BENEFITS OF REVEALING 69

Whereas Willie's confession to the police may have been motivated

by his desire to appear dangerous and powerful, his confession to Sister
Prejean seemed to be motivated by a genuine desire to purify himself of
his terrible deeds. Even though ironically Willie's confession to the police
almost surely contributed to losing his life to execution, revealing secrets
has been shown to be beneficial under certain circumstances.
As mentioned in the previous chapter, it is difficult from an ethical
standpoint to try to demonstrate experimentally that keeping secrets
makes people become ill. However, researchers have been able to study
whether revealing secrets leads to health benefits. Investigators conduct-
ing experiments in confidential and anonymous settings have shown that
talking or writing about private traumatic experiences leads to health
benefits, such as fewer physician visits (Pennebaker & Beall, 1986; Pen-
nebaker, Colder, & Sharp, 1990), improved immunological functioning
(Pennebaker, Kiecolt-Glaser, & Glaser, 1988; Petrie, Booth, Pennebaker,
Davison, & Thomas, 1995), and longer survival periods in advanced breast
cancer patients (Spiegel, Bloom, Kraemer, & Gottheil, 1989). In this chapter,
I review this remarkable evidence in detail, describing the many experi-
ments that have established a cause-and-effect relationship between re-
vealing and healing.
It is worth emphasizing that the experiments reviewed in the follow-
ing paragraphs usually took place in confidential and anonymous settings,
thus avoiding the confounding influence of confidant feedback on the
health effects of revealing. It also is worth noting that many of the follow-
ing experiments did not address the revealing of secrets per se. However,
they are relevant in this volume on secrets because they did address the
effects of revealing traumatic or distressing events, which often are kept
secret.

REVEALING TRAUMATIC VERSUS TRIVIAL EVENTS

In a seminal experiment, Pennebaker and Beall (1986) studied the

health effects of writing about private traumatic experiences as compared
with writing about trivial events in a confidential and anonymous setting.
The more specific aim of the experiment was to determine whether reveal-
ing the facts or revealing the emotions surrounding such traumas had a
greater influence on the relief associated with revelation. Undergraduates
were randomly assigned to groups in which they were asked to write for
four consecutive days about either (1) a trivial event, (2) the facts surround-
ing a personal traumatic event, (3) the emotions surrounding a personal
traumatic event, or (4) the facts and emotions surrounding a personal
70 CHAPTER 4

traumatic event. As it turned out, those participants who wrote about the
facts and emotions surrounding the trauma actually made fewer health
center visits during the 6 months after the writing experience than did
participants in the other groups. Given that one of the three trauma-
writing groups significantly differed from the group that wrote about
trivial events, this experiment provided initial support for the idea that
writing about traumas leads to health benefits. Pennebaker and Beall
(1986) reported not only that writing about both the thoughts and emo-
tions surrounding a trauma were important for health benefits but also
that the health benefits depended, at least in part, on how deep the
expressions of thoughts and emotions were.
Pennebaker and Beall's (1986) findings spawned efforts to explain
why participants who wrote about the facts and emotions surrounding
traumatic events had fewer physician visits. In one such effort, Penne-
baker et a1. (1988) sought to determine whether revealing traumas actually
could lead to improved immune functioning. Undergraduates were asked
to write for 4 consecutive days either about trivial events or about personal
traumatic events (Le., the facts and emotions surrounding the traumas)
(Pennebaker et a1., 1988). The participants' proliferative response of lym-
phocytes (white blood cells) to stimulation of two mitogens (substances
foreign to the body) were assessed both before and after the writing
period. Any increase in lymphocytes in response to these mitogens would
be a sign of improved immune functioning. As it turned out, by the end of
the writing intervention, participants who were in the trauma-writing
group as compared with those in the trivial writing group did indeed have
significantly higher proliferative responses to one of the mitogens (Penne-
baker et a1., 1988). Thus, these findings extended the findings of Penne-
baker and Beall (1986) by demonstrating that writing about traumas can
actually improve the functioning of the immune system.
In another experiment on immune functioning, medical students
were randomly assigned to write about either private traumatic events or
control topics for 4 consecutive days and then were vaccinated against
hepatitis B (Petrie et a1., 1995). As it turned out, the group who wrote about
traumatic events as compared with the control group had significantly
higher antibody levels against hepatitis B at the 4- and 6-month follow-up
periods. These findings once again demonstrated that the emotional expres-
sion of traumatic experiences can lead to improved immune functioning.
Lutgendorf, Antoni, Kumar, and Schneiderman (1994) also studied
the effects of revealing distressing events on immune functioning. In
particular, they investigated how cognitive changes and experimental
involvement during an emotional disclosure relate to changes in antibody
titers to the Epstein-Barr virus (EBV) in healthy undergraduates. In this
HEALTH BENEFITS OF REVEALING 71

case, any significant decrements in the presence of these antibody titers

would be a sign of improved immune functioning. Undergraduates were
assigned randomly either to discuss a stressful topic that they had previ-
ously discussed only minimally or to undergo assessments (including
blood tests and a mood checklist) only. As it turned out, the participants in
the disclosure condition did not differ from the assessment-only control
group in terms of EBV antibody titers. However, the participants who
disclosed more about and avoided less of the stressful topic during the
course of the 3-week period did experience antibody decrements. As such,
Lutgendorf et a1. provided correlational, although not experimental, sup-
port for the notion that revealing emotional events is associated with
improved immune functiOning.
Given that Lutgendorf and coworkers' (1994) experimental induction
did not result in improved immune functioning, researchers assessed once
again whether disclosure of emotions through writing about upsetting
events would result in changes in blood-associated immune variables
(Booth, Petrie, & Pennebaker, 1997). Using healthy adult volunteers in two
different experiments, Booth et a1. (1997) randomly assigned them to write
about either emotional issues or trivial topics for 4 consecutive days. The
participants' levels of what are called circulating lymphocytes and T-Iym-
phocyte subsets (CD4 and CD8), as well as a variety of standard hemato-
logical markers, were measured both before and after the writing interven-
tion. Immediately following the interventions in the two experiments,
there were significant differences between the emotional disclosure and
control groups in CD4, CD8, and total circulating lymphocyte numbers
(but not in CD4jCD8 ratios or any other hematological variables), sug-
gesting better immune functioning for the emotional disclosure group.
More specifically, the circulating lymphocyte numbers in the emotional
writing group remained relatively stable over the course of the experi-
ments, suggesting that the difference between the groups was due to a
fleeting elevation in postwriting blood lymphocyte levels (i.e., a sign of
poorer immunological functioning) in the control group. Moreover, the
participants in the emotional disclosure group as compared with those in
the control group reported that the writing intervention was more stress-
ful. In a nutshell, Booth et al. (1997) offered an important contribution to
the literature by showing that although participants felt distressed by their
writing about traumas, they did not experience increases in immunologi-
cal markers of stress, and in fact they showed better immune functioning
after their writing than did the control group.
Spera, Buhrfeind, and Pennebaker (1994) conducted an experiment
that went beyond assessing the internal markers of the benefits of reveal-
ing private traumas to looking at how such writing could influence adjust-
72 CHAPTER 4

ment to job loss. Their specific objective in this experiment was to deter-
mine the influence of disclosive writing on subsequent reemployment
activity and success. Sixty-three adults who had just lost their jobs were
recruited for the study and 41 of them were assigned randomly to either
experimental writing or control writing conditions. The remaining 22
participants who did not sign up for the writing phase of the study were
included as nonwriting controls. As it turned out, the participants who
were in the group that wrote about the trauma of losing their jobs as
compared to those in the control groups were more likely to find re-
employment in the months following the study. Interestingly enough,
these effects did not seem to be caused by any heightened motivation or
job-seeking efforts of those in the writing group. Specifically, the partici-
pants in the experimental group did not receive more phone calls, make
more contacts, or mail out more job-related letters than did those in the
control groups.

SUMMARY

Using the basic paradigm comparing the health effects of writing

about traumatic events versus writing about trivial ones in an anonymous
and confidential setting, researchers have discovered that writing about
traumatic events leads to fewer physician visits, improved immunologic
functioning, and even increased reemployment rates. Because these re-
sults were obtained using experimental designs, one can conclude that
there is a cause-and-effect relationship between writing about traumas
and receiving health benefits.

REVEALING TO A PSYCHOTHERAPIST
VERSUS REVEALING PRIVATELY

Given that writing alone led to health benefits in the experiments just
described, one may wonder to what extent psychotherapists are needed to
help people with their problems. Murray, Lamnin, and Carver (1989)
compared the effects of emotional expression in written essays versus in
psychotherapy on changes in emotions surrounding distressing events.
Undergraduates completed a measure of their current moods and then
were assigned randomly to written expression, psychotherapy, or written
trivial (control) conditions. Participants in the written expression condi-
tion wrote about a disturbing event and its accompanying emotional
experiences, participants in the psychotherapy condition described (orally)
a disturbing event and their emotional experiences to a therapist, and
HEALTH BENEFITS OF REVEALING 73

participants in the control group wrote descriptions of their room and

clothes closet. The participants in the psychotherapy group met with
psychotherapists who focused on reflecting feelings and content, using
empathy, clarifying meanings, and encouraging the participants to explore
the deep understandings of their specific distressing events. Participants
in all three conditions engaged in their respective experimental tasks for
30 minutes. They all completed a mood inventory again, rested for 15
minutes, and completed the mood inventory a third time. They performed
all these tasks again 2 days later. As it turned out, the participants in
written expression condition temporarily experienced increased negative
affect, and they ultimately did not change their feelings about the disturb-
ing events. In contrast, the participants in the psychotherapist condition
experienced less negative affect, and they showed more cognitive re-
appraisal (Le., made more new meanings out of the disturbing events) and
a dramatic shift to positive affect than did participants in either of the
writing groups (Murray et al., 1989). Thus, talking to a psychotherapist had
a more positive impact than writing about distressing events.
Donnelly and Murray (1991) obtained a slightly different pattern of
results in their study in which they randomly assigned undergraduates to
one of three conditions. These conditions involved writing about trau-
matic events, receiving psychotherapy about traumatic events, or writing
about trivial events (control group) over 4 consecutive days. The re-
searchers conducted content analyses of what was expressed during the
intervention phase of the study and discovered the following pattern: Both
the written expression and psychotherapy groups as compared with the
control group expressed more positive and negative emotions, and they
showed more cognitive, self-esteem, and behavior changes during the
intervention. Moreover, the participants' positive emotions, cognitive
changes, and self-esteem improvements increased over the 4 days, where-
as the negative emotions decreased. Their degree of pain and upset about
the traumatic events decreased over the days for both treatment groups.
After the experiment, both treatment groups reported feeling more posi-
tive about their topics and themselves. Despite the positive changes on
those measures, measures of mood taken before and after each session
showed that participants in the written expression group reported an
increase in negative mood immediately after each session, whereas the
psychotherapy participants did not. Based on these findings, Donnelly
and Murray (1991) suggested that the key role of the therapist may be to
provide support in facing emotional traumas.
Segal and Murray (1994) followed up that study with a test of whether
psychotherapy is more effective than an expression of thoughts and feel-
ings into a tape recorder. They recruited 60 undergraduates with un-
74 CHAPTER 4

resolved traumatic experiences and asked them either to participate in a

brief course of cognitive therapy or to talk into a tape recorder. As it turned
out, both procedures were similarly effective in reducing negative mood
and negative thoughts, although the course of cognitive therapy was some-
what more effective on two outcome measures. Specifically, at I-month
follow-up, the participants in the cognitive therapy group as compared
with those in the tape-recorder group felt significantly better about them-
selves and thought differently about their topic to an even greater extent.
In sum, Murray and colleagues conducted a series of studies with
healthy participants who either underwent very brief therapy or revealed
their distressing experiences privately. Under these conditions, therapy
was found either to have a similarly favorable outcome as or to have a
slight advantage over revealing privately. It seems that in particular one's
immediate mood after a revelation may be more positive with than with-
out a therapist. The reader must keep in mind, however, that although
participants talked about events that they rated as equally personal, the
participants in the writing group generated more negative emotion during
their expression task than did participants in the psychotherapist condi-
tion (see Murray et al., 1989). It may be that participants disclosed more
negative details during their expression period, and therefore their mood
was lower than it was for participants in the psychotherapist condition. In
essence, it may have been what they chose to reveal, rather than the
support of the therapist, that led to the group differences.

WRITING VERSUS TALKING

In Murray and his colleagues' experiments, psychotherapy was supe-

rior in terms of improving affect to writing about distressing events (Don-
nelly & Murray, 1991; Murray et al., 1989), but it was not superior to talking
alone (Segal & Murray, 1994). This pattern begs the question of whether
talking is superior to writing about distressing events. Therefore, Murray
and Segal (1994) conducted another experiment in which they compared
these two processes. They randomly assigned undergraduates either to
express vocally or to write down their feelings about either interpersonal
traumatic events or trivial events in 20-minute sessions over 4 days. As it
turned out, similar emotional processing was produced by vocal and
written expression of feeling about traumatic events. Specifically, the pain-
fulness of the topic decreased steadily over the 4 days, with both groups
reporting positive cognitive changes and feeling better about their topics
and themselves. The researchers conducted a content analysis of the ses-
sions and discovered greater overt expression of emotion and related
HEALTH BENEFITS OF REVEALING 75

changes for participants in the vocal condition than for those in the writing
condition. Moreover, there was an upsurge in negative emotion after each
session of either vocal or written expression. Murray and Segal (1994)
suggested that previous findings that psychotherapy ameliorated this
negative mood upsurge cannot be attributed to the vocal character of
psychotherapy.
Pennebaker (1997a) and Esterling, L' Abate, Murray, and Pennebaker
(1999) conducted separate literature reviews on the health benefits of
revealing private traumatic experience and concluded that expression in
either written or vocal form is healthful. In one of the key experiments
described in both reviews, healthy undergraduates who were seropositive
for the EBV (which is extremely common) were randomly assigned to
write or talk about stressful events or to write about trivial events (control
group) during three weekly 20-minute sessions (Esterling, Antoni, Fletcher,
Margulies, & Schneiderman, 1994). The participants completed a person-
ality inventory and provided blood samples before their writing or talking
period and then provided a blood sample after the final intervention
period. After the final intervention, the participants in the talking/
stressful group as compared with those in the written/ stressful group had
significantly lower EBV antibody titers, which suggests better cellular
immune control over the latent virus. In tum, the participants in the
written/ stressful group had significantly lower EBV antibody titers than
did the participants in the control group. Thus, the investigators demon-
strated that either talking or writing about distressing experiences leads to
improved immunolOgic functioning. Moreover, content analysis of the
written or oral expression indicated that the talking/stressful group
achieved the greatest improvements in cognitive change, self-esteem, and
adaptive coping strategies (Esterling et al., 1994).

REVEALING PREVIOUSLY DISCLOSED

VERSUS UNDISCLOSED EVENTS

As explained in the previous chapter, at one point, Pennebaker (e.g.,

1989) postulated that part of the benefit in revealing comes from no longer
having to expend energy in inhibiting one's expression of the traumatic
experience. Greenberg and Stone (1992) subsequently tested Pennebaker's
(e.g., 1989) inhibition model directly. Following Pennebaker's (e.g., see
1989) writing paradigm, they asked healthy undergraduates to write for 20
minutes across 4 consecutive days about undisclosed traumas, previously
disclosed traumas, or trivial events (Greenberg & Stone, 1992). As it turned
out, the three groups did not significantly differ in their subsequent use
76 CHAPTER 4

of health care services and/or in their self-reports of physical symptoms.

With the follow-up analyses, however, Greenberg and Stone were able to
show that the participants who disclosed more severe traumas as com-
pared with those who disclosed less severe traumas and control partici-
pants reported fewer physical symptoms in the months following the
study. Greenberg and Stone suggested that health benefits occur when
severe traumas are disclosed, regardless of whether previous disclosure
has occurred. However, they noted that this conclusion was based on their
correlational findings only, given that they found no between-groups
differences to suggest that writing about traumas causes these health im-
provements. They also noted that their one of their main manipulation
checks (i.e., that assessed the extent to which participants reported that
they previously had held back from revealing their traumatic events) did
not show the expected differences between the undisclosed trauma and
the previously disclosed trauma groups. This result calls into question
whether they really offered an adequate test of Pennebaker's inhibition
theory.

REVEALING REAL VERSUS IMAGINED TRAUMAS

In another clever experiment, Greenberg, Wortman, and Stone (1996)

questioned the basic assumption behind the writing studies: they chal-
lenged the idea that writing about real experiences is healthful. Specifi-
cally, they examined whether disclosing emotions generated by imagina-
tive immersion in a novel traumatic event would enhance health and
adjustment in a manner similar to that caused by disclosing real traumatic
events. College women were recruited on the basis of their having experi-
enced a traumatic event that was perceived as severe (e.g., rape, violent
assault, abandonment by a parent, witnessing a gruesome event). The
women then were assigned randomly to write about their own real
traumas, imaginary traumas, or trivial events. Those in the imaginary
trauma group were given real traumas to write about, so that the content of
their writing would be equivalent to the writing of those in the real trauma
group. As it turned out, at the end of the experiment, the imaginary trauma
participants were significantly less depressed than real trauma partici-
pants, but they were similarly angry, fearful, and happy. Interestingly
enough, at I-month follow-up, both trauma groups made significantly
fewer illness visits than did the control group participants. However, the
real trauma participants reported more fatigue and more avoidance of
certain ideas, feelings, or situations than did the other groups. Greenberg
et al. (1996) speculated that the health benefits for the participants in the
HEALTH BENEFITS OF REVEALING 77

imaginary trauma group could have resulted from their gaining catharsis,
engaging in emotional regulation, or constructing resilient possible selves
(i.e., views of self as competent and successful; see Markus & Nurius,
1986). Most relevant for the present review, these findings point to the
possibility that simply writing about traumatic events-not even neces-
sarily one's own-is healthful.

HEALTHY UNDERGRADUATES VERSUS CLINICAL SAMPLES

Bootzin (1997) and Kevin McKillop and I (Kelly & McKillop, 1996)
criticized Pennebaker's (e.g., see 1997a;b) work on the health benefits of
revealing traumatic events by noting that the experiments have relied
almost exclusively on relatively healthy individuals, begging the question,
What happens when groups from clinical populations reveal their private
traumas through writing? Do they experience health benefits?
One experiment prOVided evidence that sharing private, negative
information in therapy may actually help diseased individuals live longer
(Spiegel et al., 1989). Advanced-stage breast cancer patients were ran-
domly assigned to psychosocial support groups or to routine oncological
care groups. Those who had participated in the support groups survived
significantly longer (M = 36.6 months) than did those who received the
routine care (M = 18.9 months). The researchers argued that it was the
group members' revealing of their private fears to other group members
that helped them maximize the benefits from therapy and contributed to
their living longer. However, there was no direct empirical evidence that
it was the revealing of fears per se that led to the group differences. There
were other differences between the two groups that also could have led to
these findings. For instance, it may be that listening to others' problems
and realizing that the support group members were not alone in the
troubles led to the group differences.
Smyth, Stone, Hurewitz, and Kaell (1999) more recently addressed
this issue of the reliance on samples of healthy undergraduate by conduct-
ing a trauma-writing experiment with outpatients suffering from mild to
moderately severe asthma or rheumatoid arthritis. The outpatients were
assigned randomly to conditions in which they either wrote for 3 consecu-
tive days about "the most stressful life experience that they had ever
undergone" (p. 1305) or about their plans for the day (control group).
Amazingly, and much like the results from experiments with healthy
undergraduates, at 4-month follow-up, 47.1% of the 70 outpatients in the
experimental group as compared with only 24.3% of the 37 outpatients in
the control group experienced clinically meaningful improvement in
78 CHAPTER 4

health outcomes. In particular, the asthma patients and arthritis patients

in the experimental group experienced significantly more improvement in
lung functioning and in overall disease activity, respectively, as compared
with the asthma and arthritis patients in the control group.

META-ANALYSES
As alluded to earlier, several reviews have been conducted on the
health effects of revealing private traumatic or distressing experiences. For
example, Esterling et al. (1999) reviewed studies on the use of writing,
alone or in conjunction with traditional psychotherapy, noting that an
interest in such effects has increased substantially in recent years. They
concluded that talking with friends, confiding to a therapist, praying, and
even writing about thoughts and feelings can be physically and mentally
beneficial.
Smyth (1998) conducted a more specific review of 13 well-controlled
published and unpublished experiments on the health benefits of writing
about emotional events. He concluded that the overall effect size of the
writing on health was .42, which is considered to be between a medium
and a large effect size. He also noted that Pennebaker was involved in 8 of
the 13 experiments. Smyth compared the eight studies involving Penne-
baker with the five studies that did not involve him and found that if
anything the effects of writing were even greater in the studies not involv-
ing Pennebaker (although the difference was not statistically significant).
Thus, it seems clear that revealing private traumatic experiences in confi-
dential settings is healthful, and while Pennebaker has been the leading
researcher in this area, he is not the only researcher to have observed these
remarkable benefits of revealing secrets.

SUMMARY
Although to date there is no direct (i.e., experimental) evidence that
keeping secrets causes health problems, there is direct evidence that
through keeping a secret a person may miss out on the health benefits of
revealing it. These documented health benefits have ranged from healthy
undergraduates' improved immunologic functioning (e.g., Petrie et al.,
1995) to advanced-stage breast cancer patients' living longer (Spiegel et al.,
1989). It seems that whether talking or writing, revealing anonymously or
to a therapist, or revealing previously disclosed or undisclosed traumatic
events, revealing private and disturbing experiences is healthful. The one
glitch that emerges from drawing clear implications from this research is
HEALTH BENEFITS OF REVEALING 79

that revealing either an imagined or real trauma has been found to be

healthful (Greenberg et al., 1996). This finding does not undermine the
conclusion that writing anonymously about disturbing events is healthful,
but it does call into question what it is about revealing such events that
is healthful. This issue is addressed in the next chapter.

CONCLUSION

The implications of the results from these experiments are clear: If one
wants to feel better about a private traumatic experience as well as feel
healthier overall, one should write or talk about it either anonymously or
to a trusted person, such as a psychotherapist. There is evidence that such
expression about both the facts and feelings surrounding the traumatic
event actually can lead to improved immune functioning and fewer physi-
cian visits (e.g., Pennebaker & Beall, 1986). The findings have been ob-
tained not only by Pennebaker and colleagues but also by other re-
searchers who have observed even larger effect sizes for these health
benefits (see Smyth, 1998). Such findings may make the reader wonder
why everyone does not write about his or her traumatic experiences in
either a diary or journal. Murray and colleagues (e.g., 1989) shed some
light on this question by demonstrating that people feel negative in the
short run as they are writing and that a therapist may be able to help offset
that negativity. Greenberg et al. (1996) too showed that participants who
wrote about their own traumatic experiences as opposed to those who
wrote about imagined ones experienced more negativity and avoidance
of certain thoughts and feelings surrounding the trauma. These findings
help to explain why people do not always take advantage of such a
seemingly simple route to improved health. However, if they can tolerate
the negative feelings in the short run, the potential benefits seem worth the
trouble.
CHAPTER 5

WHAT IS IT ABOUT REVEALING

SECRETS THAT IS BENEFICIAL?

Given the substantial evidence described in the previous chapter that

revealing private traumatic or negative experiences is associated with
psychological and physiological benefits, the question remains, Why is
revealing secrets helpful? One commonly held view among scientists is
that through gaining new insights into the secrets people feel better about
them. Another idea is that through venting emotions or gaining catharsis
people come to feel better about their secrets. In this chapter, I review the
evidence on these two ideas and then describe in detail a pair of studies
that my students and I conducted (Kelly et ai., 2001) in which we teased
apart the effects of gaining new insights versus catharsis.

NEW INSIGHTS

Searching for meaning in life has been described as one of the primary
motivations for human beings (see Frankl, 1976/1959), and this search
seems particularly acute after a traumatic experience. Making sense of
such an experience is considered to be a necessary part of regaining mental
and physical health (Antonovsky, 1990; Baumeister, 1991; Lifton, 1986).
Baumeister (1991) stated that "when a major personal trauma or setback
occurs, for example, the initial response may be the rejection of meaningful
thought, as in denial. Then, gradually, the person copes with the crisis by
finding a new way to interpret what went wrong and to put the world back
together" (p. 75). If people do not talk or write about their traumatic
experiences and choose instead to keep them secret, they may miss pre-
81
82 CHAPTER 5

cious opportunities to get new perspectives on the secrets and develop a

sense of closure on the events (Kelly & McKillop, 1996; Pennebaker, 1989,
1997b; Pennebaker & Hoover, 1985; Tait & Silver, 1989). People often find
meaning in the experiences by getting a new perspective on them and then
assimilating them into their worldviews (Horowitz, 1986; Meichenbaum,
1977; Pennebaker et al., 1988; Silver et al., 1983). In fact, many counseling
approaches have grown out of the idea that reinterpreting or reframing
experiences is critical in helping clients get better (e.g., see Dowd & Milne,
1986; Kelly, 1955). Across many types of brief therapy interventions, inter-
pretations made by the therapist have been found to be associated with
client improvement (see Hill, 1992), apparently because the interpretations
provide clients with new perspectives on their problems.
When people are not able to incorporate their personal traumas into
their lives, they may find ways to escape themselves through such self-
destructive means as suicide, masochism, alcoholism, bulimia, or exces-
sive religiosity (see Baumeister, 1990, 1991). A "tunnel vision" thought
process seems to precede suicide, such that suicidal people become rigid in
their thinking and will no longer try to make meaning out of the events in
their lives or gain insight from experiences (Shneidman & Farberow, 1961,
1970). In one study, suicidal people as compared with nonsuicidal people
were significantly less able to come up with solutions to interpersonal
problems in their lives (Shneidman & Farberow, 1970). Moreover, in Silver
and co-workers' (1983) study of survivors of father-daughter incest, the
women who had not made meaning of their incest experiences as com-
pared with the women who had done so reported more psychological
distress, poorer social adjustment, and lower levels of self-esteem.
Foa, Molnar, and Cashman (1995) observed the importance of mean-
ing making in women who had experienced sexual assault and then had
undergone therapy that involved their repeatedly reliving and recounting
their trauma in a treatment referred to as exposure therapy, which is used for
treating anxiety disorders such as phobias and posttraumatic stress dis-
order (PTSD) (Foa & Kozak, 1986; Foa & Rothbaum, 1989). Following the
treatment, the women had more thoughts in which they attempted to
organize the trauma memories, and these organized thoughts were nega-
tively correlated with depression.
Along these same lines, after reviewing the studies on the physical
and mental health benefits of revealing private traumatic experience, Pen-
nebaker (1997a) concluded that gaining new insights is an important part
of recovery from distressing experiences. Pennebaker (1997a) observed
that although a reduction in inhibition may contribute to the disclosure
phenomenon, changes in basic cognitive and linguistic processes during
writing predict better health. The key, he noted, was the level of insight
WHAT IS IT THAT IS BENEFICIAL? 83

people used during their writing. In particular, he found that a high use of
positive emotion words and moderate levels of negative emotion words
were associated with positive health outcomes.
Suedfeld and Pennebaker (1997) observed this pattern when they
explored whether the recall of very unpleasant memories would occur at a
different level of complexity in meanings from that of neutral memories
and whether differences in such complexity would be related to health
outcomes. Two groups of undergraduates wrote an essay each day for 4
days: one group wrote about a trivial topic and the other wrote about a
negative (traumatic) life event for a previous study. The complexity scores
of these two types of essays were compared and they were correlated with
a composite measure of well-being (immunologic assays, visits to the
Student Health Center, and self-reported distress and substance abuse). As
it turned out, the essays about negative experiences were significantly
higher in complexity, suggesting that the participants put more mental
effort to their writing. Among these essays, there was a significant relation-
ship between the complexity of meanings in the writing and improvement
in the participants' health, such that moderate levels of complexity (Le.,
scores closest to the median for the group) were associated with the most
improvement. Apparently, very high levels of complexity represent feeling
quite troubled and may be an indication of a continued lack of resolution
surrounding the negative event, and low levels of complexity may mean
that the person did not put in the mental energy needed to gain closure on
the negative event (see Pennebaker, 1997a).
In a separate analysis of a series of writing about trauma studies,
Pennebaker, Mayne, and Francis (1997, Study 1) observed a mixed pattern
regarding the links between gaining new insights into the trauma and
experiencing changes in physical versus mental health. When participants
increased their use of words associated with insightful and causal thinking
over the course of their writing, they experienced improved physical
health. However, their increased use of insightful words was not associ-
ated with improved mental health (Pennebaker et aI., 1997).
Nolen-Hoeksema, McBride, and Larson (1997) also observed mixed
results regarding attempts to make sense of a trauma. They found that
among recently bereaved men, those who analyzed the meaning of their
loss as compared with those who engaged in less analysis reported greater
positive morale 1 month after the loss. However, these men experienced
more persistent depression and lack of positive states of mind over the 12
months after the loss (Nolen-Hoeksema et al., 1997). As the researchers
themselves noted, because of the correlational nature of the study, it was
not clear whether the analysis caused the depression, or whether the
particularly troubled men simply tended to engage in more analysis.
84 CHAPTERS

Lepore (1997) acknowledged the importance of making meaning out

of distressing experiences and tried to go one step further in understand-
ing how it reduces stress surrounding the events. In particular, he explored
whether expressive writing improves emotional adaptation to distressing
events by reducing event-related intrusive thoughts-unwanted thoughts
that pop into people's heads-or by desensitizing people to such thoughts.
He asked undergraduates who were preparing for their graduate school
entrance exams either to write their deepest thoughts and feelings about
the exam (i.e., experimental group) or to write about trivial events (i.e.,
control group). Participants in the experimental group experienced a sig-
nificant decline in depressive symptoms from 1 month (Time 1) to 3 days
(TIme 2) before the exam. In contrast, the participants in the control group
maintained a relatively high level of depressive symptoms over the same
period. Expressive writing did not affect the frequency of intrusive
thoughts, but it did seem to affect the impact of intrusive thoughts on
depressive symptoms. Specifically, intrusive thoughts at Time 1 were pos-
itively related to depressive symptoms at Time 2 in the control group and
were unrelated to symptoms in the expressive writing group. Lepore
speculated that one possible explanation for this finding was that "people
who engage in thinking about and expressing their stress-related thoughts
and feelings may gain some insight into the stressor, which in tum renders
any reminders or memories of the stressor comprehensible and non-
threatening" (p. 1034). However, he acknowledged that it is not possible to
know whether that explanation is better than the notion that the partici-
pants habituated to (i.e., got used to) the stressors when they wrote or
talked about them (d. Bootzin, 1997).
Bootzin (1997) criticized the research on the effects of writing about
traumas in general by pointing out that efforts to explain these health
benefits have been correlational in nature. As such, researchers have not
been able to establish cause-and-effect relationships between the proposed
mechanisms that underlie the benefits of writing and the health outcomes.
For example, one cannot be certain from studies such as Silver and co-
workers' (1983) study of survivors of incest whether meaning making
caused the psychological benefits observed or whether a third variable,
such as participants' intelligence levels, could explain these benefits (see
Kelly et al., 2001). The participants with higher intelligence as compared to
those with lower intelligence simply may have been better at developing
new insights and coping with their problems.
In sum, it has been observed in a number of writing studies that there
is a link between making meaning of traumatic events and improved
health. However, given the correlational nature of these analyses, more
research is needed to try to establish whether a cause-and-effect relation
WHAT IS IT THAT IS BENEFICIAL? 85

exists between meaning making and subsequent mental states (see Boot-
zin, 1997; Kelly et al., 2001).

CATHARSIS

As mentioned earlier, another popular explanation for why revealing

secrets leads to health benefits is that revealers have the opportunity to
gain catharsis. Breuer and Freud (1975) were early proponents of catharsis
and stated that "the patient only gets free from the hysterical symptoms by
reproducing the pathogenic impressions that caused it and by giving
utterances to them with an expression of affect, and thus the therapeutic
task consists solely in inducing him to do so" (p. 283). Consistent with his
interpretation of Freud's cathartic method, Pennebaker (1997b) defined
catharsis as the linking of emotion and insight. However, researchers and
therapists in general have come to use the term catharsis to mean a venting
of pent-up emotions or an expression of emotions behaViorally (e.g., Bush-
man, Baumeister, & Strack, 1999; Polivy, 1998; Tice & Ciarocco, 1998). In his
catharsis theory, for example, Scheff (1979) contended that recalling the
facts of an emotional experience is unnecessary, whereas discharging
one's emotions is both necessary and sufficient for therapy. Catharsis will
be used here to mean a venting of emotions.
Many psychotherapists today continue to believe that catharsis is
beneficial, as illustrated by the fact that it has been an integral part of a
number of the more recent approaches to psychotherapy, such as reevalua-
tion therapy ijackins, 1962), primal therapy ijanov, 1970), new identity
therapy (Casriel, 1972), psychodrama (Moreno, 1958), and short-term dy-
namic psychotherapy (Davanloo, 1980; Nichols & Efran, 1985; Sifneos,
1979). Laypersons, too, often consider catharsis to be helpful. For example,
in a survey I (Kelly, 1997) conducted a few years ago, undergraduates rated
gaining catharsis as a highly effective strategy for dealing with their own
troubling secrets and perceived that other people would view gaining
catharsis as significantly more beneficial than gaining new insights into
troubling secrets.
In her review of the research on catharsis, Polivy (1998) concluded that
even though some studies have raised the question of whether catharsis
actually purges or provokes emotions, catharsis seems to reduce one's
level of emotional arousal surrounding a troubling event more often than
not. Two studies with Holocaust survivors provided indirect empirical
support for the idea that catharsis is beneficial (Pennebaker, Barger, &
Tiebout, 1989; Shortt & Pennebaker, 1992). As the survivors related their
previously suppressed stories to the researchers, their skin conductance
86 CHAPTERS

levels fell, indicating that they became more relaxed (Pennebaker et al.,
1989; Shortt & Pennebaker, 1992). Similarly, informal studies of suspected
criminals who were given lie detector tests demonstrated that when the
suspects confessed, their skin conductance levels dropped, and they ap-
peared more relaxed to observers (see Pennebaker, 1985, 1990). Moreover,
student outpatients at a university clinic who received cathartic psycho-
therapy, as compared with those who received insight-oriented psycho-
therapy, reported more improvement on a measure of life satisfaction
(Nichols, 1974). However, in that same experiment, the outpatients who
received insight-oriented therapy reported a greater reduction in symp-
tomatology.
Mendolia and Kleck (1993) conducted a pair of experiments that
suggested that there may be delayed benefits to catharsis. They examined
the effects of talking to another person about either the facts or emotions
surrounding a stressful event on the physiological arousal of the discloser.
In Experiment 1, undergraduates viewed an affect-neutral videotape fol-
lowed by a stress-inducing videotape (i.e., of gruesome woodshop acci-
dents). The participants then talked about either their emotional reactions
to the stressful videotape (emotion condition), the sequence of events
within it (fact condition), or the sequence of events within the neutral
videotape (distraction condition). All the participants were then reexposed
to the stressful accident episode. As it turned out, the participants in the
emotion condition as compared with those in the fact condition were more
autonomically aroused (i.e., distressed) during the second exposure to the
accident episode, demonstrating that expressing negative emotions had an
untoward immediate effect. In Experiment 2, 48 hours separated partici-
pants' talking about their first exposure to the stressful stimulus from their
second exposure to it. This time, participants in the emotion condition as
compared with those in the fact condition had lower levels of autonomic
arousal while viewing the stimulus again and reported more positive
affect after watching it. In essence, although gaining catharsis was not
helpful over the short run, it was beneficial over the longer duration
perhaps because it allowed participants to get used to their negative
emotions surrounding the distressing videotape (Mendolia & Kleck, 1993).
However, other researchers have speculated that the scientific com-
munity has abandoned catharsis theory because of the research that has
failed to support the benefits of venting one's emotions (Bushman, Bau-
meister, & Stack, 1999). For example, in one experiment, participants who
expressed a fear of public speaking were asked to make speeches before an
audience (Tesser, Leone, & Clary, 1978). They were randomly assigned to a
condition that required them to focus on why they had those feelings, a
catharsis-only condition, or a control group. It was found that participants
WHAT IS IT THAT IS BENEFICIAL? 87

exhibited the highest levels of digital perspiration in the catharsis condi-

tion and reported feeling significantly worse than did the participants in
the condition that focused on why they had those feelings (Tesser et al.,
1978). In another experiment, participants were assigned to take part in
one of three conditions: (1) interpersonal, which involved a 20-minute
counseling session, (2) silence, which allowed for 20 minutes to take time
to review feelings in silence, and (3) catharsis, which provided a tape
recorder so the participant could "get things off (his/her) chest" (Bohart,
Allen, Jackson, & Freyer, 1976). Participants in the interpersonal group
reported the greatest improvement in affect, whereas participants in the
catharsis group showed the least improvement (Bohart et al., 1976). In yet
another study, men in a daily stress-coping skills group who were told
to use a cathartic release strategy reported increases in negative affect
relative to their original levels (Stone, Kennedy-Moore, & Neale, 1995).
In a study by Siegman and Snow (1997), each of 24 participants took
part in three conditions: (I) anger-out, in which previously experienced
anger-arousing events were described loudly and quickly, (2) anger-in, in
which anger-arousing events were relived inwardly, in the participant's
imagination, and (3) mood-incongruent speech, in which anger-arousing
events were described softly and slowly. As it turned out, only the anger-
out condition was associated with high cardiovascular reactivity levels,
and self-reported anger was highest in the anger-out condition, moderate
in the anger-in condition, and lowest in the mood-incongruent condition.
Based on these findings, Siegman and Snow concluded that the full-blown
expression of anger is problematic and that the mere inner experience of
anger is not.
Similarly, after comparing the effects of cognitive therapy with talking
into tape recorder (described in the previous chapter), Segal and Murray
(1994) concluded that even though the talking into a tape recorder was
almost as effective as cognitive therapy for dealing with traumatic experi-
ences, the mechanisms through which the two groups benefitted seemed
to differ. Across both groups, the arousal of negative affect was inversely
related to positive outcome, whereas the reduction of negative affect and
negative thoughts was positively related to outcome. Participants in the
tape-recorder condition talked about negative emotional content persis-
tently over the 4-day period, whereas participants in the cognitive therapy
group were encouraged by their therapists to switch their focus to examin-
ing their negative thoughts and cognitive distortions. The researchers
concluded that "the sheer arousal of negative affect by itself does not seem
to be therapeutically valuable as might be expected from simple catharsis
theories" (p. 204).
In SUfi, whereas Polivy (1998) suggested that catharsis is associated
88 CHAPTERS

with benefits more often than not, Bushman et a1. (1999) argued that the
scientific community has all but abandoned catharsis theory. In a number
of experiments involving the venting of emotions, such venting has led to
increased negative affect relative to earlier emotional states (e.g., Stone et
a1., 1995). However, Mendolia and Kleck (1993) found that the negative
emotional states were only temporary and were followed by improved
affect, suggesting that any future study of catharsis must include both
short-term and long-term assessments of changes in affect.

NEW INSIGHTS VERSUS CATHARSIS

Although I have presented gaining new insights (Le., meaning mak-

ing) and catharsis as opposing explanations for the benefits of revealing
secrets, they often have been depicted as processes that combine to lead to
the physical and mental health benefits associated with the revealing of
private traumatic experiences (e.g., Bohart, 1980; Greenberg et a1., 1996;
Murray et a1., 1989; Pennebaker, 1989, 1990; Pennebaker, Colder, & Sharp,
1990). For instance, Berry and Pennebaker (1993) suggested that the effec-
tiveness of many common expressive therapies (e.g., art, music, cathartic)
would most likely be enhanced if clients were encouraged both to express
their feelings nonverbally and to put their experiences into words.
Similarly, Pennebaker et a1. (1990) proposed that emotions were a part
of the experiences that must be recognized and that the expression of
emotions would facilitate new insight or meaning into the experiences.
They asked college freshmen to write about either coming to college or
superficial events. Following the writing, only 10% of the participants
mentioned the value of venting emotions or gaining catharsis, whereas
76% of them said that they felt the study was valuable because it helped
them to understand better their own thoughts, behaviors, and moods.
Moreover, as described in the previous chapter, Pennebaker and Beall
(1986) found that participants who wrote about the facts and emotions as
compared with participants who wrote about only the facts or emotions
surrounding their traumatic experiences made fewer health center visits
during the 6 months after the writing experience. The researchers reported
that the health benefits depended at least in part on how deep the expres-
sions of thoughts and emotions were.
One limitation of those studies is that they could not tease apart the
effects of gaining new insights and catharsis because they did not involve
a direct experimental manipulation of these variables. All of the published
studies on the health benefits of revealing have utilized correlational
analyses to establish a link between gaining new insights and benefits (see
WHAT IS IT THAT IS BENEFICIAL? 89

Bootzin, 1997). Thus, my students and I (Kelly et al., 2001) conducted a pair
of studies designed to disentangle the effects of gaining catharsis from the
effects of gaining new insights on improving one's affect surrounding a
secret. We predicted that participants' trying to gain new insights into their
secrets would be more helpful than trying to gain catharsis alone, because
purely venting their emotions surrounding their secrets could actually
intensify the emotions. Furthermore, we expected that the more the partic-
ipants developed new insights into their secrets, the more relief they
would experience.
In Study 1, we assessed whether participants' having gained catharsis
or new insights into their secrets from revealing them to confidants in the
past was associated with recovery from the secrets. In Study 2, we fol-
lowed up these correlational findings by attempting to establish a cause-
and-effect relationship between participants' gaining new insights into
their secrets and experiencing improved affect surrounding the secrets.

STUDY 1

One hundred thirty-seven undergraduates were asked to (1) select the

most private, personal secret that they had ever shared with another
person, (2) describe to what extent they experienced catharsis and gained
new insights into the secret after revealing it, and (3) indicate how they felt
when thinking about the secret now. We hypothesized that having gained
new insights into the secret would be the better predictor of feeling posi-
tive about the secret at the time of the study.

PROCEDURE

The participants were given questionnaire during their regular class

time and were asked not to write their names anywhere on the question-
naires. They read the following instructions:
Virtually everyone keeps secrets, or hides personal information from others
at some point in time. In other words, we hold private information that we
would want very few other people (or no one) to know about. Please take a
moment to reflect on a time in which you shared a very private and personal
secret with someone else. Select a secret from the past or present that in-
volves you directly and personally. Select the most private, personal secret
that you have ever shared with another person, but do not write it down.
The participants then completed scales assessing the extent to which they
had gained new insights and catharsis from having revealed their secret
to the confidant. Specifically, a modified version of the Therapy Session
90 CHAPTER 5

Report (TSR) (Orlinsky & Howard, 1966) was used to measure the extent to
which participants felt that they had gained catharsis and new insights
from having disclosed their secrets to their confidants in the past. We
adjusted the questionnaire to correspond with our research goals by pref-
acing the items with, "What do you feel you got out of sharing your secret
with this person?" (Le., their confidant). The subscales were what Howard,
Orlinsky, and Hill (1970) called catharsis (e.g., "a chance to let go and get
things off my chest"; "relief from tensions or unpleasant feelings") and
mastery /insight (e.g., "more understanding of the reasons behind my
behavior and feelings"; "ideas for better ways of dealing with people and
problems"). We also assessed participants' current positive and negative
feelings about their secret. In particular, participants were asked to reflect
on the secret they thought of earlier in the session, to feel the emotions
associated with this secret, and to "indicate to what extent you feel this
way whenever you think about your secret." The positive affect measure
included items such as "interested," "alert," and "inspired"; and the
negative affect measure included items such as "irritable," "distracted,"
and "ashamed" (see Watson, Clark, & Tellegen, 1988).

RESULTS AND DISCUSSION

Study 1 assessed whether gaining catharsis or new inSights from

revealing secrets to confidants in the past was associated with improved
affect surrounding the secrets at the time of the study. As it turned out,
gaining new insights (as measured by the mastery/insight sub scale of the
TSR) was correlated significantly with positive feelings surrounding the
secrets, whereas gaining catharsis (as measured by the catharsis subscale
of the TSR) actually was correlated with negative feelings surrounding the
secrets.
The finding that gaining new insights was associated with positive
affect surrounding the secret is consistent with previous research that
showed that across therapeutic techniques interpretations provided by the
therapist were associated with positive client outcomes (Hill, 1992). Such
interpretations are likely to provide people with new perspectives on their
problems or difficulties and help them make sense of their troubles.
In addition, the finding that catharsis was associated with negative
feelings surrounding the secrets is consistent with results from a classic
study by Ebbesen, Duncan, and Konecni (1975). These authors interviewed
100 technicians and engineers who had just been laid off from their jobs at
an aerospace company and asked some of them anger-eliciting questions
like, "What instances can you think of where the company has not been
fair with you?" (p. 446). When these same individuals were later asked to
rate their attitudes toward the company, they were more hostile toward the
WHAT IS IT THAT IS BENEFICIAL? 91

company than were those who had not been asked the anger-eliciting
questions initially. Thus, getting one's feeling out in the open may intensify
the negative emotions one feels.
A qualifier to this conclusion is that it is possible that participants told
their confidants more negative secrets in cases when they needed (and got)
catharsis as compared with when they needed new insights. Because these
individuals may have started out with more negative secrets, they still
may have had more negative feelings about the secrets at the time of the
study, even if experiencing catharsis had been beneficial to them. In es-
sence, an important drawback to interpreting these findings is that the
data are correlational, thus making it impossible to determine whether
gaining catharsis and new insights actually caused participants to feel
worse or better about their secrets. Moreover, participants' reports of
whether they gained catharsis or new insights from revealing their secrets
to their confidants were retrospective, and it is unclear whether their
recollections were accurate. As such, Study 2 was conducted to address
both of these issues. In it, we utilized an experimental design to explore
whether gaining catharsis or new insights into one's secrets would lead to
differences in affect surrounding those secrets.

STUDY 2

Eighty-five undergraduates were brought into the laboratory and

randomly assigned to one of three groups in which they were asked to
write about either (1) their secrets while trying to gain new insights into
them, (2) their secrets while trying to gain catharsis, or (3) their previous
day (i.e., control group). They were asked to generate the secrets and rate
their feelings about them before the manipulation was introduced. After
two sessions of identical writing tasks spanning one week, participants
were again asked to rate their feelings surrounding the secrets. Given the
findings from Study I, we predicted that the new-insights group would
show significantly greater improvements in affect surrounding their se-
crets than would either the catharsis or control groups, and that those
participants who reported making the most meaning out of their secrets
would show the greatest improvements. Likewise, we predicted that par-
ticipants who were most able to come to terms with their secrets during
their writing would experience the greatest improvements in affect. This
prediction was based on previous theoretical work suggesting that gaining
closure is a central part of coping with private traumas (see Pennebaker,
1997b). We also expected that the catharsis group would fare no better than
the control group in terms of changes in affect.
It is important to note that we were interested in studying the effects
92 CHAPTER 5

of the process of gaining insights, as opposed to what the new insights

were. Our basis for this focus was that Silver et al. (1983) found
no evidence to suggest that the specific type of answer our respondents
generated from their search for meaning was important in terms of coping
effectiveness. Rather ... what appeared to be critical was whether the women
were able to make any sense of their incest experience at all. (p. 91, italics in
original)

Moreover, we studied the effects of gaining new insights and catharsis

both immediately and after 1 week to address the possibility that there
could be some delayed benefits to catharsis (see Mendolia & Kleck, 1993).

PROCEDURE

Participants were brought into the laboratory in groups of one to three

and seated in separate cubicles. The experimenter told them that their
responses would be confidential and anonymous and that even the experi-
menter her- or himself would not have access to the locked box where they
would leave their completed packets.
After the opening instructions, participants were asked to report on
their stream of consciousness for a 3-minute writing period. The purpose
of this task was to acclimate them to writing about their personal thoughts
and feelings. Then they were asked to generate a very private and personal
secret that they had told no one or very few people. They were asked to
describe one secret only and not to go into any detail in their description.
After they listed their secret and turned the page over, they were asked to
complete several 9-point items. These items were: "How disturbing is
this secret to you?" (from not at all = 1 to extremely = 9); "How private is
this secret to you?" (from not at all =1 to extremely = 9); "How often have
you thought about this secret during the last week?" (from never = 1 to
extremely often = 9); and "To what extent have you come to terms with this
secret?" (from not at all = 1 to full = 9). They then completed for the first
time the same measure of their positive and negative affect that was used
in Study 1. As in Study 1, we asked the participants to reflect on the secret
they thought of earlier in the session, to feel the emotions associated with
this secret, and to "indicate to what extent you feel this way as you are
th;nking about your secret now."
Next, the manipulation was introduced. Participants were randomly
assigned to spend 25 minutes writing according to one of the following
sets of instructions: (1) The new-insights group was instructed to "focus on
making sense out of the secret or gaining new insights into the secret.
Develop this new perspective on your secret by changing your thoughts
WHAT IS IT THAT IS BENEFICIAL? 93

about it. Your sole purpose in writing is to make meaning out of your
secret-to gain a new perspective or new understanding of the secret."
(2) The catharsis group was instructed to "focus on what you are feeling
about the secret and getting those feelings out in the open. Write about
your feelings without rationalizing or explaining them. Your sole purpose
in writing is to get your feelings about the secret off your chest-to really
pour out your emotions and release them." (3) The control group was
asked to describe "in detail what you did yesterday from the time you
woke up till the time you went to bed. It is important that you describe
things exactly as they occurred. Do not mention your own emotions,
feelings, or opinions. Your description should be as objective as possible."
After they were done with the writing task, all three groups com-
pleted the measure of their positive and negative affect surrounding their
secrets a second time and items that asked how many people they had told
their secret before the study, along with some distracter items. They were
thanked for completing the first portion of the study and reminded about
their session scheduled for the next week. To bolster the manipulation, the
new-insights group was told, "if you wish to discuss your secret and
continue to make meaning out of it with other people whom you trust,
please feel free to do so"; the catharsis group was told, "if you wish to
discuss your secret and continue to get your feelings about it off your chest
with other people whom you trust, please feel free to do so."
The participants returned exactly 1 week later. The experimenter re-
minded them of the anonymity and confidentiality of their responses and
gave them the measure of the positive and negative affect surrounding
their secrets a third time. After they completed this inventory, they were
given the identical 25-minute writing instructions that they had received
the previous week (i.e., the new-insights group received the instructions
to try to gain new insights, the catharsis group received the instructions to
vent their emotions, and the control group received the instructions to
write about their previous day). After this writing period, they completed
the measure of their positive and negative affect surrounding their secrets
a fourth time and were asked the following items (along with some dis-
tracter items): "To what extent did you get your feelings off your chest in
writing today"; and "To what extent did you make sense or meaning out of
your secret in writing today?" They rated these items on 9-point scales
from not at all (1) to to a great extent (9).

RESULTS AND DISCUSSION

In Study 2, we sought to determine whether there was a cause-and

effect relationship between being instructed to focus on gaining new in-
94 CHAPTER 5

sights into personal secrets and feeling more positive about those secrets.
The participants wrote down very personal secrets, some of which were
highly disturbing, including secrets about being raped, experimenting
with sexual acts, having attempted suicide, and cheating on a test (see
Chapter 1 for a complete list). There were no significant differences across
the three groups for participants' ratings of how disturbing or private the
secrets were, and none for how often they thought about the secrets and
how much they had come to terms with the secrets. There also were no
differences across the groups for how many people they had told their
secrets before the study. Participants indicated that they had told an aver-
age of between two and three people their secrets and that they had
actively held back from telling others to a fair extent.
Participants in the new-insights group experienced significantly
greater improvements in positive affect surrounding their secrets than did
participants in the catharsis and control groups. Moreover, the extent to
which participants gained new insights into their secrets from their writ-
ing, as rated by judges, was significantly correlated with participants'
reports of increased positive affect surrounding their secrets. When partic-
ipants' own ratings of the extent to which they gained catharsis and made
meaning of their secrets were entered in a multiple-regression analysis
together with initial positive affect scores, meaning making was a signifi-
cant predictor of increased positive affect, whereas catharsis was not.
Regarding negative affect, there were no significant differences across the
three groups. However, we found that participants' coming to terms with
their secrets during their writing (as rated by judges) was associated with
less negative affect. These findings provide strong support for the conclu-
sion that, not only are people able to create new insights into their secrets
on their own, hut engaging in this process is an effective coping strategy
that is superior to merely venting their emotions about the secrets.
To assess the pattern of improvement in the affect scores for the new-
insights group as compared with the catharsis group, a repeated-measures
analysis was performed on the four administrations of the affect measures
(i.e., Time) for the new-insights and catharsis conditions using the negative
affect scores subtracted from the positive affect scores at each of the four
administrations. As shown in Table 1, the participants in the new-insights
group had significantly greater improvement in affect across time than did
those in the catharsis group. The difference in mean affect scores between
the new-insights and catharsis groups was significantly greater at Time 4
than at Time 1 (P = 0.4) or at Time 3 (P = .007), but not significantly greater
than at Time 2 (P = .13). This finding is important, because Pennebaker
(e.g., 1997a) established that improvement in affect over the course of
writing about emotional events was associated with health benefits. This
WHAT IS IT THAT IS BENEFICIAL? 95

TABLE 1. Means and Standard Deviations of Negative

Affect Scores Subtracted from Positive Affect Scores
across the Four Administrations
Administration of affect scales
Condition Time 1 Time 2 Time 3 Time 4

New insights (n = 35)

M -5.19 .59 1.76 4.52
SO 13.94 12.10 10.49 11.24
Catharsis (n = 33)
M -5.55 -2.42 -.58 -1.64
SO 11.95 13.24 8.99 9.22
Control (n = 17)
M -6.18 -2.65 -1.76 -1.35
SO 13.58 9.88 9.02 8.09

pattern also is important because it illustrates that there was no delayed

benefit to catharsis relative to the other groups (see Mendolia & Kleck,
1993).
One might argue that the results were simply an artifact of the differ-
ential demand characteristics inherent in the writing tasks. Specifically,
participants in the new-insights condition may have believed that this task
was supposed to help them, whereas participants in the catharsis condi-
tion may have been skeptical about the effectiveness of catharsis. There-
fore, participants in the new-insights group may have reported that they
felt more positive about their secrets than did participants in the catharsis
group. However, a recent survey showed that a sample of 99 undergradu-
ates rated gaining catharsis and new insights into a secret as equally highly
effective for helping them cope with troubling secrets (P > .05) (Kelly,
1997). The means for new insights and catharsis were 7.44 (SD = 1.20) and
7.19 (SD = 1.31), respectively, on 9-point scales from not at all helpful (1) to
extremely helpful (9). When those same participants were asked to rate
how effective other people thought the two coping strategies were, the
participants reported that others would view catharsis as significantly
more helpful than gaining new insights (Kelly, 1997). These results directly
contradict the demand-characteristic explanation for the findings.
One reason why gaining new insights is likely to be curative may be
that people are able to find closure on the secrets and avoid what has been
termed the Zeigamik effect (Zeigamik, 1927), wherein people actively seek
to attain a goal when they have failed to attain the goal or failed to
disengage from it (Martin & Tesser, 1993; see also, Pennebaker, 1997a). As
96 CHAPTER 5

mentioned in Chapter 1, Zeigamik (1927) showed that people continue to

think about and remember interrupted tasks more than finished ones,
suggesting that they may have a need for completion or resolution of the
events. Thus, revealing a secret with the explicit intention of gaining a new
perspective on it may help people feel a sense of resolution about the
secret. Evidence for this idea comes from two separate analyses of the data
from Study 2. In one analysis, as expected, judges' ratings of the extent to
which participants gained new insights into their secrets during their
writing significantly predicted their final positive affect scores, even after
we statistically adjusted for participants' initial positive affect scores. In
another analysis, we found that not coming to terms with their secrets was
a strong predictor of feeling negative about the secrets, even after we
statistically adjusted for initial negative affect scores.

DISCUSSION

We conducted these studies in an effort to disentangle the effects of

gaining catharsis from the effects of gaining new insights, which have been
blended in previous research that has examined the benefits of revealing
emotions and facts surrounding private traumas (e.g., Pennebaker & Beall,
1986; Pennebaker et al., 1990). The experimental (and correlational) find-
ings from Study 2 and the correlational findings from Study 1 converged to
support the conclusion that focusing on gaining new insights into one's
secrets is a useful strategy for increasing positive affect surrounding the
secrets, whereas merely attempting to gain catharsis is not helpful. Study
1 even showed that gaining catharsis was associated with negative affect
surrounding the secrets.
The implications of these results are that people should try to gain
new insights into and come to terms with their troubling secrets, if their
goal is to make themselves feel better about the secrets. However, one
might argue that some secrets are difficult to construe in new ways and
that it may take a long time to create new meaning for particularly excru-
ciating or humiliating secrets. In his theory of personality, George Kelly
(1955) proposed that people are not just capable of actively constructing
their realities and seeing their problems from alternative perspectives, but
they are naturally inclined to seek these meanings. Silver et al. (1983)
showed that the great majority of their sample of survivors of long-term
father-daughter incest sought to make meaning of the experiences, and a
number of them were able to do so, despite the terrible nature of their
experiences. Other researchers also have shown that many people do
come to make meaning out of their negative life experiences (see Andrea-
WHAT IS IT THAT IS BENEFICIAL? 97

sen & Norris, 1972; Chodoff, Friedman, & Hamburg, 1964; Cornwell, Nur-
combe, & Stevens, 1977; Doka & Schwartz, 1978; Helmrath & Steinitz, 1978;
Silver & Wortman, 1980). In Study 2, we demonstrated that participants
were able to generate new perspectives on their secrets and that gaining
these new insights occurred relatively quickly (after two 25-minute writ-
ing periods spanning 1 week). This effect occurred even though many
participants described very disturbing and private secrets.
Whereas the results from Studies 1 and 2 clearly support the benefits
of gaining new insights into secrets, Nolen-Hoeksema and her colleagues
(Lyubomirsky & Nolen-Hoeksema, 1993, 1995; Nolen-Hoeksema & Mor-
row, 1993) have shown that if people are experiencing negative life events
and choose to think about their feelings surrounding the events and the
implications of those events (i.e., engage in rumination), they are more
likely to become depressed and stay depressed longer than if they choose
to distract themselves from the negative events. This observation seems to
contradict the notion that trying to make meaning of negative life events is
a buffer against suicide and other self-destructive acts (Baumeister, 1990,
1991). One way of resolving this apparent contradiction is that perhaps
only successful meaning makers feel better about their troubling secrets.
Silver et al. (1983) stated that
the extent that the search for meaning results in finding meaning in an
undesirable event, it is likely to be an adaptive process.... However, finding
meaning does not appear to terminate the search or the ruminations. More-
over, when after an extended period the search fails to bring understanding,
the continuing process of searching and repeatedly ruminating appears to be
maladaptive. (p. 81)

Another way of resolving the contradiction is to differentiate between two

kinds of meaning making: Making sense of an event and finding benefit in
the experience (Davis et al., 1998). In a longitudinal study of people coping
with the loss of a relative, Davis et al. (1998) found that making sense of
the loss was associated with less distress, but only in the first year after the
loss, whereas the finding of benefit was most strongly associated with
adjustment 13 and 18 months after the loss, a pattern that suggests that the
finding of benefit is the more long-lasting element to emotional recovery.
Perhaps researchers could utilize an experimental design in the future to
compare the effects of passive rumination with both of these types of
active meaning making.
Before closing this chapter, I must mention a couple of noteworthy
limitations to our research. First, both the new-insights and catharsis
groups, not just the neW-inSights group, in Study 2 used significantly more
emotion words (relative to total words) during their writing than did the
98 CHAPTERS

control group. Also, although the catharsis group had a statistically signifi-
cantly higher proportion of emotion words (M = .03) than did the new-
insights group (M = .02), this difference was quite small. Thus, it is not
clear whether the new-insights group actually engaged in a combination
of gaining catharsis and new insights or in gaining new insights alone.
Future researchers perhaps could prevent (more effectively than we did)
the participants in the new-insights group from writing about emotions
and then assess the effects of such a clear-cut manipulation on their affect.
Second, although the pattern of findings between Studies 1 and 2
converged concerning the effects of gaining new insights on positive
affect, the pattern was somewhat different for the two studies concerning
the effects of catharsis on negative affect. This difference could have been
the result of the fact that we studied the oral revelation of secrets to
confidants in Study 1, as compared to the written revelation of secrets
anonymously in Study 2. Specifically, the venting of emotions may have
been linked to negative affect in Study 1 because confidants are sometimes
rejecting of people who engage in such emotional expressions (e.g., see
Davidowitz & Myrick, 1984). In contrast, the venting of emotions may not
have been linked to negative affect in Study 2 because anonymous venting
has fewer negative repercussions. Because of the different methodologies
of the two studies, we cannot be sure why we obtained the different
patterns for catharsis. We can only say that it did not seem helpful in either
case.

CONCLUSION

Although no experimental research has been conducted on the nega-

tive health effects of secrecy, well-conducted experiments have been
shown that the revealing of private traumatic information or secrets in
anonymous, confidential settings can lead to health benefits, such as im-
proved immunologic functioning (e.g., Pennebaker & Beall, 1986; Penne-
baker et al., 1988; Petrie et al., 1995). It has been proposed that the reason
the revealing of secrets offers these benefits is that the revealer experiences
catharsis, gains new insights into the secrets (i.e., makes meaning out of
them), and no longer has to expend cognitive and emotional resources
actively hiding the secrets (Pennebaker, 1989, 1990; Pennebaker et al., 1990).
Along these lines, Polivy (1998) noted that expressing emotions behav-
iorally, or gaining catharsis, is helpful more often than not. However, as
described in this chapter, there is much contradictory evidence concerning
whether catharsis is indeed helpful (e.g., Bushman et al., 1999), just as there
has been mixed evidence regarding the benefits of high levels of insightful
thinking (e.g., Nolen-Hoeksema et al., 1997; Pennebaker et al., 1997).
WHAT IS IT THAT IS BENEFICIAL? 99

We (Kelly et al., 2001) conducted two studies in which we teased apart

the effects of gaining new insights and catharsis to see which makes people
feel better about their secrets. In Study I, undergraduates indicated
whether they had gained new insights or catharsis from revealing their
secrets to their confidants in the past. As it turned out, gaining insights was
associated with feeling positive about the secrets, whereas gaining ca-
tharsis was associated with feeling negative about them. In Study 2, under-
graduates were randomly assigned to write about their (1) secrets while
trying to gain new insights, (2) secrets while trying to gain catharsis, or
(3) previous day. The new-insights group felt significantly more positive
about their secrets than did the other groups. Moreover, they came to
terms with their secrets during the writing to a greater extent than did the
catharsis group. Not coming to terms with their secrets was associated
with participants' feeling negative about them. By obtaining these results,
we (Kelly et al., 2001) showed that focusing on getting a new perspective
on secrets is a superior means of making oneself feel more positive about
them. This strategy seems not only to be useful at increasing positive
affect, but also seems to encourage the person to come to terms with the
secrets, and thus diminish his or her negative emotions surrounding the
secrets. By contrast, solely focusing on stirring up and releasing negative
emotions may put the revealer at risk for feeling worse about the secret.
Integrating our findings with those of Pennebaker et al. (1997) and Nolen-
Hoeksema et al. (1997), who obtained mixed results regarding the benefits
of high levels of insightful thinking or analysis, we suggest that attempts at
meaning making are valuable to the extent that the person is able to come
to terms with their secret or trauma. Continued dwelling on a secret
without such closure may backfire, just as venting the negative emotions
surrounding a secret without gaining new insights may backfire.
CHAPTER 6

SECRECY AND OPENNESS

IN PSYCHOTHERAPY

In the previous two chapters, I described the benefits of revealing secrets

in a journal or diary, and in these remaining chapters, I move into a more
complex analysis of what happens when revealing secrets to another
person, starting with one's therapist. Psychotherapy clients could be ex-
pected to reveal their secrets completely to their therapists, no? Actually,
even though they were paying a good deal of money for their treatment,
46% of a sample of clients in long-term therapy (Hill et aI., 1993) and 60%
of a sample of short-term therapy outpatients (Kelly, 1998) reported keep-
ing some major secret from their therapists.
These findings are puzzling given that many theorists from the differ-
ent mainstream approaches to psychotherapy, such as the psychoanalytic,
family, and group therapy traditions, have emphasized the importance of
high levels of revelation from clients in psychotherapy (e.g., Foa, Roth-
baum, Riggs, & Murdock, 1991; Fong & Cox, 1983; Freud, 1958). For exam-
ple, Freud (1958) routinely explained to his patients that it was crucial for
them to reveal as much about themselves as possible in psychoanalysis, no
matter how ridiculous, unacceptable, or anxiety provoking those revela-
tions seemed to them. He referred to this directive as the fundamental rule
of psychoanalysis (see Hoyt, 1978).
Being open (i.e., truthful and revealing) is thought to allow clients to
release harmful pent-up emotions and gain an understanding of them-
selves and their problems (Hill & O'Grady, 1985; Hill et aI., 1993; Jourard,

101
102 CHAPTER 6

1971a,b; Martin, 1984; Stiles, 1987). Presumably, the more disclosing clients
can be about their thoughts and feelings, the more their therapists should
be able to help them. Fong and Cox (1983) noted that "until clients can
expose their innermost 'secrets' and make themselves vulnerable to the
counselor, the real work of counseling cannot begin" (p. 163). Likewise,
Jourard (1963) suggested that a requirement for mental health is the ability
to reveal one's inner self to at least one other person and that disclosing
secrets increases self-knowledge and psychological health. It follows that
revealing oneself, in time, to a trusted therapist would potentially have
great therapeutic effects. Along these lines, many therapists today require
a great deal of revelation from their clients, arguing that for clients to
benefit from therapy, the clients must first work through painful personal
experiences (e.g., Arnow, 1996; Courtois, 1992; Horowitz, 1986; Liotti, 1987;
Rando, 1993; Reichert, 1994).
In fact, psychotherapy techniques from a wide range of approaches
are aimed at promoting clients' self-disclosure (Sloan & Stiles, 1994). For
instance, if a client states that she is too ashamed to reveal a troubling
indiscretion from her past, the therapist may try to help her disclose by
describing what the therapist anticipates is an even more embarrassing
act. The rationale for such an intervention is that if the client hears the
description of a situation that is worse than her own, then she may have an
easier time describing her relatively minor indiscretion.
However, Bok (1982) has argued that the common, negative view of
silence and of secrecy in general may encourage people to make revela-
tions about themselves indiscriminately in social interactions. For exam-
ple, such revelations in group therapy can backfire and result in clients'
experiencing acute anxiety attacks or expressions of hostility from other
group therapy members, or in extreme cases even making suicide attempts
(Lieberman, Yalom, & Miles, 1973). Thus, the question remains, Are the
clients who resist their therapists' attempts at eliciting self-disclosure or
those who keep relevant secrets harming themselves?
In this chapter, I first provide an overview of the theories and tradi-
tions from the psychoanalytic, family, and group therapy literatures that
have led many therapists to believe that clients' secret keeping under-
mines the therapeutic process. Along the way, some alternative perspec-
tives within each tradition on the role of clients' openness are discussed. I
then review the empirical research on whether clients' openness is associ-
ated with favorable therapy process ratings and outcomes and conclude
by discussing the limitations of such evidence and offering suggestions
for future research.
SECRECY AND OPENNESS IN PSYCHOTHERAPY 103

THEORETICAL PERSPECTIVES
ON THE ROLE OF CLIENTS' OPENNESS

PSYCHOANALYSIS

Freud focused his theorizing on the concept of censorship, which en-

compasses both unconscious and conscious forms of inhibition; he tended
to use the terms repression and suppression interchangeably (see Wegner,
1992, for a discussion of this point). This is perhaps because he assumed
that one's personality was almost entirely unconscious, and so conscious
inhibition did not require much discussion. Therefore, when I discuss the
psychoanalytic emphasis on patients' openness, I refer to patients' revela-
tions of both secrets and repressed material together.

Benefits of Confession
Since the tum of the 20th century, psychoanalysts have viewed se-
crecy as problematic and accessing repressed material as being very help-
ful to patients. Psychoanalysis relies on the search for unspeakable, sealed-
off trauma; " ... behind an emotion expressed, behind a symptom mani-
fested, there lurks a contrary, repressed emotion" (Abraham, Torok, &
Rand, 1994, p. 18). Freud based his fundamental rule of psychoanalysis on
a number of case observations that patients' revealing of their secrets
allowed them to relive their repressed or buried traumatic experiences and
that this reliving was typically followed by a reduction in the patients'
symptoms (e.g., Breuer & Freud, 1975).
Jung (1933) also encouraged patients to face those things that they
typically repressed or kept hidden from themselves. He made patients'
revelations of previously hidden material the focus of his psychoanalytic
sessions, encouraging his patients to engage in a thorough confession of
the facts and suppressed affect surrounding their emotional experiences
(Jung, 1933). Jung stated that " ... every personal secret has the effect of sin
or guilt" (p. 34). He believed that if patients could be more conscious of
what they were concealing, then they would be harmed less by the buried
material. For example, if a patient admitted during his analysis that he
sometimes felt like killing his beloved mother, then he would be less
susceptible to blurting out unintended insults about her mothering skills
when visiting her during the holidays.
Even though Freud and Jung encouraged their patients to reveal their
secrets in treatment when the patients felt ready to do so, these famous
104 CHAPTER 6

analysts observed that their patients resisted complete openness (Kauf-

man, 1989). Freud thus inferred that patients are instinctively driven both
to be cured through revealing themselves completely and to avoid being
cured through keeping their unconscious images buried in the deeper
regions of their minds (Freud, 1958). This phenomenon has been described
as "a tension between the urge to retain and urge to expel" (Hoyt, 1978).
The patients fear penetration and yet they long for extraction of the secret
(Rosenfeld, 1980). When patients refuse to explore their secrets in psycho-
analysis, they are perceived to be defensive and resistant (Rosenfeld, 1980;
Minoff, 1992).
Psychoanalysis developed as a means of overcoming such resistance
to the revelation of hidden or repressed material (Bernfeld, 1941). Analysts
believe that by examining the manner in which patients disclose a secret,
they can gain insights into the patients' psychological development and
conflicts (Schoicket, 1980). Psychoanalytic sessions are devoted to trying to
remove obstacles to confession (Bernfeld, 1941). But at the same time,
analysts avoid rushing or outwardly forcing patients to reveal their deep-
est secrets (Rosenfeld, 1980). The analyst says and does things to reduce the
patient's discomfort or shame about the previously suppressed informa-
tion (Weiss, 1995). Freud drew an analogy between the psychoanalyst's
intense efforts to draw out hidden information and a criminal investiga-
tor's procedures to extract a confession (Welsh, 1994). The analyst hypoth-
esizes about what the patient's secret might be, comments on the resis-
tance in the patient, and makes an interpretation about the hidden content
to the patient (Bernfeld, 1941). The analyst also uses the primary psycho-
analytic techniques of dream analysis and free association (Le., in which
the analyst presents a series of words and the patient says what imme-
diately comes to mind) to elicit the revelation of hidden, unconscious
material (Handelman, 1981).
If patients are not willing or able to share their secrets, psychoanalysts
believe the energy that the patients spend keeping secrets repressed might
become represented in compulsive behavior or other symptoms (Abraham
et al., 1994; Freud, 1958; Hesselman, 1983; Margolis, 1966; Reik, 1945; Werge-
land,1980). For instance, a patient's denial or repression of a trauma may
contribute to the development of psychological disorders such as dissocia-
tive identity disorder (multiple personality disorder), bipolar disorder,
anorexia, and melancholia (Abraham et al., 1994). Some psychoanalysts
believe that only through the confession of secrets can the treatment of
these disorders succeed and symptoms be reduced (Abraham et al., 1994).
A number of case studies have been cited to support these claims. For
example, in one well-known case of a woman named Dora who suffered
from migraine headaches and depression, Freud identified and recon-
SECRECY AND OPENNESS IN PSYCHOTHERAPY 105

structed her traumatic childhood experiences (Mahony, 1996). He believed

that the source of Dora's illness was her childhood masturbation, and he
elicited her confession that she had masturbated. Her symptoms subse-
quently diminished. Another case involved a 41-year-old man who suf-
fered from alcoholism and drug abuse. His analysis included a focus on
uncovering his previously withheld sexual history and extracting his con-
fession of gay sexual feelings for the analyst (Wallerstein, 1986). After 7
years of analysis, the patient reached his treatment goals: his addictive
behaviors were gone, and he was leading a satisfying life with his wife and
children.

Why Revealing Is Critical for Therapeutic Progress

Psychoanalysts have offered several interconnected explanations for
why revealing secrets is essential for therapeutic progress. First, revealing
the secret is believed to allow catharsis or a discharge of pent-up emotions
such as shame, guilt, and anger that previously inhibited creative growth
(Hoyt, 1978). Hymer (1982) speculated that the benefits of revealing private
information outweigh the risks: "The initial embarrassment of confessing
is frequently outweighed by the relief that comes with the verbalization of
the darker, secretive aspects of the self" (p. 131).
Second, while helping the patients to overcome anxiety and guilt,
confession also theoretically gratifies the patients' masochistic needs for
punishment (Gillman, 1992; Reik, 1945). Patients seem compelled to re-
solve unfinished business by reenacting past traumatic events (Dushman
& Bressler, 1991). Traumatized patients "fear the return of the trauma, yet
they tend to retraumatize themselves through repeated actions symboliz-
ing aspects of the trauma" (Bergmann, 1992, p. 452). Confession allows an
emotional reliving of the Original trauma, and thus presumably provides
partial gratification of the patients' drives and impulses to punish them-
selves (Reik, 1945). Dushman and Bressler (1991) conducted a case study of
a 17-year-old male who used drugs and was in psychodrama group treat-
ment. They observed that he was able to find closure on a friend's death
through revealing this secret in treatment. The group provided him with a
safe environment to recover from his guilty feelings about having survived
his friend and find meaning in his own survival.
Third, the disclosure of secrets may help patients develop their identi-
ties (Hymer, 1982). The revealing of a secret presumably makes it less
foreign to their sense of who they are and dispels their uncomfortable
moral burden of deceptiveness (Schwartz, 1984). Thus, the revealing al-
lows an integration of the positive and negative aspects of themselves
(Hymer, 1982; Schwartz, 1984). In one case study, a psychiatric inpatient
106 CHAPTER 6

hid from other patients the fact that he had killed one of his parents
(Schwartz, 1984). The patient later reported feeling isolated from the other
patients on his ward as a result of this concealment. Schwartz (1984)
recommended that therapy patients reveal their secrets to other patients
when they feel ready to do so.
A common theme to all these explanations is that revealing secrets to a
noncondemning analyst instead of to rejecting parents is a corrective
emotional experience (Margolis, 1974). In contrast to past experiences, the
analyst presumably will not reveal the secret, will not take sides, and will
not use the information against the patient (except where bound by the
legal limits of confidentiality). Through confessing the buried impulses
and the mechanisms driving toward repression of these impulses, a better
adjustment to reality may replace the process of repression. Confession
offers a more realistic meaning for the repressed wishes and the possibility
for greater self-understanding (Reik, 1945).
In sum, psychoanalysis developed as a means of helping patients
reveal themselves to the analyst and confess their secrets, and the revela-
tion of secrets has become the cornerstone of modem psychoanalysis
(Castets, 1988). Drawing largely from case studies, psychoanalysts have
contended that keeping secrets in treatment prevents patients from bene-
fiting from the treatment. In these case studies, patients typically have
recovered following an important confession or retelling of previously
suppressed or repressed material.

Alternative Perspectives
New avenues for research exploration often can be gleaned from case
studies. However, it is very difficult to draw conclusions or generalize
from such studies. Although Freud considered himself a scientist (Barron,
Beaumont, Goldsmith, & Good, 1991), Bernfeld (1941) has noted that the
case study methods of psychoanalysis deviate greatly from the scientific
method of observation. In particular, compared with other research de-
signs, such as correlational or experimental designs, case studies are espe-
cially vulnerable to the biases of the investigators. Because psychoanalysis
grew out of the notion that very high levels of patient revelation are
essential for therapeutic progress, perhaps psychoanalytic researchers
have expected to find that secrecy in sessions is problematic. For example,
Coons (1986) surveyed psychotherapists treating patients with dissocia-
tive identity disorder and found that the therapists rated secrecy as a major
hindrance to improvement. Thus, it is possible that psychoanalytic re-
searchers may not have documented the times when revealing a secret
went awry because that result was so different from their expectations.
SECRECY AND OPENNESS IN PSYCHOTHERAPY 107

One psychoanalytic researcher concluded, JlPractica1 experience ...

suggests that total disclosure is not necessaryJl (Hoyt, 1978, p. 238). Some
psychoanalysts have proposed that there are even times when keeping
secrets may be beneficial. Margolis (1966) suggested that retaining secret
thoughts and feelings leads to a healthy sense of separateness, individu-
ality, and ego identity. He also indicated that people learn about their
uniqueness through recognizing that they have the power to divulge or
withhold information. Likewise, Hoyt (1978) argued that secrets are im-
portant for establishing interpersonal boundaries. With the knowledge of
a secret comes a certain amount of power, and others may abuse the power
accompanying secret knowledge (Hoyt, 1978).
A problem with very high levels of revelation in psychotherapy is
that, although analysts are depicted typically as noncondemning, neutral
listeners who will help the patient with the secret and will not utilize
information against the patient (Margolis, 1974), this depiction is not al-
ways accurate. Hoyt (1978) and Hoffman (1983) warned that when coun-
tertransference (i.e., the thoughts, feelings, and impulses that the therapist
has toward the clients) occurs, the analyst may not provide the nonjudg-
mental environment that patients so desperately need when the patients
are revealing painful or shameful deeds. Furthermore, Silver (1983) has
indicated that it is risky to agree to unconditional secrecy when treating
clients who have a borderline personality disorder, because such collusion
may interfere with the building of a trusting relationship. Confidentiality
may have to be broken in the event that the borderline client attempts
suicide or performs other dangerous acts (Silver, 1983). Another harmful
effect may occur if the patient becomes too ashamed or embarrassed about
the revelation and decides to terminate therapy prematurely (Hoyt, 1978).
In sum, the case studies from the psychoanalytic tradition lend only
initial, tentative support for encouraging very high levels of revelation
from clients. Furthermore, some psychoanalytic theorists have offered
examples of when secrecy in psychotherapy may actually be productive.

FAMILY THERAPY

Detriments of Secrecy
Although psychoanalysts were the first modem-day psychotherapists
to emphasize the importance of uncovering secrets in therapy, family
therapists may have had the most to say about the negative effects of
secrecy. The strain of keeping a secret is often considered by family thera-
pists to be the source of clients' problems, and symptoms are thought to be
mere by-products of the secret keeping (e.g., Saffer et al., 1979). For exam-
108 CHAPTER 6

pIe, a child may exhibit symptoms as a diversion from the family's denied
or secret problems (Eaker, 1986). The parents may bring a child into ther-
apy because it is easier to blame or scapegoat the child than it is to face
their own problems (Pincus & Dare, 1978). Indeed, there have been docu-
mented occasions when secret keeping seemed to be linked to psychotic
symptoms in children (Saffer et al., 1979).
Secrets are thought to control family members and keep them bound
to one another by making the members feel an unhealthy sense of obliga-
tion to the family (Avery, 1982). It has been suggested that when one family
member wants to separate from the secretive family, these families are
often driven into therapy (Gutheil & Avery, 1977). Hence, some therapists
have construed the main purpose of therapy as helping the family with
their secrets while also helping them to stay together (Avery, 1982).
The notion that secrecy in families is dysfunctional has received some
empirical support. Vangelisti (1994) showed that there is a negative rela-
tionship between family members' reports of the number of secrets that
their family keeps in relation to other families and the members' levels of
satisfaction with their family. However, the actual estimated number of
secrets that family members reported was not correlated with their family
satisfaction (Vangelisti, 1994). Vangelisti (1994) concluded that dissatisfied
family members may be inclined to believe that their families are unusu-
ally secretive, even if the families are not actually unusually secretive.
As described in Chapter I, nearly every family has secrets (Vangelisti
& Caughlin, 1997). The types of family secrets that have received the most
attention in the literature include secrets about incest (Swanson & Biaggio,
1985), extramarital affairs (Shlien, 1984), the sexual orientation of one or
more family members (Murphy, 1989), and death (Paul & Bloom, 1970). In
the paragraphs to follow, I describe the rationales for emphasizing the
revealing of these various types of secrets in family therapy. I have orga-
nized the material this way because the rationales vary to some extent
based on the types of secrets the family is keeping.
Secrets surrounding incest are considered particularly troubling for
the family. Incest victims have reported that they fear being abandoned
(Kaufman, Peck, & Tagiuri, 1954), not being believed (Butler, 1978), being
punished (Herman, 1981), and being blamed for complying with the inces-
tuous activity (Geiser, 1979; Goodwin, 1982; Justice & Justice, 1979; Meisel-
man, 1978). In cases in which a father has molested his daughter, secrecy is
thought to isolate the family from the outside world, thereby contributing
to the father's abusive dominance in the family (Hoorwitz, 1983). Repres-
sion or denial are typical responses to the incest experience (Lindberg &
Distad, 1985; Russell, 1986), and such responses allow the incest to con-
tinue (Hays, 1987). The child feels isolated and different from other chil-
SECRECY AND OPENNESS IN PSYCHOTHERAPY 109

dren her age (Luebell & Soong, 1982). There is tremendous pressure on her
to maintain secrecy (Swanson & Biaggio, 1985). The daughter understands
that revealing her secret could destroy her family and could mean her
being removed from it (Swanson & Biaggio, 1985).
Despite these fears, Swanson and Biaagio (1985) argued that the vic-
tim must reveal the secret during therapy sessions. A critical part of
therapy is to diminish the daughter's sense of isolation by encouraging her
to break the silence surrounding the incest trauma (Black, 1981; Hoorwitz,
1983). The victim also ultimately must discuss the incest with someone
other than the therapist so that she can rid herself of her feelings of
isolation (see Herman, 1981; Meiselman, 1978; Tsai & Wagner, 1978). Group
therapy for adult survivors of incest often requires not only disclosing the
secret of the incest to the therapist and other group members (Hays, 1987)
but also confronting the family with the secret in an effort to rebuild one-
self (Swink & Leveille, 1986).
Some family therapists have suggested that even though the initial
relief of revealing of family secrets in therapy may be followed by feelings
of guilt and an increase in symptoms, the revealer ultimately will enjoy
symptom reduction. Three documented cases involving family sexual
secrets showed this pattern (Saffer et al., 1979). In each case, an adolescent
was concealing information about the inappropriate sexual conduct of one
or more family members, and in each case the adolescent was encouraged
by the therapist to reveal this secret. After the adolescents complied, they
experienced some immediate abating of their symptoms of paranoia, act-
ing out, and severe depression. However, these symptoms then increased
during the days immediately following the disclosure. After more time
had passed, their symptoms lessened once again. Saffer et al. (1979) sug-
gested that this pattern was due to the fact that the patients initially felt
that they had been disloyal to the family by revealing their secrets and then
the patients regained a sense of equilibrium after coming to terms with this
sense of disloyalty, even though their families dissolved.
In cases in which one spouse has had an extramarital affair, many
marriage and family therapists have indicated that it is critical for the
infidel to confess to the other spouse (e.g., Brown, 1991; Pittman, 1989;
Shlien, 1984). As described in Chapter 1, their reasoning is that it is not the
extramarital sex that causes problems but rather the secrecy surrounding
the sex (Pittman, 1989). Once the person has had the affair, it takes energy
to keep the secret by omitting truths or telling lies, and these deceptive
efforts may destroy a person's sense of self-worth (Shlien, 1984). The guilt
and shame of the secret affair may simply become torturous for the person
who is having the affair (Shlien, 1984).
In addition, the "other woman" (or the" other man") also may be hurt
110 CHAPTER 6

by the secrecy surrounding the affair (Richardson, 1988). Richardson (1988)

conducted interviews with 65 women who were involved with married
men and found that the relationships typically were maintained by both
parties' efforts to keep the affairs hidden. This secrecy resulted in both
positive and negative feelings by creating intense feelings of togetherness
for the couple, but it also undermined the woman's sense of power within
the relationship (Richardson, 1988).
Some marital therapists have asserted that secret extramarital affairs
should be revealed because hiding the affair means keeping a piece of
oneself from one's spouse (Brown, 1991, p. 138). Brown (1991) postulated
that when clients bring up the affairs to their therapists, such as in a private
session away from their spouses, it is apparent to the therapists that the
affairs are still having an impact. The therapists presumably need to be
aware of the affairs to try to understand their impact on the marriage
(Westfall, 1989), and then hopefully the therapists can assist the clients in
sharing the affair with the spouse (Brown, 1991).
One case study involved a couple who vaguely expressed their goal of
improving their marriage. In an individual therapy session with the hus-
band, the therapist learned indirectly about the affair (Brown, 1991). Just
before the next session, it was the wife who canceled their treatment, and
this abrupt termination was interpreted to mean that the wife and husband
were trying to avoid facing the affair. It was thought that both could have
benefitted from an open discussion of the infidelity (Brown, 1991).
The literature on revealing one's gay or lesbian identity to others sug-
gests that haVing an open identity across all domains of one's life is
associated with increased self-esteem (e.g., Cass, 1979; Hencken & O'Dowd,
1977; Ponse, 1978). In particular, coming out to one's parents is thought to
be central to a lesbian's self-acceptance (Sophie, 1988). When a daughter
hides her lesbian identity from her parents, "what closeness exists is
considered pseudocloseness, because a relevant piece of information is
missing" (Murphy, 1989, p. 49). Through revealing her lesbian identity, the
daughter can dispel this sense of isolation from her parents (Roth, 1985).
However, many gay men and lesbians who have revealed their sexual
orientation to their parents have reported that their parents disapproved
of this fact (Chafetez, Sampson, Beck, & West, 1974; Jay & Young, 1979;
Mendola, 1980). For example, in a survey of 20 lesbians in committed
relationships who had come out to their parents, Murphy (1989) found that
70% of the women reported that their parents disapproved of their sexual
orientation. At the same time, the women in that study also reported
feeling a sense of relief that they did not have to conceal such an important
aspect of their lives. Thus, Murphy (1989) concluded that the disappoint-
ing effects of coming out were outweighed by the positive aspects of
having parents know about the nature of the couple's relationship.
SECRECY AND OPENNESS IN PSYCHOTHERAPY 111

Murphy (1989) recognized the costs to revealing one's sexual orienta-

tion to important others such as one's parents. She recommended that in
therapy with lesbian couples, the therapist should help clients balance the
effects of secrecy against the consequences of coming out in what she
referred to as "identity management." In particular, Murphy suggested
that therapists should help the couple prepare for the effects of coming out
to parents and then help the couple grieve over the loss of "heterosexual
privilege" in the family of origin.
Secrets about death and loss of a family member (e.g., through di-
vorce) also are considered to be highly detrimental to family functioning.
For example, after conducting two family therapy case studies, Evans
(1976) concluded that the dysfunctional secret surrounding the death (or
impending death) of a family member contributed to the development of
symptoms in one child in each family. The secret seemed to intensify
feelings of alienation among the members of each family. The interven-
tions aimed at bringing the secret out in the open and resolving it seemed
to lead to substantial improvements in the clients' functioning (Evans,
1976).
Paul and Bloom (1970) postulated that families in crisis, such as in the
case of an imminent death, necessarily have secret or denied emotions that
members often project onto one scapegoated family member. They pro-
posed that such secrets need to be addressed directly in therapy (Paul &
Bloom, 1970). A case of several families treated in multifamily group
therapy illustrates this phenomenon (Paul & Bloom, 1970). For one family
in the group, the suicidal thoughts of the son, John, were viewed by
therapists as part of the effort to keep the secret of his father's terminal
illness from the father: "It became apparent that the strain of hiding his
serious illness from Mr. Henry had produced frantic behavior in the
family, leading to reinforced scapegoating of John" (p. 43). With the thera-
pist's encouragement, the family told the father of his illness. Although
John's self-esteem improved after the disclosure, the family abruptly ter-
minated treatment, and it was the parents who initiated the termination.
The authors indicated that if the family had remained in treatment, they
ultimately might have developed more functional and open patterns of
communication.
In another case study, an asthmatic to-year-old son revealed in play
therapy his secret wish that his divorced father would come back to the
family or would at least befriend his mother (Eaker, 1986). Shortly after
such discussions, the boy's symptoms lessened. However, his 12-year-old
brother subsequently developed some misbehaviors at home and at
school. Despite these mixed findings, Eaker concluded that although fami-
lies may resist revealing their secrets, they often are helped by such open
discussion (Eaker, 1986). Play therapy is seen as a "cushion" that can allow
112 CHAPTER 6

the family to tolerate the anxiety associated with the revealing, for many
families drop out after secrets emerge in standard family therapy (see
Winnicott, 1980).
Researchers who have studied families with a member who has AIDS
have found that the families expend a great deal of energy keeping the
secret because of the social stigma associated with the disease (e.g., Bor,
Miller, Scher, & Salt, 1991; Greif & Porembski, 1988; Miller, Goldman, &
Bor, 1994). Two case studies of families dealing with HIV-infected mem-
bers showed that, regardless of how the disease was contracted (i.e.,
because of hemophilia or sexual promiscuity), both families worked very
hard to hide the disease (Miller et al., 1994). Moreover, extensive interviews
of 11 significant others of person who died of AIDS showed that all 11
reported that there was some attempt to keep the AIDS a secret from
others. Forty-four percent of the families of the AIDS victims reported that
it was stressful for them to maintain the secrecy (Greif & Porembski, 1988).
It seemed that the burden of the secret over time left these significant
others feeling isolated from their own support networks (Greif & Poremb-
ski, 1988). Thus, Bor et al. (1991) suggested that to address feelings of
loneliness and isolation, the therapist should attend early on in treatment
to the secrecy of the illness and who could be told about the illness.
Greif and Porembski (1988) provided a more cautious perspective on
revealing the secret of AIDS by suggesting that despite the potential strain
of secret keeping, secrecy may be the best way for the person with the dis-
ease to cope with his or her immediate crisis. Dealing with the disease itself
is so stressful that the patient may not have the resources to cope with
the added strain of managing other people's reactions to the disease.
Insum, as with the psychoanalytic literature, there is a preponderance
of theoretical work that suggests the revealing of secrets is a critical part of
family therapy. Yet the evidence supporting this claim has been composed
almost entirely of case reports, which offer limited generalizability. In the
next section, I suggest some alternative ways of looking at the functions
of secrecy in family therapy.

Alternative Perspectives
Some therapists have put secret keeping in the family in a positive
light by describing shared family secrets as ones that strengthen the family
boundaries from the outside world (Karpel, 1980). It is believed that the
occasions in which the therapist must insist on disclosure of a secret occur
when she or he learns that one family member is keeping a secret for the
sake of another member (Karpel, 1980; Palazzoli & Prata, 1982). Such
concealment places the therapist in an ethical dilemma that involves be-
SECRECY AND OPENNESS IN PSYCHOTHERAPY 113

traying the trust of the person kept unaware (Karpel, 1980; Palazzoli &
Prata, 1982). These therapists systematically refuse to allow secret revela-
tions, unless the revelations are made in the presence of the entire family
in treatment. The content of the secret is not considered important, nor is
the fact that there is a secret, but rather that the therapist is being offered a
coalition against someone else in treatment, an undertaking that presum-
ably can undermine treatment completely (Palazzoli & Prata, 1982).
Kaslow (1993) also has taken a stance that is contradictory to most
family therapy perspectives on secrecy and on secret affairs in particular.
She has complained that in some marriages full disclosure between the
dyad is unduly expected or demanded. Thus, having an extramarital affair
may seem like the only way for one partner to assert a sense of identity that
is separate from that of the other partner (Kaslow, 1993). Kaslow has asked
difficult questions of therapists, such as: Who are therapists to claim that
they know some absolute truth about the benefits of revealing secrets?
Should clients not be permitted to decide that for themselves?
A number of women have reported that their secret affairs actually
added to their happy marriages and reawakened important aspects of
their identities (Heyn, 1992). If a therapist requires that no secrets be kept
between the couple, such a requirement may cause the partner who is
having the affair to flee treatment rather than admit to the affair (Kaslow,
1993). Therefore, it has been suggested that the therapist should take a
neutral stand in helping a client decide whether revealing a secret is in the
client's best interest (Moultrop, 1992).
Some family therapists have postulated that keeping secrets is part of
healthy family functioning, such as when parents appropriately keep
private details of their sex lives from the children (Grolnick, 1983; Imber-
Black, 1993), and some have even prescribed secret keeping for family
members who have no secrets (Adams, 1993). Adams (1993) suggested that
such withholding could revitalize romantic relationships by adding a
sense of intrigue to them and could allow for differentiating relationships
that are too enmeshed.
How the family interacts over a secret is considered more critical in
determining its function in the family than the actual content of the secret
(Imber-Black, 1993). One case study involved parents who were keeping
the secret that the child's biological father had committed suicide (Adams,
1993). Because of the parents' fears of disclosure, the therapist intervened
with questions targeting the functioning of the secret (e.g., who knew and
who did not know the secret), rather than the revelation of the content
itself. The recommendation was that the timing of the revelation would be
best left up to the family (Adams, 1993).
In sum, although family therapy theorists generally have taken a
114 CHAPTER 6

negative stance toward keeping secrets in therapy, there are several note-
worthy exceptions. Some therapists have even prescribed secret keeping
and boundary setting in families where there is too little secrecy.

GROUP PSYCHOTHERAPY

Therapeutic Effects of Revealing Secrets

Yalom (1985) argued that actively hiding information from the thera-
pist or other group members can take energy away from constructive
revelation and group interaction. A premise of group psychotherapy is
that over time when people are able to be open and honest with one
another they can learn about themselves through getting feedback about
how they come across to other people (Yalom, 1985). Theoretically, if they
expend effort hiding a secret, they may become less spontaneous in their
interactions with other group members and participate in the group on
only a superficial level (Yalom, 1985). Moreover, revealing in group ther-
apy is thought to allow clients to release and accept their feelings of shame
surrounding their secrets (Asner, 1990; Silverstein, 1993; Winter, 1985). For
example, in a psychotherapy group composed of six Jewish members, the
members shared their secrets about the perceived stigma of being Jewish,
and this sharing of shame seemed to build a sense of trust among the
members (Klein, 1976). Keeping secrets also is thought to lead to subgroup
formation and the premature termination of group members (Silverstein,
1993). In a case study of Jews and Arabs and their relationships in psycho-
therapy training groups, the secrets the ethnic subgroups kept among
themselves seemed to undermine the progress of the entire training group
(Rippa, 1994). Presumably, a group becomes "stalemated" until a group
member who has been keeping a secret from the group finally reveals it
(Hough, 1992).
Hough (1992) stated that the "destructiveness of secrets in group
psychotherapy is well known" (p. 107). As such, the revealing of secrets is
often the focus of group psychotherapy (Bloch & Reibstein, 1980), espe-
cially for problems such as incest abuse (Hays, 1987; Sturkie, 1983), alcohol-
ism (Bingham & Bargar, 1985), bulimia (Asner, 1990), and dissociative
disorders (Buchele, 1993). For example, Bingham and Bargar (1985) pro-
posed that a central task in the group treatment "of the latency-age child
from an alcoholic home is to encourage the child to share the dreaded
family secret with other people in a supportive and trusting atmosphere"
(p.15).
In the group treatment of persons with bulimia, sharing secrets may
attenuate the strong feelings of alienation and shame that often accom-
SECRECY AND OPENNESS IN PSYCHOTHERAPY 115

pany bingeing and purging (see Asner, 1990). Case studies have pointed to
secrecy and the "bad self" as one of six themes that emerge in group
psychotherapy for women with bulimia (Weinstein & Richman, 1984).
Across three groups, Weinstein and Richman (1984) found that of the 16
women who attended at least 20 meetings, 6 had totally stopped binge
eating and 8 had reduced the frequency of bingeing. All group members
reported improvements in self-concept, ability to identify feeling states,
and experiencing a sense of control over their lives. The authors suggested
that it was the elaboration of these themes, the foremost of which was
secrecy, and their expression in the group that were therapeutic for these
women.
Often clients reveal secrets in individual therapy sessions that they
feel too ashamed of to share with the group. Wright (1990) described a case
of a woman who remained silent during the first 2 years of group therapy,
even though she had shared private material during individual therapy
sessions. She ultimately emerged as a very powerful and articulate group
member who simultaneously became successful in her management ca-
reer. Wright (1990) attributed this development to providing her with the
time she required to feel safe enough to share secrets with the other group
members, and he noted that individual and group therapy work well
together to prepare a client for essential revealing in the group context.

Alternative Perspectives
Some group psychotherapists believe that the content of the commu-
nication of secrets is generally irrelevant and that the goal of therapy must
be to enhance mature communications among group members, not merely
to provide an atmosphere for revelation of secrets and probing the uncon-
scious (Kirman, 1991). Too much probing in group therapy may be counter-
productive for psychiatric inpatient adolescents in particular (Amini,
Burke, & Edgerton, 1978). Amini et al. (1978) observed 11 male and 11 fe-
male inpatients who attended numerous meetings with other patients and
staff in which tensions between patients and staff were openly expressed.
After 3 months, the adolescents reported that they felt overexposed and
overunderstood. This need for secrecy and privacy seemed to be a func-
tion of the adolescents' stage of development and needing a sense of
independence from adults (Amini et al., 1978).
Five time-limited (10 sessions) psychotherapy groups for women who
were survivors of incest abuse provided additional examples of group
members that may have been overexposed (Herman & Schatzow, 1984).
During the middle sessions, the group leaders encouraged members to
discuss in greater detail their incest experiences. Their aim was to help
116 CHAPTER 6

members resolve issues of secrecy, shame, and stigmatization associated

with incest, but the effectiveness of this approach was questionable. At
6-month follow-up, although most of the women reported that they had
higher self-esteem, some of them reported that sex was worse for them.
Revealing private information can have negative effects beyond what
may occur during the group therapy sessions. Davis and Meara (1982)
warned that it is difficult to ensure group confidentiality because group
members are not bound by the same legal and ethical rules as are thera-
pists. Group members may reveal the details of another group member's
sexual encounters, private fears, and personal traumas to people outside
of the group. Hence, group members may be hurt by their revealing secrets
to other members (Davis & Meara, 1982).
In sum, as demonstrated above, a number of group therapists believe
that it is important for clients to reveal their secrets to the group when they
feel ready to do so. When group members do keep secrets, their withhold-
ing sometimes is seen as a form of acting out in the group sessions (Silver-
stein, 1993). Thus, the expectations therapists sometimes place on their
clients may set clients up to feel what Bok (1982) has called a counter-
productive "compulsion to confess" (p. 79). Members may come to feel
guilty about violating the expectations for participation when they do
choose to keep secrets, which could have negative effects on the members.

SUMMARY

The various theoretical approaches to psychotherapy offer different

yet overlapping explanations for why revealing secrets in therapy is bene-
ficial. Psychoanalysts believe that revealing secrets allows patients to
avoid the neurotic symptoms associated with expending energy to keep
undesirable impulses suppressed (e.g., Abraham et al., 1994). Confession
allows a release of pent-up energy, or catharsis, and alleviates feelings of
shame and guilt while presumably satisfying one's masochistic impulses
(e.g., Gillman, 1992; Reik, 1945). Family therapists believe that addressing
secrets helps the family avoid scapegoating a particular family member
and allows them to deal with dysfunctional interactions among family
members (e.g., Paul & Bloom, 1970). Addressing secrets can help alleviate
the symptoms (e.g., hostile outbursts, depression, and suicide attempts)
exhibited by the scapegoated family member that may have been caused
by the strain of keeping the family secrets. In particular, in cases of reveal-
ing incest or other types of abuse, the victim can feel protected and less
isolated from other family members and the outside world (e.g., Black,
1981; Hoorwitz, 1983). In group therapy, revealing secrets allows group
members to have meaningful interactions with other group members and
SECRECY AND OPENNESS IN PSYCHOTHERAPY 117

can provide them with a sense that they are not alone in their troubles (e.g.,
Yalom, 1985).
As noted above, the majority of the studies that have pointed to the
importance of being open in therapy have been case studies. These case
studies offer a good starting point for subsequent research, but they are too
vulnerable to the biases of the researchers (who are likely to expect that
secret keeping is problematic) to represent a conclusive perspective on
secret keeping in therapy. I believe that it is now time to reevaluate the
general efficacy of high levels of revelation in therapy and to assess when
such revealing leads to positive or negative outcomes.

EMPIRICAL FINDINGS ON THE ROLE

OF CLIENTS' OPENNESS IN THERAPY

A Consumer Reports (November, 1995) survey of roughly 4000 psycho-

therapy clients showed that the clients who "formed a real partnership
with their therapist-by being open, even with painful subjects, and by
working on issues between sessions-were more likely to progress" (p.
739). This finding resonates with the widely accepted belief that high levels
of clients' openness are therapeutic. In the following paragraphs, I provide
a close look at the empirical evidence on this position by reviewing the
research on clients' disclosure and covert processes in psychotherapy.

CLIENTS' DISCLOSURE

Clients' disclosure is defined as the clients' revealing of thoughts, feel-

ings, perceptions, or goals in therapy (Stiles, 1995; Stiles & Sultan, 1979).
Researchers have frequently measured clients' disclosure by having ob-
servers rate these intentions and then tallying the percentage of utterances
during the therapy session that qualify as disclosure (e.g., Stiles, 1995; Stiles
& Sultan, 1979). Stiles and colleagues (McDaniel, Stiles, & McGaughey,
1981; Stiles & Shapiro, 1994; Stiles & Sultan, 1979) showed that across all
types of psychotherapy, clients typically had disclosure intent for 40-60%
of their utterances. This high percentage makes the therapy client role
distinct from other expository roles in which disclosure occurs less fre-
quently, such as the medical patient and the courtroom eyewitness roles
(McGaughey & Stiles, 1983; Premo & Stiles, 1983; Stiles, Putnam, & Jacob,
1982).
The high percentage also supports the claim that "disclosure is at the
heart of psychotherapy" (Stiles, 1995, p. 71). High levels of clients' dis-
closure are seen as a very promising indicator of good therapy process
118 CHAPTER 6

(Stiles, McDaniel, & McGaughey, 1979). It is believed that by disclosing in

the first person (i.e., using "I" statements), clients come to understand their
own points of view and to take responsibility for their feelings and actions
(Stiles et al., 1979).
There is surprisingly little research on the relation between clients'
disclosure in psychotherapy and improvement rates. The research is scant
on this relation because investigators often have used clients' disclosure
itself as a measure, as opposed to a predictor, of outcome (e.g., Halpern,
1977; Kremer & Gesten, 1998; Strassberg & Anchor, 1977).
Perhaps surprising too is the fact that the research that has specifically
studied levels of clients' disclosure in psychotherapy has not supported
the assertion that more disclosure is associated with improved therapy
outcomes. For example, in a study of 18 hospitalized schizophreniCs, the
patients who made fewer self-disclosures during a 10-week period of
group therapy sessions actually showed greater posttherapy improve-
ments in their intelligence test and Minnesota Multiphasic Personality
Inventory (MMPI) scores (Strassberg et al., 1975). These patients also re-
ported a greater reduction in symptoms than did their more disclosing
counterparts (Strassberg, Roback, Anchor, & Abramowitz, 1975).
In another example, Stiles and Shapiro (1994) studied 39 clients who
participated in the 16-session Sheffield Psychotherapy Project (Shapiro &
Firth, 1987) and found that increases in clients' disclosures across a
cognitive-behavioral phase of treatment were significantly correlated
with increases in their depression. They also found that overall, after a
combination of both interpersonal and cognitive-behavioral phases of
treatment, there were no significant relations between (1) the percentage of
clients' disclosures in sessions and (2) changes in symptomatology and
depression scores from intake to termination (Sloan & Stiles, 1994; Stiles,
1995; Stiles & Shapiro, 1994).
Likewise, McDaniel et al. (1981) studied 31 male college students in
time-limited psychotherapy and found that there was no relationship
between the percentage of clients' disclosures in sessions (as rated by
trained coders) and clients' improvement rates. These outcome ratings
were made across intake, termination, and I-year follow-up by the clients,
their therapists (intake and termination only), and independent clinicians
(McDaniel et al., 1981). In addition, in an investigation of the ongoing
psychotherapy of 11 clients (seen at a university psychology department
clinic or in private practice), Stiles (1984) found that the percentage of
clients' disclosures was not correlated with the clients', therapists', and
external raters' perceptions of session depth and value. The three raters
also observed that the sessions that were unusually high in disclosure
(relative to the clients' usual levels) were rough, unpleasant, difficult, and
dangerous (Stiles, 1984).
SECRECY AND OPENNESS IN PSYCHOTHERAPY 119

Researchers studying encounter groups have shown that revealing

powerful negative emotions to other group members can backfire with
devastating effects. Lieberman et al. (1973) conducted what has been called
"the best-executed study" on encounter groups among those studies con-
ducted in the 1960s and early 1970s (Hartley, Roback, & Abramowitz, 1976).
At the height of the encounter group movement, they studied 15 encounter
groups composed of undergraduates led by well-trained, established facil-
itators with varying theoretical approaches. The groups were all limited to
30 hours and the members were assessed for their psychological function-
ing at the beginning of the groups, at group termination, and at 6- to
8-month follow-up. Their psychological functioning was assessed based
on questionnaires given to the leaders, participants, co-participants, and
observers and based on interviews with participants. Using a criterion of
downward change on three or more measures, the researchers discovered
that 16% of the group members showed negative change, including 8%
that were considered "casualties" who showed more psychological dis-
tress and/or used more maladaptive defenses as a direct result of the
group experience. One member even committed suicide after the second
group therapy session. A main finding was that encouragement of con-
frontation and expression of anger were associated with negative out-
comes for the targets of those attacks (Lieberman et al., 1973).
How can the fact that high levels of clients' disclosure are not linked to
favorable therapy outcomes be explained? One possible explanation stems
from the fact that the clients who are the most distressed tend to disclose
the most (McDaniel et al., 1981; Stiles, 1984; Strupp & Hadley, 1979). Stiles
(1995) has argued that because the most distressed clients tend to get the
least out of therapy (see Garfield, 1994; Stiles, 1987), the benefits of high
levels of disclosure are not apparent.
Another explanation is that it may be the quality (i.e., the intimacy and
relevance), as opposed to the quantity, of clients' disclosures is related to
improvement. There again, though, the evidence does not support the
claim that more disclosure is better for clients. In a study of 53 clients, it
was discovered that therapists' perceptions of their clients' "willingness to
share material of a personal and intimate nature" were positively related
to the therapists' ratings of the clients' improvement (Strassberg, Anchor,
Gabel, & Cohen, 1978). However, an independent coder's ratings of the
intimacy of clients' actual self-disclosure in the sessions were not signifi-
cantly related to the therapists' ratings of the clients' improvement
(Strassberg et al., 1978). Thus, it seems that therapists may perceive that
intimate self-disclosure is related to outcome more than it actually is.
Another possibility is that therapy clients experience delayed benefits
to disclosure that have not yet been documented in the client disclosure
literature. Along these lines, Foa et al. (1991) observed a pattern of delayed
uo CHAPTER 6

benefits of exposure therapy when they compared the effectiveness of

exposure therapy with other treatments for reducing posttraumatic stress
disorder (PTSD) in female rape victims. The treatments used were pro-
longed exposure (PE), stress inoculation training (SIT), and supportive
counseling (SC). The rationale behind using PE is that the fundamental
task of any therapy for fear or anxiety is to identify the stimuli that elicit
the fear and then provide corrective information to alter the memory
structures (Foa & Kozak, 1986; Foa & Rothbaum, 1989). Exposure therapy
theoretically allows habituation to the fear and may be especially useful
when the disorder involves excessive avoidance of the problem (Foa &
Kozak, 1986). Foa and Rothbaum (1989) offered the following illustration
of the treatment of PTSD in rape victims:
Seven sessions are devoted to re-experiencing in imagination the rape scene.
Clients are instructed to try to imagine as vividly as possible the assault scene
and describe it aloud. The exposure is paced to allow the most anxiety-
provoking details only during the later sessions. The narrations are repeated
in their entirety several times for 60 minutes per session and tape-recorded
for playback at home at least once daily (p. 223).

Also it turned out in Foa and coworkers' (1991) study, SIT seemed superior
immediately after treatment, but at the follow-up assessment, PE evi-
denced more lasting improvement. These researchers explained this pat-
tern of results by stating that the procedures utilized in PE are expected to
produce high levels of arousal initially because patients are asked to
repeatedly confront the rape memory, and then these procedures are likely
to cause permanent positive changes in the rape memory.
Likewise, in their experiments with undergraduates, Pennebaker and
his colleagues (Harber & Pennebaker, 1992; Pennebaker, 1990; Pennebaker
& Beall, 1986; Pennebaker et aI., 1988) observed that even though partici-
pants often reported feeling worse after anonymously writing about
traumas, they later felt better as a result of the revelation. Pennebaker
(1990) suggested that to be reminded of pain may be painful initially, but
then the writing may lead to the lasting relief associated with facing a
trauma. It is important to keep in mind, however, that neither Penne-
baker's nor Foa's experiments specifically involved a comparison of high
levels of client disclosure with low levels of disclosure in therapy.

Summary
To date, there is no direct evidence that greater clients' disclosure in
therapy is linked with posttherapy benefits. At this point, the most parsi-
monious interpretation of the findings on clients' disclosure is that greater
SECRECY AND OPENNESS IN PSYCHOTHERAPY 121

disclosure is not associated with therapy benefits, even though this issue
needs far more exploration. In the future, researchers will need to tease
apart the effects of high levels of distress and high levels of disclosure on
clients' improvement rates and will need to explore the long-term effects
of clients' in-depth disclosures in therapy.

COVERT PROCESSES

Hidden reactions are those hidden thoughts and feelings that clients
have in response to therapists' interventions, and things left unsaid are
those thoughts and feelings that clients have during sessions that they do
not share with their therapists (Hill et al., 1993). They are both covert
responses to the therapy itself. Unlike hidden reactions and things left
unsaid, secrets arise over a relatively long time period and do not neces-
sarily stem from events within the therapy sessions (Hill et al., 1993).
Secrets are major life experiences, facts, or feelings that clients do not share
with their therapists (Hill et al., 1993). Because both hidden reactions and
things left unsaid are very similar processes, I present the findings pertain-
ing to them together in this chapter, followed by a separate discussion of
secrets.

Hidden Reactions and Things Left Unsaid about Therapy

Researchers have shown that clients frequently conceal their reactions
from their therapists and that the most of these hidden reactions are
negative (Hill, Thompson, & Corbett, 1992; Regan & Hill, 1992; Rennie,
1985,1992; Thompson & Hill, 1991). For example, using a qualitative meth-
odology, Rennie (1985,1992) discovered that clients acted cooperative and
amiable, even when they were secretly questioning the therapists' inter-
ventions and feeling resentful toward the therapists. The clients did not
reveal such negative reactions because they thought it was not their place
to question the experts, felt it was immature to criticize the therapists when
the therapy was generally appreciated, or were concerned that criticism
might threaten the relationship with the therapists (Rennie, 1985, 1992).
It is commonly believed that therapists need to encourage clients to
reveal private infomlation that the clients might be disinclined to reveal
(e.g., Hill et al., 1993; Stiles, 1995) and that therapists need to have greater
awareness of clients' hidden negative reactions to therapists' interventions
(e.g., Hill, Helms, Spiegel, & Tichenor, 1988a). If therapists are aware of
such hidden reactions, then presumably they can make helpful adjust-
ments to their interventions (e.g., Hill et al., 1988a; Rhodes, Hill, Thomp-
son, & Elliott, 1994). Rhodes et al. (1994) conducted a qualitative study of
122 CHAPTER 6

19 therapy cases and obtained results that were consistent with this idea.
Clients recalled an event in therapy in which there was a major misunder-
standing and described how it was or was not resolved (Rhodes et aI.,
1994). A good relationship, clients' willingness to assert negative feelings
about being misunderstood, and therapists' facilitation of a mutual repair
effort through maintaining a flexible and accepting stance were factors
that were associated with resolution. In contrast, a poor relationship,
therapists' unwillingness to discuss or accept clients' assertion of negative
reactions to being misunderstood, and therapists' lack of awareness of
clients' negative feelings were associated with unresolved misunderstand-
ings and clients' terminating therapy (Rhodes et aI., 1994).
Despite the intuitive appeal of a very open therapist-client relation-
ship, only one (Wright, Ingraham, Chemtob, & Perez-Arce, 1985) of a re-
cent string of correlational studies on covert processes in therapy (e.g., Hill
et aI., 1992, 1993; Kelly, 1998; Regan & Hill, 1992; Thompson & Hill, 1991)
produced findings that are consistent with the results from the qualitative
Rhodes et a1. (1994) study. Fifteen graduate students participated in two
experiential training groups that met for an hour and a half each week for
13 weeks as part of a course in group psychotherapy (Wright et aI., 1985).
After each group therapy class, the students rated their perceptions of the
session. The more that members of a group therapy class withheld from
other group members and the group leaders, the less satisfied they were
with the group sessions. However, the reader should be alerted to the fact
that the two first authors (i.e., Wright and Ingraham) were the two group
leaders in that study, who had hypothesized that the participants would
report less satisfaction in group meetings in which the participants held
things back. As conscientious group leaders, the authors used the weekly
feedback to try to enhance the subsequent sessions. Thus, because of the
possible demands on the students to fulfill their instructors' expectations,
these findings would need to be replicated in studies where the demand
characteristics are less prominent before researchers could conclude that
concealment in group therapy is associated with less satisfaction with the
sessions.
In another study, Regan and Hill (1992) observed that the more things
that the clients in time-limited therapy left unsaid with behavioral!
cognitive content (e.g., "I wasn't able to express all my feelings articu-
lately," p. 169), the less satisfied the clients were with the amount of
progress they had made in their sessions. However, there was a positive
relation between the proportion of things with emotional content that
clients left unsaid in therapy (e.g., "I was feeling anxious about being
videotaped and having to review the videotape at the end," p. 169) and the
clients' satisfaction with therapy and change (Regan & Hill, 1992). Further-
SECRECY AND OPENNESS IN PSYCHOTHERAPY 123

more, the therapists were aware of only 17% of the things clients left unsaid
in sessions. When the therapists could identify what clients left unsaid, the
therapists rated the sessions as being rougher, and the clients were less
satisfied with the treatment (Regan & Hill, 1992).
The findings from several investigations involving therapists' and
clients' reviews of video- or audiotapes of a just-concluded therapy ses-
sion have pointed to a clearer conclusion concerning whether therapists'
recognizing clients' hidden negative reactions is beneficial to clients (Hill
et a1., 1992, 1993; Thompson & Hill, 1991). Such recognition of negative
reactions does not seem to be helpful.
In one of the studies, Hill et a1. (1992) showed that clients in time-
limited therapy reported that they hid more negative reactions (e.g.,
scared, confused, misunderstood) than any other kind of reaction. Al-
though the therapists did guess that the clients were hiding reactions from
them, the therapists typically could not identify when the clients were
hiding their reactions or what the reactions were (Hill et a1., 1992). Like-
wise, 65% of a sample of long-term therapy clients reported that they left
something unsaid in therapy; yet only 27% of the therapists were able to
match what the clients left unsaid (Hill et a1., 1993). Perhaps most interest-
ing is the fact that both Hill et a1. (1992) and Thompson and Hill (1991)
found that when therapists recognized the hidden negative reactions of
their clients, the therapists themselves and the clients rated subsequent
therapist interventions as less helpful than when the therapists did not
recognize these reactions.
It is noteworthy that the therapists from two of these studies were
able to match the clients' supported reactions (i.e., understood, hopeful,
relief), which were not hidden, at a better-than-chance rate (Hill et a1., 1992,
1993). Furthermore, Thompson and Hill (1991) found that the therapist's
accurate perceptions of positive client reactions were related to the thera-
pist's ability to generate helpful interventions. Hence, the findings from
these studies taken together support the notion that it is fruitful for thera-
pists to recognize positive, overt reactions from their clients. However, the
findings clearly do not support the idea that therapy is enhanced by the
therapists' recognizing covert, negative reactions from their clients. As Hill
(1992) succinctly put it, "clients often hide negative reactions; and when
therapists are aware of negative reactions, there may be negative effects on
the therapy" (p. 689).
What causes therapists to be less effective after recognizing negative
client reactions? Thompson and Hill (1991) suggested that therapists' be-
coming anxious after such recognition may be somewhat debilitating to
them. Supporting this suggestion are the findings from a study involving
graduate student counselors with varying levels of experience (Russell &
124 CHAPTER 6

Snyder, 1963). Russell and Snyder (1963) showed that regardless of their
level of experience, the counselors became more anxious when interacting
with clients who acted negative and hostile than they did with clients who
acted positive and friendly. I suggest that the therapists' subsequent inef-
fectiveness may stem from their responding to the clients in ways that
reduce the therapists' own anxiety rather than in ways that help the
clients. For example, a client angrily tells her therapist that she sees him as
unsupportive and condescending for not backing her efforts at becoming a
famous actress. He may respond by saying that the sessions are a good
place for her to learn to manage her anger, and these words may make her
even angrier because she perceives them to be an invalidation of her
complaint.
Researchers in fact have shown that psychotherapists tend to avoid
hostility that is directed at themselves more than when it is directed at
others and that following therapists' avoidant reactions, clients are more
likely to drop the hostile topic or change the object of the hostility (Ban-
dura, Lipsher, & Miller, 1960). Given these findings and the ones described
in the previous paragraphs, the statement that "perhaps counselors oper-
ate more effectively if they operate under a slight delusion that clients are
reacting positively to them" (Regan & Hill, 1992, p. 173) seems to be an
appropriate conclusion.

Secrets
As mentioned earlier, Hill et al. (1993) found that about half (46%) a
sample of individual therapy clients who had received an average of 86
sessions reported that they were keeping secrets from their therapists. The
length of time that the clients had spent in therapy was not related to the
number of secrets they were keeping (Hill et al., 1993). Because so many of
the clients reported keeping secrets from their therapists, Hill et al. (1993)
recommended that "therapists might attend to methods to enable clients
to feel more comfortable and less embarrassed about revealing secrets" (p.
285). However, secret keeping was not significantly related to the clients'
satisfaction with therapy; and symptomatology was not assessed, so there
was no evidence that clients who kept secrets actually had greater symp-
tomatology.
In another study mentioned in previous chapters, I (Kelly, 1998) as-
sessed the relation between secret keeping in therapy and symptom reduc-
tion, using a sample of 42 clients who had received an average of 11
therapy sessions. Much like the clients in the Hill et al. (1993) study, just
over 40% of the clients reported keeping relevant secrets (i.e., ones that
they perceived to be related to their presenting problems) in therapy. Most
SECRECY AND OPENNESS IN PSYCHOTHERAPY 125

important is the finding that, after 1 statistically adjusted for clients' social
desirability and self-concealment scores, keeping relevant secrets in ther-
apy was associated with a reduction in the clients' reports of symptoms
since the intake. (I adjusted for self-concealment scores to see whether
keeping a particular secret from the therapist was associated with a reduc-
tion in symptoms above and beyond the effects of the client's general
tendency to keep secrets.) This result supports the idea that clients who in
practice do conceal some unfavorable aspects of themselves from their
therapists may benefit more from therapy than those who do not (Kelly,
1998). 1 suggest that it is possible that these benefits occur because the
clients who do keep secrets may have an easier time imagining that their
therapists see them favorably (Kelly, 1998, 2000a,b). See the next chapter
for an elaboration of this point.

SUMMARY

On the whole, the research on clients' disclosure and covert processes

in therapy does not offer much support to the notion that a high degree of
clients' openness is therapeutic. In a study of schizophrenics, researchers
showed that more disclosure in group therapy was associated with less
improvement on the MMPI and on measures of intelligence and self-
reported symptomatology (Strassberg et al., 1975). In another study, more
disclosure either was not associated with symptom reduction or was
associated with an increase in depression (see Stiles & Shapiro, 1994). In
addition, higher levels of clients' disclosure have been associated with
negative therapy process ratings (Stiles, 1984). Likewise, even though in
some case studies, therapists' lack of awareness of clients' negative feel-
ings have been problematic (e.g., Rhodes et al., 1994), researchers conduct-
ing correlational studies generally have demonstrated that therapists'
awareness of clients' negative covert processes is associated with negative
therapy ratings from both the clients and therapists (e.g., Hill et al., 1992;
Regan & Hill, 1992; Thompson & Hill, 1991). Moreover, 1 (Kelly, 1998)
recently showed that clients' keeping relevant secrets from their therapists
was associated with a reduction in symptoms, after the analyses adjusted
for clients' social desirability and general self-concealment scores. Hence,
the best conclusion that can be drawn from all this research is the qualified
statement that higher levels of clients' disclosure are not helpful in therapy.
The statement is qualified by the fact that there are gaps in the literature
concerning whether (1) a greater depth of clients' disclosure is associated
with therapy benefits and (2) there are long-term benefits associated with
more disclosure in therapy. The statement also is qualified by the fact that
in a number of these studies, researchers primarily used clients' self-
126 CHAPTER 6

reports to measure outcome. Even though it seems reasonable to ask

clients about their perceptions of themselves if the aim of therapy is to
promote positive self-concept change, it is still possible that clients either
deliberately or inadvertently misrepresent their views of themselves on
questionnaires.

CONCLUSION

Many theorists from various approaches to psychotherapy believe

that high levels of revelation from clients, particularly carefully timed
revelations, are important for therapeutic progress. This emphasis on very
high levels of openness in therapy has its origins in Freudian psycho-
analysis and may have been perpetuated by the humanistic/encounter
group movement in the 1960s.
However, it is far from clear from the research on clients' disclosure
and covert process that high levels of clients' openness are beneficial to
clients (e.g., Hill et al., 1993; Kelly, 1998). There are even some occasions
when the benefits of revealing secrets in therapy might outweigh the costs.
For example, some case studies have shown that clients' families dissolved
after the clients revealed a family secret in treatment (e.g., Saffer et al.,
1979). Moreover, confidentiality is not a guaranteed part of treatment,
particularly in family and group therapy contexts (see Davis & Meara,
1982), and clients may be wise to protect their self-interests of privacy (e.g.,
in cases of pending divorce and child custody battles) (McCarthy, 1995).
In addition, the revelation of shameful secrets may cause some clients to
flee therapy prematurely before deriving many therapeutic benefits (e.g.,
Kaslow, 1993). Again, though, these too are case studies, which offer
limited generalizability, so it is unclear when and which clients should
hold back from revealing negative material in therapy. Also, there have
been no follow-up studies comparing cases in which clients did or did not
disclose important secrets in therapy. It is possible, for example, that the
clients whose families dissolve because of the revelation of a secret are
better off than clients who live under the shroud of secrecy.
Perhaps future researchers could examine the characteristics of clients
to see which clients fare best when they are very open with their therapists.
One example of work that already has been done is the study by Amini et
al. (1978), which demonstrated the negative effects of too much probing for
psychiatric inpatient adolescents who may need more privacy than adults.
Another promising line of research is that on counseling men (e.g., Good,
Dell, & Mintz, 1989; Robertson & Fitzgerald, 1992). These researchers have
gathered evidence that a task-oriented, as opposed to a more open, expres-
SECRECY AND OPENNESS IN PSYCHOTHERAPY 127

sive, and feeling-oriented, approach to treatment could be more efficacious

with individuals with a masculine gender role orientation.
In closing, I suggest that psychotherapy may create difficult situations
for clients who are inclined to keep information private. The clients are
likely to perceive that their therapists want them to divulge completely, yet
hiding some negative aspects from the therapist may be healthier for them
(Kelly, 1998). Specifically, it may allow them to maintain their own sense of
personal boundaries and privacy (Hoyt, 1978; Margolis, 1966, 1974). Per-
haps therapists who encourage clients to make careful decisions about
what to reveal and what not to reveal in therapy are providing an oppor-
tunity for the clients to practice judicious revelation that could be invalu-
able to them in dealing with people outside of therapy.
CHAPTER 7

WHY OPENNESS MAY

NOT BE THERAPEUTIC
A SELF-PRESENTATIONAL
VIEW OF PSYCHOTHERAPY

As outlined in the previous chapter, psychotherapy researchers have dem-

onstrated not only that clients withhold personal information and reac-
tions from their therapists but also that such discretion is associated with
positive therapy process ratings and outcomes. The question I address in
this chapter is, how can these puzzling findings be explained?
One obvious reason for the clients' concealment is that clients want to
avoid making a bad impression on their therapists and might feel too
ashamed or embarrassed to reveal their secrets (see Hill et al., 1993; Kelly,
1998). If the clients were to tell their therapists abhorrent things about
themselves, the clients would run the risk of constructing unwanted im-
ages of themselves. For example, recall the case of a New York City woman
who abused her 6-year-old daughter Elisa Izqueido, to the point of killing
the child in 1997. It is reported that on one occasion, when Elisa had gone to
the bathroom on the floor, the mother had turned her upside down to use
her hair to mop up the excrement. If the mother were to tell her therapist
about such an event, she may come to imagine, or accurately perceive, that
the therapist sees her as a horrifyingly abusive, self-centered (or antisocial)
person who is in need of deep-seated personality change. In everyday
interactions, people normally refrain from telling others highly objection-
able information about themselves because of fears that they will receive
negative evaluations from others (see Kelly & McKillop, 1996). Presum-

U9
130 CHAPTER 7

ably, in therapy, clients have the chance to reveal personal information

without the fear of receiving negative judgments about themselves as
people (e.g., Rogers, 1951, 1957). However, I contend that the same con-
cerns that people have about looking bad in front of others when they say
unfavorable things about themselves are likely to apply to interactions
with a therapist. The concerns may even be accentuated because of the
therapist's being an expert, knowledgeable audience (see Kelly, McKillop,
& Neimeyer, 1991; McKillop & Schlenker, 1988).
After all, therapists are trained to diagnose their clients' pathologies,
and they often make broad-based diagnoses such as labeling the clients as
having narcissistic, borderline, or psychotic tendencies (see American Psy-
chiatric Association, 1994). These diagnoses may not seem evaluative to
the therapists, because they are trying to help their clients, but the clients
would probably find such diagnoses to be very negative and demoraliz-
ing. In their study of brief therapy, Regan and Hill (1992) showed that
therapists did indeed form diagnoses or explanations for their clients'
behaviors that were very negative (as rated by judges) and hid these
conjectures from the clients. Yet ironically, therapists secretly may deem
the clients who express worries about being judged to have paranoid
tendencies.
When one considers the fact that therapists form negative opinions
about their clients, the findings that clients frequently hide negative mate-
rial from their therapists and that high levels of clients' openness are not
linked to greater therapy benefits (e.g., Hill et al., 1992, 1993; Kelly, 1998;
Regan & Hill, 1992) begin to make sense. I suggest that the findings can be
explained by conceptualizing psychotherapy as a self-presentational pro-
cess. Specifically, clients may come to benefit from therapy by perceiving
that their therapists have favorable views of them. Part of creating these
favorable impressions can involve clients' hiding some undesirable as-
pects of themselves from their therapists, because if the clients are too re-
vealing, they might perceive that their therapists view them unfavorably.
The components of this self-presentational perspective are as follows:
(1) Clients present themselves in various ways to their therapists, (2) thera-
pists then offer feedback to their clients based on those self-presentations,
(3) this feedback can lead to the clients' altering their self-beliefs to be in
line with the feedback, particularly because the feedback is from an expert
audience, and (4) this shifting of self-beliefs followed by similar self-
presentations and feedback may eventually lead to changes in the clients'
self-concepts (Le., in their relatively stable collections of self-beliefs) (Kelly,
2000a). At the heart of this perspective is the suggestion that clients may
benefit from therapy through presenting themselves to their therapists in a
manner that is believable to the clients and consistent with how they wish
WHY OPENNESS MAY NOT BE THERAPEUTIC 131

to be seen by the therapists (e.g., as decent people who are courageous

enough to deal with their problems). The clients may then come to see
themselves in these desirable ways by getting the impression that their
therapists do see the clients in these ways.
More precisely, self-concept change should be enhanced when the
clients find their self-presentations to be believable and when they think
that their therapists find them to be believable. It really should not matter
for clients' self-concept change whether the therapists do in fact find the
clients' self-presentations to be believable (Kelly, 2000b). Nevertheless,
there is still a good, separate argument for the suggestion that clients' self-
presentations should be believable to their therapists. This argument
comes from the evidence that extremely positively biased self-perceptions
are not associated with favorable mental health (see Taylor & Brown, 1994).
Having therapists find the self-presentations believable is likely to provide
some indication that the clients' self-presentations are not too extreme or mal-
adaptive, even if the therapists' actually believing the self-presentations
may not be crucial for clients' self-concept change (see Kelly, 2000b).
In formulating this theoretical perspective, I have drawn from self-
identification theory (Schlenker, 1980, 1986; Schlenker & Weigold, 1989,
1992; Schlenker, Britt, & Pennington, 1996) and from social psychological
research on self-concept change. In this research, people have been in-
duced to describe themselves in particular ways, such as extraverted or
introverted, and then have reported changes in their self-beliefs to be
consistent with their self-presentations (e.g., Fazio, Effrein, & Falender,
1981; Gergen, 1965; Jones, Rhodewalt, Berglas, & Skelton, 1981; Kelly et al.,
1991; Kulik, Sledge, & Mahler, 1986; Rhodewalt & Agustsdottir, 1986;
Schlenker, Dlugolecki, & Doherty, 1994; Schlenker & Trudeau, 1990; Tice,
1992).
There are two major sections of this chapter. In the first major section,
I lay the empirical and theoretical foundation for the proposed view of
psychotherapy. I describe the psychotherapy research findings that clients
do try, either consciously or unconsciously, to manage the impressions that
therapists have of them. This research is included because in order to
embrace the notion that clients may benefit from presenting themselves in
believable and desirable ways, one needs to see the evidence that clients
do vary their self-presentations in therapy, depending on their goals for
the interactions. I then present the social psychological research on self-
concept change to illustrate how the clients' impression-management at-
tempts can be a part of healthy client self-concept change. In the second
major section, I indicate how the proposed self-presentational view of
psychotherapy can explain previously confusing findings in the psycho-
therapy literature. In addition, I provide guidelines concerning (1) what
132 CHAPTER 7

might be optimal levels of self-revelation from clients and (2) how thera-
pists might appropriately respond to these revelations. I conclude by
addressing some alternative points of view and limitations of this theoreti-
cal perspective, as well as by offering new research directions.

SELF-PRESENTATION RESEARCH
CLIENTS'SELF-PRESENTATIONS
I take a broad view of self-presentation, defining it as showing oneself
to be a particular kind of person for various audiences (See Schlenker,
1986). The types of self-presentations people perform are influenced by
their goals for interacting with others, and the subsequent real or imagined
feedback from others is thought to be an integral part of how people come
to view themselves (Kelly et al., 1991; Schlenker, 1980, 1986, 1987; Schlenker
et al., 1996; Schlenker & Weigold, 1992). For example, an adolescent client
who has been physically abused by her father and who also has been
engaging in shoplifting may discuss the abuse but may not tell her thera-
pist about the shoplifting. Her private goal is to have her therapist see her
as the survivor of abuse, not as a criminal. Researchers have shown in fact
that counselors' judgments of clients are affected by the clients' initial self-
presentations (Schwartz, Friedlander, & Tedeschi, 1986) and that such
initial impressions are stable and may influence the course of therapy (see
Wills, 1978).
This view of self-presentation contrasts with the more narrow tradi-
tional view of self-presentation as a way of strategically describing oneself
to others in order to get one's needs met by them Oones & Pittman, 1982). It
also contrasts with the way the term has been used in the psychotherapy
literature, where clients' self-presentations have been construed as coun-
terproductive efforts to manipulate or influence their counselors (e.g.,
Friedlander & Schwartz, 1985; Haley, 1963; Strong, 1968). In that literature,
therapists have been advised to offset such attempts because it is believed
that if they are manipulated or controlled by their clients' self-presentations,
then they will not be able to help the clients (Friedlander & Schwartz, 1985;
Haley, 1963; Kiesler, 1981; Strong & Claiborn, 1982; Strong, 1987).
When self-presentation is viewed in this broad way, the implications
are different for psychotherapy than would be the case with the more
traditional view of self-presentation as a form of deliberate manipulation
(see Friedlander & Schwartz, 1985, for a discussion of the latter). According
to this broad view, clients cannot avoid trying to construct desirable images
of themselves in the presence of their therapists; it happens automatically,
much like communication itself (see Schlenker & Weigold, 1989). Clients
may invest different levels of effort in constructing their desirable identi-
WHY OPENNESS MAY NOT BE THERAPEUTIC 133

ties, may have varying degrees of awareness of their self-presentational

motives, and may have additional motives for why they fail to disclose
information to their therapists. Nonetheless, self-presentational motives
are likely always to be present in the counseling relationship (see SchInker
& Weigold, 1989). For example, a man who is being treated for his shyness
with women may be reluctant to bring up details about his having grown
up in fear of his mother's cruel disciplinary tacts. He may fear that the
therapist will weigh those details too heavily in developing a treatment
plan (a non-self-presentational concern). At the same time, he may uncon-
sciously wish to avoid having the therapist view him as weak or pathetic (a
self-presentational concern).
In general, people put more effort into creating desirable impressions
when the audience is significant (i.e., expert, powerful, attractive) and
psychologically close to them (see Leary & Kowalski, 1990; Nowak,
Szamrej, & Latane, 1990). People perceive clinical and counseling psychol-
ogists to be more knowledgeable about human behavior, expert, intel-
ligent, caring, supportive, and helpful than their own peers (Kelly et al.,
1991; McKillop & Schlenker, 1988). Therefore, it is not surprising that
researchers have demonstrated in several in-the-field experiments that
clients do try, either consciously or unconsciously, to manage the impres-
sions that their therapists have of them (Braginsky & Braginsky, 1967;
Braginsky, Grosse, & Ring, 1966; Kelly et al., 1996; McArdle, 1974).
For example, in early experiment with psychiatric inpatients, the
instructions to a mental health assessment were manipulated so that a
group of "short-timers" and a group of "old-timers" (Le., patients who
were most and least motivated to be discharged, respectively) believed
that their mental hospitalization status would potentially be determined
by the test, whereas a control group of old-timers received no such instruc-
tions (Braginsky et al., 1966). As it turned out, the experimental group of
old-timers presented themselves as mentally ill and unprepared for dis-
charge, whereas the short-timers presented themselves as self-insightful
and ready for discharge. The control group of old-timers did not present
themselves as poorly as the other old-timers nor as well as the short-
timers.
In a similar experiment, hospitalized chronic schizophrenics were
randomly assigned to groups who were led to believe that either their
continued eligibility for open-ward status or their appropriateness for
continued hospitalization would be determined by an interview (Bra gin-
sky & Braginsky, 1967). During the interview, the patients who believed
that their open-ward status was in question presented themselves as
healthy, whereas the patients who believed that their hospitalization status
was in question presented themselves as mentally ill. Moreover, these self-
presentations were convincing to a group of experienced psychiatrists
134 CHAPTER 7

(Braginsky & Braginsky, 1967). The findings from this study and the Bra-
ginsky et al. (1966) study are relevant to the proposed view of psycho-
therapy, because they demonstrate that even patients with severe psychi-
atric disorders can and do vary how they present themselves to their
therapists, depending on their goals for the interactions.
More recently, my colleagues and I conducted an in-the-field experi-
ment in which we asked 92 therapy outpatients to complete intake forms
that would be reviewed by either (1) their counselor, (2) their counselor
who requested that they reveal their innermost thoughts in completing the
forms, or (3) researchers who would not know who completed the forms
(Kelly et al., 1996, Exp. 2). We found no significant differences in the
outpatients' reports of well-being across the three conditions. Specifically,
the outpatients in all three groups reported on average that they were
depressed and had low self-esteem. However, the outpatients in the coun-
selor conditions had higher social desirability scores than did those in the
anonymous condition, suggesting that the clients attempted to look like
good people to their counselors (Kelly et al., 1996, Exp. 2).
Moreover, at the end of that experiment, we asked the outpatients in
all three conditions to rate anonymously how representative of themselves
their self-descriptions had been. Interestingly enough, those who had
described themselves as depressed tended to rate those self-descriptions
as less representative of themselves than did the outpatients who had
described themselves as nondepressed (Kelly et al., 1996, Exp. 2). Based on
this finding, I (Kelly, 2000a) suggested that even when clients present
themselves in a way that could be considered appropriate for the client
role (i.e., as depressed individuals), they still are motivated to see those
self-presentations as being unrepresentative of themselves. Hence, the
findings support the idea that clients are quite similar to the general
population in that people predominantly prefer to construct beneficial
self-images and avoid detrimental ones (se Baumeister, 1982; Jones, 1990;
Schlenker & Weigold, 1992; Tesser, 1988). A restriction to this claim is that
a couple of the clients in that sample reported very high levels of depres-
sion and very low levels of self-esteem, and they rated these presentations
as highly representative of themselves. Thus, there may be some clients
who have such firmly entrenched depreSSive self-views that they would
view feeling good about themselves as atypical, and those clients might be
most resistant to change (see Swann, 1996).

SELF-CONCEPT CHANGE

What makes the fact that clients vary how they present themselves to
their counselors so critical is that these self-presentations are likely to play
WHY OPENNESS MAY NOT BE THERAPEUTIC 135

a pivotal role in the way that the clients come to see themselves (Kelly,
1968). For example, imagine a client who sometimes experiences intense
feelings of grief, anger, and hopelessness when thinking about his ex-
wife's taking sole custody of their children and who at other times feels
that he is recovering from these negative emotions. He may not be sure
whether he is making progress in therapy. During his therapy sessions, he
may emphasize the periods of relief and tell his therapist that he thinks
that he is making good progress. Theoretically, the client should come to
see himself as less grief-stricken and angry in part because he comes to
perceive that the therapist sees him as making progress (see Quintana &
Meara, 1990; Schlenker, 1986; Schlenker & Trudeau, 1990; Schlenker &
Weigold,1989).
Clients who incorporate desirable images into their self-beliefs should
become even more likely to describe themselves in these favorable ways in
the future. Such self-presentations followed by additional desirable feed-
back from the therapist-and additional internalization of that feedback-
may ultimately lead to changes in the clients' self-concepts (see Schlenker,
1986; Schlenker et al., 1994). It is through this process of self-concept change
that clients may come to benefit from therapy (see Kelly, 2000a). This idea
is consistent with research that suggests that positive views of oneself
promote psychological well-being, the ability to care about others, and the
ability to engage in productive work (Taylor & Armor, 1996; Taylor &
Brown, 1988, 1994; Taylor & Gollwitzer, 1995).
This characterization of how clients may experience self-concept
change is compatible with the notion that the self-concept has some flex-
ibility, as it has been depicted in the social psychological literature. Bau-
meister (1998) noted that although a person can have only one self, he or
she can have multiple self-concepts across time. The self-concept may be
seen as having a relatively solid nucleus of strong self-beliefs with a
permeable periphery of weaker, more situationally dependent self-beliefs
(Rosenberg, 1979; Schlenker, 1985, 1986; Schlenker & Trudeau, 1990). These
flexible boundaries of the self-concept account for the idea that people can
choose from a range of self-presentations that all would be considered
representative of themselves (Schlenker, 1986; Schlenker et al., 1994;
Schlenker & Trudeau, 1990).
Psychologists have long theorized that people form their self-concepts
in part through presenting particular self-images to others and then receiv-
ing reactions to these images (e.g., Baldwin, 1992; Cooley, 1902; James, 1890;
Mead, 1934; Goffman, 1959; Schlenker, 1980, 1986; Schlenker & Weigold,
1992). Whereas there is a fair amount of evidence that self-concepts are
relatively stable and resistant to change (e.g., Maracek & Mettee, 1972;
McFarlin & Blascovich, 1981; Sullivan, 1953; Swann, 1987, 1996; Swann &
136 CHAPTER 7

Ely, 1984; Swan & Hill, 1982; Swann & Predmore, 1985; Swann & Read,
1981), there also is evidence that following their self-presentations, people
do shift even strong self-beliefs (i.e., ones that people perceive themselves
to hold consistently) in the direction of the self-presentations (e.g., Schlen-
ker & Trudeau, 1990). In addition, research supports the idea that such
temporary shifts in people's self-beliefs can affect their behaviors, which
then can have an impact on their self-concepts (e.g., Fazio et al., 1981;
Schlenker et al., 1994).
This process whereby people incorporate aspects of their self-presen-
tations into their own identities often is referred to within the social
psychological literature as internalization (e.g., Tice, 1992). Its meaning is
similar to the way the same term has been used in the counseling literature
to describe clients' incorporating feedback from others into their own self-
images (e.g., Quintana & Meara, 1990). I review the research on this phe-
nomenon next and discuss how the findings help explain how self-concept
change may take place in clients.

Effects of Self-Presentations on Self-Beliefs

In a well-known series of experiments, Jones et al. (1981) studied the
effects of strategic self-enhancement or self-deprecation on subsequent
ratings of self-esteem. In the first two experiments, participants were in-
duced to present themselves in either a self-enhancing or self-deprecating
manner during an interview. The participants in the self-enhancing condi-
tion later showed increased self-esteem, whereas the participants in the
self-deprecating condition showed decreased self-esteem. Likewise, in the
third experiment, participants who played a self-enhancing role for the
interview subsequently showed an increase in their self-esteem. However,
this shift only occurred for participants who had the freedom to respond
in-role to interview questions. It did not occur for participants who gave
preplanned, in-role answers. In contrast, participants who played the self-
deprecating role during the interview subsequently showed lowered self-
esteem only in the condition in which they had been given a clear choice
concerning whether to engage in the interview. Schlenker (1986) suggested
that "participants in the Jones et al. (1981) study may simply have shifted
their self-feelings in the direction of their behavior whenever the behavior
seemed to be representative of the self" (p. 39). Schlenker proposed that
the believability of one's self-presentations in general enhances the inter-
nalization of those presentations.
Applying this idea to counseling, I content that the likelihood that
the client will come to internalize the therapist's feedback is enhanced
when a client performs self-presentations that are within a range of self-
WHY OPENNESS MAY NOT BE THERAPEUTIC 137

presentations that the client finds to be believable. (It would probably help
if the therapist also could find the client's self-presentations to be believ-
able; what really matters, though, is that the client is able to believe the
therapist's feedback.) Believability is likely to be important because if the
client tells himself that he lied to the therapist and that the therapist would
not view him in a desirable way if the therapist knew more accurate
information about him, then the client should not be able to place much
weight on what he perceives as the therapist's uninformed opinions of him
(see Schlenker, 1986; Swann, 1996). By contrast, if the client feels that his
self-presentations have generally been honest and accurate (despite his
having kept some secrets), then he can internalize what he sees as the
therapist's informed and favorable opinions of him (see Schlenker, 1986).
Another experiment showed that self-presentations affect subsequent
self-appraisals, behaviors, and recall of events that are relevant to the self-
presentations (Schlenker et al., 1994, Exp. 1). Participants were either in-
structed to present themselves as sociable during an interview or were
given information about the importance of sociability but were not inter-
viewed. All the participants then waited in a room with a confederate who
later judged how socially each participant behaved (e.g., whether the
subject initiated conversation, how much he or she spoke). For the final
portion of the experiment, participants assessed their own sociability and
a recall measure concerning experiences that had had (outside the labora-
tory) that were relevant to sociability. As it turned out, the participants
who presented themselves as sociable later behaved more sociably, rated
themselves as more sociable, and recalled more past experiences in which
they had behaved sociably than did the participants who were not inter-
viewed. Another noteworthy finding was that no differences emerged for
assessments of features other than SOciability, such as self-esteem, intel-
ligence, leadership abilities, or affective states. Therefore, an important
contribution of this study was that it demonstrated not only that self-
presentations have an impact on self-evaluations and behaviors but also
that these effects specifically correspond to the self-images that are por-
trayed in the self-presentations. The effects are not the result of a mere
shift in affect or self-esteem (Schlenker et al., 1994).
In yet another experiment, participants were asked biased questions
that elicited either introverted or extraverted responses (e.g., "What things
do you dislike about loud parties?"; Fazio et al., 1981). After the interview,
participants interacted with a female confederate who would later make
judgments concerning how introverted or extraverted each participant
appeared (e.g., how close he or she sat to the confederate, whether he or
she initiated conversation). For the final portion of the experiment, partici-
pants rated how introverted or extraverted they perceived themselves to
138 CHAPTER 7

be. Results from both their own self-ratings and from the confederate's
behavioral ratings of them supported the idea that participants inter-
nalized their earlier self-presentations. The authors interpreted their find-
ings in terms of self-perception (Bem, 1972) and self-fulfilling prophecy
(Merton, 1948). Specifically, they indicated that once an individual behaves
in a particular manner, that person looks to her behaviors to determine her
internal state and internalizes the very traits that the audience expected
her to possess (Fazio et al., 1981). The authors concluded that such a change
in her self-concept is apt to affect her behavior in future and different
situations that do not even involve the original audience.

Audience Effects
Not only do one's self-presentations to various audiences influence
one's self-beliefs but also the feedback from those audiences seems to play
a central role in such shifts (Gergen, 1965). In one early study, participants
were interviewed under the pretext that they would be helping to train
interviewers who would later conduct large-scale personality surveys
(Gergen, 1965). The participants were instructed either to present them-
selves accurately or to try to make a good impression. During the inter-
view, they either received reflective reinforcement (e.g., "very good," "yes,
I would agree") or did not receive reinforcement. After the interviews,
participants in the reflective-reinforcement condition described them-
selves significantly more positively following their self-presentations than
did participants in the no-feedback condition. Also, participants who had
been instructed to try to make a good impression during the interview
became even more positive in their self-descriptions following the inter-
view than those who had been instructed to be accurate. Gergen (1965)
concluded that these results support the contention that affirming social
feedback is effective in elevating people's self-evaluations.
Even the mere presence of an audience can augment the internaliza-
tion of one's self-presentations (Tice, 1992). In two studies, Tice (1992)
demonstrated that participants who were induced to describe themselves
in particular ways (Le., as emotionally stable or emotionally responsive in
Study 1 and as introverted or extraverted in Study 2) for an audience
tended to bring their self-beliefs more in line with their self-descriptions
than did participants who described themselves anonymously. This shift
in self-beliefs extended to changes in the participants' behaviors. More-
over, these behavioral changes occurred even when participants were
unaware that they were being observed (Tice, 1992, Study 2). In Study 3,
participants were asked to describe themselves as introverted to a clinical
psychology graduate student whom they would meet or whom they
WHY OPENNESS MAY NOT BE THERAPEUTIC 139

would not meet. In addition, they were instructed to draw on their own
past experiences (self-reference group) or to read from a script of re-
sponses that was generated by the self-reference group (yoked group) in
their portrayals of themselves as introverted. As it turned out, the partici-
pants who expected future interaction with the clinical graduate student
internalized their self-presentations more than those who did not. More-
over, the participants who were in the self-reference group, as compared
with those in the yoked group, experienced internalization to a greater
extent, and the combined effect of these two manipulations on internaliza-
tion was larger than either effect alone (Tice, 1992).
Researchers have demonstrated that the type of audience (i.e., clinical
psychologists vs. untrained peers) has a differential impact on the internal-
ization of self-presentations (McKillop & Schlenker, 1988). Participants
were asked to present themselves in either a depressed or nondepressed
manner for a videotaped interview and were informed after the interview
that the audience who would be viewing their videotapes would either be
clinical psychologists or general psychology students (McKillop & Schlen-
ker, 1988). The participants who thought that clinicians would view their
sessions became more depressed following a depressed presentation and
less depressed following a nondepressed presentations. However, these
shifts did not occur in the conditions in which participants thought that the
untrained peers would view their session. The authors explained the fact
that the clinical psychologist audience had a greater impact on partici-
pants' internalization than did the peer audience by saying that "the
expert audience caused participants to focus their attention more on their
prior self-presentation which, in tum, enhanced its impact" (p. 7).
The idea that audiences playa key role in the construction of the self-
concept has particular relevance for counseling, which can be concep-
tualized as a process in which therapists use their role as an expert on
human behavior to induce change in their clients (Strong, 1968). In one
study, clients did seem to come to internalize the positive dispositions that
they perceived that their counselors held toward them (Quintana &
Meara, 1990). Forty-eight short-term therapy clients rated their counselors'
dispositions toward them (e.g., "he/she shows understanding of my
views and has empathy for me"), their dispositions toward the counselors
(e.g., "I freely, openly disclose my inner self when he/she is listening"),
and their dispositions toward themselves (e.g., "I tell myself things to
make me feel bad, guilty, or unworthy") both after their first and last
counseling sessions. After the initial session, clients' ratings of their dispo-
sitions toward themselves were quite different from (i.e., more negative
than) their ratings of the counselors' dispositions toward them. But by the
end of their treatment, clients shifted their self-views in the direction of
140 CHAPTER 7

their therapists' favorable dispositions toward them, rating their disposi-

tions toward themselves and their counselors' dispositions toward them to
be very similar. Quintana and Meara (1990) suggested that when clients
seek therapy, they may be looking for an interpersonal relationship that
will provide what they are not providing for themselves (e.g., self-
affirmation and self-empathy). They concluded that at termination, the
clients provided for themselves some of the positive actions and attitudes
that counselors had been providing for them. The findings from this in-
the-field study are especially important because most of the internalization
research has been conducted in the laboratory with undergraduates. These
findings support the possibility that the internalization processes observed
in the laboratory may indeed generalize to therapy contexts.

SUMMARY

I have reviewed seven sets of findings on self-presentation to help

provide insights into how self-concept change may take place in clients.
First, psychotherapy researchers have shown that patients vary how
they present themselves to their therapists according to their goals for
the interaction (e.g., Braginsky & Braginsky, 1967; Braginsky et al., 1966;
McArdle, 1974). They present themselves as good people to their coun-
selors at intake (Kelly et al., 1996, Exp. 2), and even when they appro-
priately (given their role as clients) present themselves as depressed, they
still view those presentations as relatively unrepresentative of themselves
(Kelly et al., 1996, Exp. 2). Second, it has been demonstrated in the labora-
tory that self-presentations have effects on subsequent self-appraisals,
such that participants change their self-beliefs to be in line with their self-
presentations (Schlenker et al., 1994; Fazio et al., 1981; Gergen, 1965; Kellyet
al., 1991; Jones et al., 1981; McKillop & Schlenker, 1988; Tice, 1992). Third,
these internalization effects have been shown to be enhanced when partic-
ipants see their self-presentations as believable (see Jones et al., 1981;
Schlenker & Trudeau, 1990). Fourth, these effects extended to subsequent
behaviors with a new audience (Schlenker et al., 1994; Tice, 1992) and to
recall of events relevant to the self-presentations (Schlenker et al., 1994;
Fazio et al., 1981). Fifth, these internalization effects are specific to the types
of presentations performed and do not seem to be the mere result of shifts
in affect (Schlenker et al., 1994; Tice, 1992). Sixth, the presence of different
types of audiences (e.g., clinical psychology graduate student vs. a private
audience) has been shown to have different effects on the internalization of
people's self-presentations (McKillop & Schlenker, 1988; Tice, 1992). Fi-
nally, clients do seem to come to internalize the positive dispositions that
they perceive that their counselors hold toward them (Quintana & Meara,
WHY OPENNESS MAY NOT BE THERAPEUTIC 141

1990). These findings taken together support my contention that clients

may experience desirable changes in their self-concepts by presenting
themselves to their therapists in a way that is both believable and consis-
tent with how the clients wish to be seen by the therapists, as specified by
the proposed self-presentational view of psychotherapy (Kelly, 2000a).
This contention must be seen as tentative, however, because the experi-
ments (i.e., involving random assignment to experimental and control
groups) in which the internalization of one's self-presentations have been
observed have not yet been tried in psychotherapy contexts.

SELF-PRESENTATIONAL VIEW OF CLIENT CHANGE

Although the findings that high levels of openness in therapy are

associated with negative therapy process and outcome ratings are not
consistent with popular mainstream theories, such as the psychoanalytic
and client-centered approaches, they are consistent with the proposed self-
presentational perspective. In particular, even though the therapy setting
often is intended as a haven for extensive revelation (e.g., Freud, 1958;
Fromm-Reichmann, 1950; Lodge, 1995; Rogers, 1951, 1957), many clients
in practice do keep secrets from their therapists (Hill et al., 1993; Kelly,
1998). This practice may actually enable them to construct more desirable
images of themselves before the credible, knowledgeable therapist audi-
ence (see Kelly, 1998). Likewise, therapists in practice do hide their nega-
tive clinical conjectures from their clients (Regan & Hill, 1992). Therapists
have been rated as more effective when their clients do not know what are
the intentions behind their therapists' interventions (Hill et al., 1993; Mar-
tin, Martin, Meyer, & SIemon, 1986; Martin, Martin, & SIemon, 1987). My
contention is that therapists may be especially effective when instead of re-
vealing their intentions, they spend more time reflecting desirable images
of their clients and showing them that they are "on the client's side" (see
Elliott, 1985).
Some may think that I am advocating deliberate manipulation at-
tempts in therapy. However, what I am doing is describing what goes on
in therapy: some censorship on the part of both clients and therapists. I am
suggesting that these practices are not necessarily harmful and they may
even be productive. In this section of the chapter, I identify the recent
theoretical trends in the psychotherapy literature as a backdrop for under-
standing what the proposed self-presentational view of psychotherapy
contributes to the psychotherapy literature. I then describe the self-
presentational view and demonstrate how this view can explain previ-
ously counterintuitive findings from the psychotherapy literature. I offer
142 CHAPTER 7

several practical suggestions for how therapists and clients might interact
and address some limitations and alternative perspectives to this view.

COMMON FACTORS IN PSYCHOTHERAPY

Psychotherapists today are faced with many choices for how to pro-
ceed with their practice because there are roughly 400 or more variants of
psychotherapy available (Kazdin, 1986). There has been a recent trend for
psychotherapists to use a mixture of approaches, with more therapists'
identifying themselves as eclectic than any other orientation (Lambert &
Bergin, 1994). This trend may have been spawned by meta-analytic studies
that showed that therapy is more beneficial than no treatment and that no
one approach is clearly superior to another (Lambert, Shapiro, & Bergin,
1986; Landman & Dawes, 1982; Smith, Glass, & Miller, 1980).
Researchers have moved away from trying to sort out which of these
many therapies are the best, and some have tried instead to assess the
"common factors" that may underlie the effectiveness of all types of
therapies. These factors include the creation of hope, opportunity for ven-
tilating feelings, interpretation of one's problems, support, advice, experi-
menting with new behaviors, and modification of cognition (see Frank,
1973; Garfield, 1980, 1991). In addition, some researchers have proposed
that instead of thinking about therapy as an application of techniques to
clients' problems, therapeutic processes may be best understood as recip-
rocal interactions between the therapist and client (e.g., Strong, 1995; Wein-
berger, 1995).
A meta-analysis of 24 studies of the therapeutic alliance, or working
bond between clients and their therapists, showed that the quality of the
alliance was a good predictor of clients' improvement, especially when the
clients rated both the bond and the outcome of their therapy (Horvath &
Symonds, 1991). Likewise, a review of what psychotherapy researchers
across 50 publications believed to be effective elements of therapy showed
the most frequently proposed factor (56% of all authors) was the develop-
ment of the therapeutic alliance (Grencavage & Norcross, 1990). Similarly,
in their review of the effects of a variety of therapist variables on counsel-
ing outcome, most of which showed no consistent relations to outcomes,
Beutler, Machado, and Neufeldt (1994) concluded that "a warm and sup-
portive therapeutic relationship facilitates therapeutic success" (p. 259).
Moreover, in their meta-analysis of paradoxical interventions, Shoham-
Salomon and Rosenthal (1987) found that interventions such as symptom
prescription were significantly more effective when the therapist used a
positive connotation than a negative connotation. For example, a therapist
who praises her depressed client for his tolerance of solitude and willing-
ness to sacrifice for the good of others is more likely to help the client than a
WHY OPENNESS MAY NOT BE THERAPEUTIC 143

therapist who tells the client that he is using his depression as a passive-
aggressive way of making others feel guilty (see Feldman, Strong, &
Danser, 1982).
In his study of brief counseling, Elliott (1985) found that clients rated
their therapists' encouraging a new perspective on the clients' problems as
being particularly helpful, along with the therapists' offering understand-
ing, assurance, personal contact, and so forth. Elliott (1985) concluded that
what the clients deemed to be helpful went beyond the therapists' accurate
following of client material to include the clients' experiencing the thera-
pist as being lion the client's side" (p. 319).
I suggest that these central findings concerning what is helpful in
counseling (Le., the therapists' offering a new perspective on the clients'
problems and a warm, supportive relationship-showing clients that the
therapists are on their side) all support the notion that it is crucial for
therapists to convey that they view their clients in ways that the clients
would find to be desirable. This idea is the basic tenet of the proposed self-
presentational view of psychotherapy. The notion that therapy should be
conceptualized as an interactional process (e.g., Strong, 1995) also is en-
tirely consistent with the proposed view of psychotherapy.

THE PROPOSED SELF-PRESENTATIONAL VIEW

The findings that clients who successfully conceal their negative reac-
tions tend to rate their sessions as more helpful (e.g., Hill et al., 1992) and
that clients who keep relevant secrets from their therapists experience
greater symptom reduction (Kelly, 1998) are congruent with the proposed
self-presentational view of psychotherapy. If the therapist knows ex-
tremely objectionable things about the client, then the client could come to
imagine or in some cases accurately perceive that the therapist is reflecting
undesirable images of the client during therapy sessions. Supporting this
idea is the finding that several of the clients in the Kelly (1998) study
indicated that the reason they did not share their secrets was that they
were afraid the therapist would see how little progress they had made.
What was especially interesting about the Kelly (1998) study is that a
number of the clients' secrets involved lies about verifiable facts, such as
their continuing to have a sexual relationship with an abusive former
partner, yet they still experienced more symptom reduction than did the
clients who were not keeping any relevant secrets. It may be that those
clients saw their presentations as more representative of themselves, even
though they lied about some things, than did clients who were completely
truthful about their shortcomings. People tend to see more desirable self-
presentations as more truthful (Schlenker, 1980, 1986), as did the under-
graduate and client participants in the Kelly et al. (1996) experiments who
144 CHAPTER 7

rated their more depressed self-descriptions as relatively atypical of them-

selves.
As described earlier, the components of the self-presentational view of
psychotherapy are as follows: (I) Clients perform various self-presentations
in therapy; (2) therapists then offer feedback to their clients based on those
self-presentations; (3) this feedback can lead to clients' internalization of
their self-presentations; and (4) this internalization process followed by
similar self-presentations and feedback eventually may lead to changes in
their self-concepts (Kelly, 2000a). In addition, the internalization effects
resulting from counselor feedback may be enhanced by the perceived
expertise of the counselor (McKillop & Schlenker, 1988). Hence, in the
proposed view of psychotherapy, I emphasize the importance of clients'
making good decisions about what to reveal to their therapists that will
enable the clients to construct believable and desirable images of them-
selves before this expert audience.
In the proposed view of psychotherapy, I have identified both the self-
presentations and the corresponding feedback as important to self-concept
change, even though the relative contributions of each to self-concept
change are not known (see Kelly, 2000b). Emphasizing both, as opposed to
only the self-presentations, provides a parsimonious yet comprehensive
accounting for the evidence. Any theory of psychotherapy should account
for the interactional nature of the relationship between the therapist and
client, including the feedback from the therapist (Strong, 1987). Even
though clients keep some undesirable information from their therapists,
they also tend to disclose their problems such as feelings of low self-esteem
and depression (e.g., Kelly et a1., 1996). Because the clients are performing
so many negative self-presentations to an important audience, one might
expect that they would emerge from therapy worse off than if they had
received no treatment. However, researchers using meta-analytic tech-
niques (e.g., Smith et a1., 1980) have shown that the average treated therapy
client is better off than 80% of the comparable untreated clients (i.e., those
on a waiting list for treatment). Why do clients fare so well in therapy
despite their many negative self-presentations? I argue that it is the real or
imagined favorable feedback from the therapists, along with correspond-
ing favorable changes in the clients' self-presentations, that can explain
how clients come to experience positive self-concept change (Kelly, 2000a).

Explaining Previous Findings

This self-presentational view of psychotherapy provides an explana-
tion for many of the puzzling findings that have surfaced in the psycho-
therapy literature. First, it may help to explain the findings that therapists
WHY OPENNESS MAY NOT BE THERAPEUTIC 145

and clients typically .do not agree about how the clients are faring in
therapy (e.g., Caracena & Vicory, 1969; Fish, 1970; Hansen, Moore, &
Carkhuff, 1968; Hill, 1974; Hill et al., 1988b; Kurtz & Grummon, 1972;
Tichenor & Hill, 1989; Truax, 1966; see Lambert & Hill, 1994, for a review).
This perplexing set of results makes sense when one considers that clients
may tend to view therapy as more helpful when their therapists are giving
them favorable feedback (as specified by the proposed view of therapy). In
contrast, the therapists may be more inclined to view therapy as effective
when their clients are in the midst of divulging painful or troubling
personal information (as specified by the more traditional approaches to
therapy).
Along these lines, when Eugster and Wampold (1996) identified what
therapy process variables (e.g., patient and therapist involvement in a
session, patient and therapist comfort, etc.) 114 therapists and 119 of their
patients perceived to be important in their separate evaluations of how
useful or good a psychotherapy session was, the researchers discovered
that there were Significant differences between the two groups. For the
therapists, it was their perceptions of their own expertness that were most
closely associated with their ratings of how good a session was. However,
for the patients, it was their perceptions of how much the therapists
seemed to like them and seemed to act outside of the constraints of the
therapist role (i.e., the therapist real relationship) (Gelso & Carter, 1994)
that were most closely related to how good the patients thought a session
was. At the same time, the therapists' perceptions of the therapist real
relationship were negatively associated with their evaluations of a session
when all the process variables were considered simultaneously. What this
pattern suggests is that therapists may be underestimating how much
liking their clients plays a role in helping them. It is possible that if thera-
pists change their ideas about what is helpful to clients, there will be
greater congruence between therapists' and clients' ratings of therapy
outcome.
Second, Hill and colleagues (Hill et al., 1992; Regan & Hill, 1992;
Thompson & Hill, 1991) have demonstrated that therapists and clients rate
therapists' interventions as less helpful when the therapists accurately
perceive their clients' hidden (usually negative) reactions or things left
unsaid. This counterintuitive set of findings may be explained both by the
fact that clients who express negative reactions to their therapists tend
to be less well liked (Russell & Snyder, 1963) and by the fact that clients
who are less well liked by their therapists tend to have poorer therapy
outcomes (Ehrlich & Bauer, 1967; Nash et al., 1965; Stoler, 1963). In essence,
when clients express negative reactions, their therapists may like them less
and therefore the therapists may be less inclined to express desirable views
146 CHAPTER 7

of the clients. These clients may benefit less from therapy because they are
missing out on the most crucial element of the proposed view of psycho-
therapy: getting desirable feedback about themselves from their therapists.
The notion that clients should be concerned about creating desirable
images before their therapists challenges traditional forms of psycho-
therapy that conceptualize "working" or "experiencing" in therapy as
involving clients' revealing previously hidden private information (e.g.,
Klein, Mathieu, Gendlin, & Keisler, 1969; Klein, Mathieu-Coughlan, &
Kiesler, 1986; Stiles et al., 1979). However, this idea is quite consistent with
the rationale behind solution-focused therapies, which depict the delving
into client problems as an unnecessary part of treatment (e.g., Chevalier,
1995; Fish, 1996; O'Brien & Pilar, 1997; Okun, 1997). It also is consistent with
other cognitive-behavioral approaches in which the therapists emphasize
the positive aspects of their clients in an effort both to reinforce (or shape)
the clients' optimistic statements about themselves. These efforts are de-
signed to enhance the clients' feelings of being capable of solving their
problems (e.g., Kyrios, 1998). The proposed view of psychotherapy may be
seen as offering a deeper level of understanding these processes of change,
because it addresses the prospect that the clients' perceptions of how their
therapists view them may come to influence how the clients view them-
selves.
Certainly, clients do need to reveal personal information to be able to
receive helpful interpretations of their difficulties. At the same time, the
fact that some clients keep secrets from their therapists may not be prob-
lematic. The following are descriptions of what might be appropriate
levels of revelation from both clients and therapists.

What Clients May Benefit from Revealing

An example of a series of self-presentations that a client might hap-
pen to make that would be both believable and desirable is a case of a
31-year-old married male client who is dealing with the problem of sexual
compulsion. He has been extremely promiscuous, including having a one-
time sexual encounter with a 12-year-old girl. The man does not tell his
female therapist about the sex with the girl, because he fears that the
therapist will view him unfavorably. However, he does discuss his feelings
of being out of control of his sexual impulses, giving examples of nu-
merous sexual affairs and offering the therapist enough information to
provide useful, believable interpretations of his difficulties. By hearing
disclosures about his sexual behavior only with adult women, the thera-
pist may be better able to reflect desirable images (i.e., of a client who is a
good person who is struggling to control his sexual impulses) to the client.
WHY OPENNESS MAY NOT BE THERAPEUTIC 147

As therapy gets under way, the client is likely to receive both helpful and
nonhelpful interventions. If he happens to acknowledge how much he has
gained from receiving the helpful interventions even though he also ac-
knowledges the times when he feels that he is not making progress, the
therapist should be able to continue to reflect desirable images to the client.
I am speculating that this result would take place both because she likes
her client more (see Ehrlich & Bauer, 1967; Nash et al., 1965; Stoler, 1963)
and because she feels less anxious about her ability to help the client (see
Russell & Snyder, 1963; Thompson & Hill, 1991) than if he just complained
about the lack of progress. A client who presents himself in this way may
come to internalize the view of himself as one who is making good
progress. In contrast, if such a client happens to emphasize the times when
he feels that he is not making progress and finally admits to the statutory
rape, he may perceive that the therapist sees him as a therapeutic failure and child
molester, and thus may come to see himself this way.*

How Therapists May Benefit Their Clients

Therapists are likely to help their clients by assessing early on in
therapy what kinds of images the clients perceive to be desirable. The
therapists then can encourage the clients to make desirable statements
about themselves. Simultaneously, the therapists can encourage the clients
to reveal some of their problems, so that the clients will see their thera-
pists as having informed, believable opinions of them. The therapists can
respond to those revelations with interpretations that convey a new per-
spective on the clients' problems (see Elliott, 1985; Hill, 1992) and can
reflect desirable (see Shoham-Salomon & Rosenthal, 1987) and believable
images back to the clients. For instance, the therapist of the 31-year-old
male client who is dealing with feeling out of control of his sexuality may
ask how the client likes to be seen by others. The client may respond by
saying that he likes to be seen as a leader who gets what he wants, as one
who can take care of others, and as someone who is accommodating and
flexible. He also may indicate that his lack of control over his sexual
impulses makes him feel as though he is a weak, immoral person. The
therapist might interpret this revelation by stating that expressing his

"It may seem unethical to indicate that it is acceptable for clients to hold back information
from their therapists. Imagine, for example, if he continues to have sex with children. In
that case, the therapist could have prevented the very negative outcome by eliciting the
client's disclosure of the statutory rape and reporting it to the authorities. However, many
therapists already do prompt clients to avoid revealing acts that the therapists would have to
report to the police by informing the clients at the outset of therapy about the legal limits of
confidentiality (see Roback & Shelton, 1995).
148 CHAPTER 7

sexuality may stem from the lesson he learned as an attractive adolescent

that strong men get what they want through physical means, including
offering sexual gratification to women. The therapist also might state that
it is because of the success of his seductiveness as an adolescent that he
continues to be seductive as an adult and at the same time he now has the
flexibility to change his behavior.
The therapists may enhance positive self-concept change by explicitly,
as opposed to subtly, giving believable and desirable feedback to clients.
Therapists already do this in practice in the form of interpretations, and
research on brief therapy techniques shows that this is one technique that
has been linked with positive outcomes (Hill,1992).
An implication of applying these principles to psychotherapy is that
therapists could offer therapy on a relatively short-term basis, because
such therapy would not necessarily involve the delving into buried mate-
rial such as unconscious intrapsychic conflicts and repressed memories. A
practical consideration for therapists today is that even though there are
many long-term approaches to therapy and even though some clients may
need longer-term treatment, research shows that the average client who
begins treatment (such as a client who has prepaid mental health insur-
ance) is likely to receive only about six sessions (see Garfield, 1994). More-
over, when clients and their therapists are asked to indicate how many
sessions the clients will need, the clients' estimates are lower, and it is these
lower estimates that are the better predictor of how many sessions the
clients actually receive (see Garfield, 1994).

BoUNDARY CONDITIONS

In this section, I describe several key boundary conditions to the

proposed self-presentational view of psychotherapy to help the reader
determine when it makes the most sense to think of therapy in these self-
presentational terms. After reading the example about the child molester,
the reader might misconstrue what I am saying to mean that therapists
should encourage clients to hide essential information from them that
may be necessary for their treatment. For example, my theoretical perspec-
tive may be mistakenly seen as encouraging people with personality
disorders to practice manipulating their therapists as they do others. I am
not suggesting that clients should set out deliberately to hide information
from their therapists that could be essential to their recovery. Rather, I am
suggesting that if the clients feel that the revealing ofa specific piece of informa-
tion about themselves would make it very difficult for them to imagine that the
therapist still sees them favorably, even if the therapist offers them support, then
it is acceptable for the clients to discuss the themes as opposed to the details of those
WHY OPENNESS MAY NOT BE THERAPEUTIC 149

revelations. That way, the clients will still be able to talk about relevant
information without making the sessions so embarrassing that they may
not want to return to face their therapist. For example, a client who is
struggling to overcome her debilitating fear of germs may benefit from
describing her oven-cleaning, floor-scrubbing, and hand-washing behav-
iors in an effort to work through these rituals, as long as describing these
rituals would not make it difficult or impossible for the client to imagine
that the therapist views her favorably. In contrast, a client like Jeffrey
Dahmer who has dissected people and animals for pleasure may want to
discuss his "sadistic urges" and leave out the ghastly details. The problem
with revealing these details is that such revelations are likely to leave the
client feeling that the therapist sees him in undesirable ways, given that
society in general condemns such individuals.
The proposed view of psychotherapy seems to place a fair amount of
responsibility on the client to decide what he or she is comfortable reveal-
ing. The fact is that clients already do censor their revelations in practice
(Hill et al., 1993; Kelly, 1998), and there is some evidence that clients who
engage in such censorship do better in therapy (Kelly, 1998). The therapist
can facilitate the client's decision making by reminding the client that it
is his or her choice to reveal private information and that complete open-
ness is not a requirement of successful therapy. It might make the clients
who do hide information from their therapists feel relieved that they are
not necessarily undermining their treatment.
At the same time, there are likely to be some clients who begin therapy
with very negative self-views surrounding specific transgressions or hu-
miliating events, and who over time could come to believe that their
therapists would still have favorable impressions of them, even if the ther-
apists knew about these transgressions. Such clients might benefit tremen-
dously from disclosing these events and then hearing the therapists' chal-
lenges of their negative self-views surrounding the events. An example
might be a woman who enters therapy feeling extremely ashamed and
guilty about having shaken her crying baby brother when she was 10,
leaving him with mild brain damage. She does not reveal this event until
after a year of psychotherapy, whereupon the therapist reframes the event
in a believable way by telling her, "You weren't capable of thinking like an
adult when you were 10, and you need not hold yourself accountable for
what you did as a kid. In fact, that event may have played a role in why
you have developed to be so kind and considerate of others now."
What should therapists do when clients choose to reveal information
about themselves that the clients find to be desirable but that therapists
and others would find to be abhorrent or maladaptive? An example might
be a client who feels that she would enhance her existence by taking
150 CHAPTER 7

revenge on all the men who ever abused her. Another example might be a
former police officer who believes that he knows what is best for the city
and is considering shooting all the local drug dealers. Should therapists be
restricted to reflecting what their clients believe to be desirable? Or should
therapists try to change these potentially destructive beliefs? A potential
solution lies in the therapist's ability to reframe the clients' articulated
desirable self-images in such a way that the clients still find them to be
desirable and believable, and that the therapist perceives as not being
problematic. For instance, the therapist might accurately perceive that
what underlies the desire to kill the drug dealers is the wish to become a
respected, powerful hero. The therapist can convey his respect for the
client by saying, "You have been a wonderful leader who has protected the
community for many years. At the same time, being a great citizen in-
volves your upholding the law. How can you put your energies into
helping the community within the limits of the law?"
Rogerian therapists have encountered similar challenges with the
notion of their offering unconditional positive regard to antisocial clients
who may express the desire to harm others. Rogers's (1951, 1957) theory
accounted for such concerns by assuming that human nature is essentially
good and if clients were truly left to their own accord, they would not
choose to do such things. I also think that it would be unlikely for clients to
articulate such antisocial goals to their therapists. However, rather than
my assuming that this censorship stems from the fact that humans are
essentially good, I am postulating that the great majority of clients would
have some idea that their therapists would not view such a statement in a
positive way. Because therapy is an interactional process, the clients' initial
expressions of desirable self-images would already be affected by what the
clients anticipate that the therapists would view as favorable. Thus, it
would be a small minority of clients who would articulate such destructive
tendencies or maladaptive self-beliefs. In such cases, the therapists would
most likely need to concentrate on modifying the grossly impaired judg-
ment or potential dangerousness of the clients.
How can therapists inoculate clients against receiving feedback about
themselves from other important audiences outside the therapy context
that contradict the desirable feedback from their therapists? For example,
if a therapist reflects images to a female client that she is effervescent and
intelligent, yet the client's boyfriend sees her as somewhat frivolous and
simpleminded, then the client may have a difficult time maintaining the
benefits of the favorable feedback from the therapist. She may come to see
that she feels better about herself after being with the therapist than with
her boyfriend, and thus may want to break off the relationship with the
boyfriend. In such cases, the effectiveness of therapy might be enhanced
either by including the fiance in the therapy and helping him provide her
WHY OPENNESS MAY NOT BE THERAPEUTIC 151

with desirable feedback or by supporting the client's decision to disentan-

gle herself from relationships with people who view her in ways that she
finds to be particularly undesirable.
One last boundary condition is that the clients are most likely and
perhaps are only likely to benefit from their desirable self-presentations
when those presentations are within a range of self-presentations that are
believable (see Schlenker, 1986). For instance, imagine a client who tells
his therapist that he is no longer snorting cocaine, yet he comes to therapy
with dilated pupils. The client is not likely to benefit from that declaration
because both he and the therapist will know that it is not true. But if that
same client says that he has recently snorted a few lines but is really trying
hard to give it up (and he sees that statement as being truthful at least some
of the time), he is likely to fare better than if he flatly states that he does not
think that he will ever be able to stop (a statement that he also sees as being
truthful at least some of the time).

ADDRESSING ALTERNATIVE PERSPECTIVES

In this section, I address a number of anticipated counterarguments

and limitations to the proposed self-presentational view of psychotherapy.

Having One True Self

The suggestion that clients' concealing some undesirable aspects of
themselves may be helpful to them seems to be an endorsement of dis-
honesty that may impede the therapeutic process. One might argue that if
a client really thinks that his or her "true self" is undesirable, then he or she
should be as revealing as possible with the therapist to allow the therapist
to help negate those self-beliefs. However, a core assumption that I am
making in proposing this self-presentational view of therapy is that clients
have multiple self-beliefs that all fall within a fairly wide range of believ-
able self-views, as opposed to "one true" self-concept. Many of the ways
that they describe themselves to their therapists fall within this range of
believable self-presentations (see Schlenker, 1986; Schlenker et a1., 1994;
Schlenker & Trudeau, 1990). The findings from a study in which four
clients interacted with four experienced therapists and presented them-
selves very differently to the therapists (Fuller & Hill, 1985) are congruent
with this idea:
[T]he helpees presented different problems to the four counselors which in
effect made them appear to be different persons in each pairing. Undoubt-
edly, each person has a range of possible behaviors and behaves in a specific
way depending on what he/ she perceives to be appropriate in that situation.
(p.337)
152 CHAPTER 7

As such, it does not seem unreasonable to suggest that clients present

those aspects of themselves that are consistent with how they would like
to see themselves.

Recognizing the Concealment

Even though therapists and clients do conceal the unfavorable reac-
tions that they are having from each other, they might have concerns that
the other person will be able to recognize the concealment. This concern is
offset by research that shows that even highly trained psychiatrists are
generally very poor, at least initially, at determining when others are
concealing thoughts and feelings from them (see Ekman, 1991; Ekman &
O'Sullivan, 1991). Moreover, as mentioned earlier, therapists in both long-
term (e.g., Hill et al., 1993) and short-term therapy contexts (e.g., Regan &
Hill, 1992) can identify what their clients leave unsaid only a small percent-
age of the time. Therapists are typically able to guess accurately that their
clients are hiding reactions from them, but they are unlikely to be able to
detect what their clients' specific hidden reactions are or when they are
having them (Hill et al., 1992). Therefore, I suggest that even though their
concealment may be detected as a general or nonverbal level, clients and
therapists should not feel overly concerned that they will be "found out"
about a specific omission, simply because it is difficult to detect things
that go unsaid.

Not Caring What the Therapists Think

One might argue that some clients, such as those who are highly
reactant (i.e., motivated to restore their freedom) (Brehm & Brehm, 1981;
Dowd, Milne, & Wise, 1991) or high in private self-consciousness (i.e.,
"tuned out" to the expectations of others and behaving in a way that is
consistent with internal motivations) (Buss & Briggs, 1984; Carver &
Scheier, 1985), are not likely to be influenced by what their counselors think
of them. However, Baumeister (1982) gathered a substantial amount of
evidence that although reactant people often appear to be behaving in a
manner that is consistent with their internal standards, many times they
are actually behaving in a manner that is intended to create the public
impression that they are independent and autonomous (see Baer, Hinkle,
Smith, & Fenton, 1980; Heilman & Gamer, 1975; Heilman & Toffler, 1976).
Likewise, Schlenker and Weigold (1990) found that participants who were
high in private self-consciousness were highly responsive to feedback
from a partner (who regarded them as dependent or independent) and
publicly changed their attitudes in an effort to make themselves appear
WHY OPENNESS MAY NOT BE THERAPEUTIC 153

autonomous. In contrast, participants who were high in public self-

consciousness shifted their attitudes to conform to whatever the expecta-
tions of the partner were (i.e., they altered their attitudes to appear to be
independent or dependent, depending on the expectations of the partner).
Schlenker and coworkers (1996) interpreted these findings to mean that
audiences matter for everyone, although different people have different
goals when relating to others.

Gaining Catharsis
One also could argue that clients need the opportunity to ventilate
pent-up negative emotions (i.e., experience catharsis) as a means of feeling
better about their problems. This expression may include the clients' tell-
ing their therapists how much they resent the therapists and disapprove of
the therapists' attempts to help them. However, as described in Chapter 5,
the therapeutic effects of catharsis have not been supported generally (see
Bohart et aI., 1976; Ebbesen et aI., 1975; Kelly et aI., 2001; Stone et aI., 1995;
Tesser et aI., 1978).

Contradicting Assumptions of Openness

The notion that clients might benefit from censoring particularly hei-
nous details from their revelations in therapy seems to contradict psycho-
dynamiC approaches to trauma that involve remembering and working
through images of very negative or traumatic experiences. It also seems to
contradict some cognitive-behavioral therapies that require the revelation
of what can be excruciating or humiliating details. For example, as de-
scribed earlier, in exposure therapy, clients who have experienced traumas
are asked to "relive" the trauma in an effort to modify the memory
structures that may serve as blueprints for the client's fear behavior (Foa &
Kozak, 1986; Roa & Rothbaum, 1989). Foa and colleagues have demon-
strated that exposure therapy resulted in reduced anxiety-related symp-
tomatology (Foa et aI., 1991; Kozak, Foa, & Steketee, 1988) and in more
organized thoughts surrounding a rape trauma, which were negatively
correlated with depression (Foa et aI., 1995). At the surface, these findings
seem to disaffirm the proposed view of psychotherapy, but I suggest that
there is no real contradiction because the successfully treated women in
these studies probably were able to continue to imagine that their thera-
pists viewed them favorably after their revelations. A central principle of
the proposed view of psychotherapy is not that deliberate censorship on
the part of the clients is necessary, but rather that such censorship is
recommended if revealing a certain piece of information about themselves
154 CHAPTER 7

would make it difficult for the clients to imagine that their therapists still
see them in desirable ways.

Truly Knowing the Clients

Some clients may feel bad or guilty about leaving out objectionable
details about themselves. They might feel as though the therapists could
not really know them unless the therapists knew about these very negative
things. However, does truly knowing someone mean knowing the worst
things that they have ever felt or done? In most circumstances other than in
therapy settings, people would feel that if others knew such bad things,
those others would actually get the wrong impression of them. People
tend to believe that more favorable things about themselves are more, not
less, truthful (see Kelly et al., 1996; Schlenker, 1980, 1986).
One reason why clients might feel guilty about not telling their thera-
pists everything may be that they assume that the therapists have the
expectation of full disclosure, rather than that the clients have some justi-
fiable worry that they can only be truly known with full disclosure. Vir-
tually all the clients in my study on secret keeping in psychotherapy (Kelly,
1998) indicated they believed that the therapists expected full disclosure
from them. In nontherapy relationships in which there is no expectation of
full disclosure, it would be unlikely for the clients to worry about whether
they would still be accepted if the others knew about their "skeletons in
the closet." For instance, imagine a 57-year-old married woman who
enters therapy for the treatment of depression and feelings of alienation
from her colleagues. After a few sessions, she begins to ruminate over
whether she should tell her therapist about the time in high school when
she experimented with sex acts with multiple partners and even spent a
night in jail for indecent exposure. She feels both ashamed about this
promiscuous phase and guilty about not disclosing this information to her
therapist. She has never told her husband about the experience because
they have had a long-standing agreement that their previous sexual histo-
ries are private matters that do not need to be discussed. In the context of
her marriage, she considers it to be her right to keep such information
private; she feels reasonably close to her husband and does not ruminate
over what he would think of her if he knew about the previous promis-
cuity. But in the therapy setting, her belief that the therapist expects
revelation might cause her to feel bad and exert energy in hiding that
information from the therapist.
It is understandable that clients would believe that therapists expect
full disclosure. After all, as discussed earlier, modern psychotherapy has
grown out of Freudian psychoanalysis, wherein the fundamental rule of
WHY OPENNESS MAY NOT BE THERAPEUTIC 155

psychoanalysis required patients to be as open about themselves as pos-

sible (Freud, 1958; Hoyt, 1978). In addition, most counseling psychology
graduate programs have been heavily influenced by Rogers's approach to
therapy (Hill & Corbett, 1993), in which the therapeutic ideal involves high
levels of openness on the part of both clients and therapists (Rogers, 1951,
1957). My intention in proposing the self-presentational view of psycho-
therapy is to stimulate a rethinking of common assumptions about how
psychotherapy leads to change. Specifically, I believe that if clients and
therapists could change their expectations that therapy be a place for
full disclosure, then clients might not have those bad feelings about
whether the therapists would still accept them if the therapists knew about
the most negative things.

Manipulating the Therapist

The assumptions underlying the proposed self-presentational view of
psychotherapy are directly opposed to the common notions of client self-
presentations as maladaptive manipulations (e.g., Strong, 1987). Accord-
ing to Strong's (1987, 1995) interpersonal influence theory of counseling,
the beginning of therapy is based on patients' perceptions of their own
ineffective self-presentations. The therapist's role is to cause a change in
the client's understanding of his or her environment in order to promote a
change in the client's self-presentations. In inducing such change, the
therapist needs to avoid being manipulated and to offer clients a valida-
tion of the clients' most feared selves: to give clients unexpected responses
while also supporting them (Strong, 1987, 1995). However, there is no
evidence to back the idea that giving clients feedback that they would fear
or find to be undesirable is helpful, and there is a good deal of evidence
that providing clients with feedback that suggests that the therapist is
"on the client's side" is helpful (see Elliott, 1985).

Just Providing Social Support

Yet another concern may be that this view of psychotherapy is a
euphemistic reframing of the idea that therapists are merely available to
provide social support, just as friends do. In actuality, friends often do not
provide effective social support when people really need it, such as when
victims disclose details surrounding the death of a loved one (e.g., Davido-
witz & Myrick, 1984). Moreover, as Swann's (1996) research has demon-
strated, people's romantic partners and close friends often make attempts
to confine them to acting in a manner that is consistent with the views that
the friends already have of them. These views may be undesirable to the
156 CHAPTER 7

individuals, thus leaving them feeling trapped and unable to improve

themselves. Because of the therapists' training in providing desirable and
believable feedback, they should be better able to offer such feedback
consistently than friends should be. Given the status associated with the
expert role of the therapist, the therapists' feedback should have an even
greater effect on internalization of these positive self-images than should
friends' feedback (see McKillop & Schlenker, 1988).

Having a Weak Empirical Foundation

Finally, there are some constraints to the empirical foundation of the
proposed view of psychotherapy. As mentioned earlier, the internalization
experiments from the social psychology literature have yet to be attempted
in psychotherapy contexts. Moreover, because most of the studies from the
clinical-counseling literature on clients' openness have been correlational
in nature and conducted in short-term therapy contexts, the conclusions
that can be drawn from them are restricted. For example, because of the
correlational design of my secret keeping in psychotherapy study (Kelly,
1998), I could not establish that keeping relevant secrets in therapy caused
the symptom reduction in clients. I merely showed that there was a rela-
tionship between these two variables. Also, because the clients in my
study were in relatively short-term therapy, it is not known whether this
relationship would hold up in long-term therapy contexts (Kelly, 1998). In
essence, more research is needed to see if the proposed view of psycho-
therapy predicts future psychotherapy outcomes in both short- and long-
term therapy settings. Experiments could involve randomly assigning
clients to therapists who either encourage the revelation of highly objec-
tionable material or explain that such material is private and up to the
client's discretion to reveal. The experimenters could then assess the extent
to which clients do indeed reveal or conceal very undesirable information
about themselves to their therapists. The clients' perceptions of the desir-
ability of the therapists' feedback also could be measured to see how those
perceptions are related to both short- and long-term changes in the clients'
self-images.
Such studies would provide answers to some of the questions sur-
rounding the intriguing phenomenon of one's making multiple self-
presentations to important audiences over time. To date, multiple self-
presentations followed by audience feedback and internalization of that
feedback over time have not been adequately assessed by self-presentation
researchers (McKillop, 1991). These studies could assess how crucial it is
for one's self-presentations to be believable when interacting with re-
spected audiences, namely the therapists. It has been postulated that
WHY OPENNESS MAY NOT BE THERAPEUTIC 157

believability is important in predicting whether one will internalize one's

self-presentations (e.g., Schlenker, 1986), but the parameters of this be-
lievability have not been clearly delineated in the context of multiple,
complex self-presentations. The fact that one study (Kelly, 1998) showed
that clients who actually lied to their therapists about verifiable facts fared
well in terms of their experiencing symptom reduction begs the question
of what kinds of complex sets of self-presentations are seen as most
believable and most likely to be internalized. Are these self-presentations
ones that are completely free of lies or ones that include them? The inter-
action of clinical-counseling and social psychological research offers enor-
mous potential for understanding self-concept change more fully.

CONCLUSION

Many therapists encourage their clients to discuss undesirable aspects

of themselves in therapy in an effort to help the clients change those
aspects. Yet given the social psychological research on self-concept change
(e.g., Fazio et al., 1981; Jones et al., 1981; Rhodewalt & Agustsdottir, 1986;
Schlenker & Trudeau, 1990; Tice, 1992), I suggest that this process of
extracting undesirable information actually may be encouraging clients to
solidify these negative images of themselves. Clients attempt to manage
the impressions therapists have of them and they view more depressed
self-presentations as less representative of themselves (e.g., Kelly et al.,
1996). They also seem to come to internalize the images that they perceive
their therapists have of them (Quintana & Meara, 1990). Assuming that
the goal of therapy is to induce positive self-concept change and that the
self-concept change processes observed by social psychologists generalize
to psychotherapy contexts, it may make sense for counselors to reflect
images of their clients that the clients would find to be desirable and
representative of themselves.
There certainly will be times when clients reveal undesirable informa-
tion about themselves to their therapists. I am not suggesting that the
therapists' responding in a normalizing, accepting manner is harmful. The
therapists should accept and reframe the clients' problems, just as the re-
search suggests (e.g., Elliott, 1985; Shoham-Salomon & Rosenthal, 1987).
What I am adding here is the suggestion that when clients reveal terrible
things about themselves, such as their having savagely beaten their chil-
dren or cruelly neglected their pets, the clients are likely to imagine that
their seemingly accepting therapists are judging them negatively. In fact,
therapists do make very negative clinical conjectures about their clients'
pathologies (Regan & Hill, 1992). Because therapists tend to use the details
158 CHAPTER 7

of the clients' problems to develop themes or explanations for the prob-

lems during the course of therapy, perhaps the clients themselves could
focus on the themes to their problems. This option might be preferable to
the clients' running the risk of perceiving that they have horrified their
therapists with the details of their sometimes abhorrent actions.
In a nutshell, instead of suggesting that clients deliberately should
hide information from their therapists to make a good impression, I note
that clients do conceal information as part of either conscious or uncon-
scious attempts to construct desirable images before the therapists. There
is evidence that such discretion does not necessarily undermine their
treatment (and it may even enhance their treatment). In those cases when
clients believe that their therapists could not view them favorably after
hearing particularly heinous revelations, I suggest that it is acceptable for
the clients to discuss themes as opposed to details of those revelations. At
the same time, there may be some clients who enter therapy with very
negative self-views surrounding specific transgressions or humiliating
events and who may come to believe that their therapists would still have
favorable impressions of them even if the therapists knew about these
events. I suggest that these clients might benefit immensely from telling
the therapists about these events and then hearing the therapists' chal-
lenges of their negative self-views surrounding the events (Kelly, 2000a).
Rather than my suggesting that clients show only positive self-images to
their therapists, I propose that it is acceptable for clients to use some
discretion based on their perceptions of the therapists' responses (Kelly,
2000a,b).
So, what is the main contribution of this reconceptualization of psy-
chotherapy? For some therapists, such as some cognitive-behaviorists,
adopting this new perspective would not change anything that they do in
therapy. These therapists already spend time reinforcing the favorable
statements that clients make about themselves. Yet for other therapists,
such as some classically trained psychoanalysts, adopting this view would
dramatically alter their efforts at eliciting the patients' expression of re-
pressed or buried negative material. I see the major contribution of this
proposed view not so much as offering a new set of techniques for inter-
vention, but rather I see it as offering a ne'", understanding of how psycho-
therapy works. Specifically, in the proposed view, I emphasize the role of
the therapists' opinions of their clients in effecting client change. I also see
this proposal as a challenge to the idea that therapy is a place where clients
are supposed to be completely revealing of themselves. If both therapists
and clients could change their expectations that revelation is required of
the clients, then the clients could potentially benefit even more from
psychotherapy. This idea gets back to what I described in Chapter 1 as the
WHY OPENNESS MAY NOT BE THERAPEUTIC 159

distinction between secrecy and privacy. If clients can view the material
that they keep from their therapists as private material that the therapists
do not expect access to, then the clients may not feel bad about keeping
such information from their therapists.
In clOSing, I suspect that social psychologists (e.g., Arkin & Hermann,
2000) are likely to accept the notion that the "masks" people wear when
interacting with others are a very important part of who they are and will
become. In contrast, clinical and counseling psychologists (e.g., Hill, Gelso,
& Mohr, 2000) may reject much of the evidence that people's self-concepts
are influenced by what they show others. For example, Hill et al. (2000)
stated that "we believe that one of the most important things that can
happen in therapy is for therapists to accept clients deeply for themselves
as they are" (p. 498). Besides rejecting my position, Hill and coworkers'
comment also implies that self-concepts are more stable than they actually
may be, given that people seem to have multiple self-concepts over time
(see Baumeister, 1998).
Perhaps social psychologists and clinical-counseling psychologists
will seem to have in common a greater sense of optimism about the
psychotherapy process than what I have described here. For example,
Arkin and Hermann (2000) recommended that clients fully describe their
very negative behaviors, so that their therapists can help them see their
behaviors as separate from the broader implications of who they are.
Likewise, Hill et al. (2000) seemed to suggest that therapists can and do
truly hold their clients in high regard, even when the clients reveal heinous
details. I have seemed somewhat less optimistic in suggesting that being
judgmental is part of human nature and that clients are rightfully sensitive
to the possibility that their therapists may form very negative clinical
conjectures about them if the clients say really heinous things about them-
selves (see Regan & Hill, 1992). The key message from my proposal is that
at its core psychotherapy is an interpersonal endeavor that is not exempt
from social processes that characterize normal discourse. In particular,
people put on masks when interacting with others. These "masks" have
important implications for how they perceive that others see them and for
how they see themselves. I am optimistic that once this perspective is ack-
nowledged, scientists can exploit what is known about self-presentation to
enhance their understanding of psychotherapy and potentially increase its
effectiveness (Kelly, 2000b).
CHAPTER 8

DILEMMAS TO REVEALING
SECRETS AND THE ROLE
OF THE CONFIDANT

Now I move away from a discussion of revealing secrets in therapy to the

broader dilemmas associated with revealing secrets to confidants in gen-
eral. In Chapter 4, I noted that some critics thought that the ending of the
film "Secret and Lies" seemed false, because it is unlikely that people
would so readily accept some secrets that they previously considered
loathsome. In particular, the movie simplistically depicted uneducated
people with racist beliefs readily embracing a family member from a dif-
ferent race. What the movie illustrated is that the responses of the confi-
dant should largely determine whether someone will benefit from reveal-
ing a secret; all was well in the end because the family could accept the
secret. However, in real life, some confidants cannot be expected to accept
some secrets, and they might even be obligated to reject the revealer.
For this reason, I believe that pop star Michael Jackson made the right
decision never to admit to having behaved inappropriately with a 13-year-
old roughly a decade ago. If he did admit to any sexual misconduct, then
(among other problems) the people who spend time with him like Eliza-
beth Taylor would be put in an awkward position; they would seem as
though they were defending a child molester. In the absence of a confes-
sion, his friends and fans can avoid being compelled to act on what would
become public knowledge and avoid having to reject the pop star. Like-
wise, I think Hillary Clinton made the right decision never to admit
knowing about the Monica Lewinsky affair as it was happening. If she

161
162 CHAPTER 8

had, people might see her lack of intervening on behalf of the young in-
tern as a sign of weakness or immorality.
What strings these examples together is that revealing secrets typ-
ically has a social context in which the revealer is held accountable for the
disclosed information, unlike the revealing that went on in the writing
experiments described in Chapters 4 and 5. As John Steinbeck (1961) wrote,
"The king told his secrets down a well, and his secrets were safe. A man
who tells secrets or stories must think of who is hearing or reading, for a
story has as many versions as it has readers" (p. 89). Unfortunately,
researchers studying disclosure have not paid enough attention to the role
of the confidant (see Kelly & McKillop, 1996). For example, even though
Pennebaker (1990) acknowledged the importance of having an accepting
confidant when disclosing private information, at one point he approv-
ingly described a psychiatrist's advice to a traumatized woman "to tell
her story to everyone she met" (p. 37, italics added). I contend that such a
recommendation can backfire with stigmatizing secrets and that the health
benefits of revealing most likely would be wiped away if one were to
reveal to an indiscreet, judgmental, or rejecting confidant. The experiments
on the health benefits of revealing were extremely important in capturing
the mechanisms that can lead to recovery from troubling secrets. However,
the reader must keep in mind that the participants in those experiments
were assured that their responses would remain confidential and anony-
mous. Thus, it is difficult to extend the findings from such experiments to
everyday situations. "In the laboratory we have inadvertently provided
people with a safe, nonjudgmental haven for disclosure" (Pennebaker,
1990, p. 65). Such protection allowed participants to avoid the usual worry-
ing about whether their confidants would tell all their peers about their
wrongdoings, shortcomings, or private embarrassments, or would reject
them outright (Kelly & McKillop, 1996).
Along these lines, there is evidence that although supportive social
networks often reduce distress in individuals experiencing stressful life
events, unsupportive or critical social networks can increase it (e.g., Abbey,
Abramis, & Caplan, 1985; Holahan, Moos, Holahan, & Brennan, 1997;
Kennedy, Kiecolt-Glaser, & Glaser, 1990; Lepore, 1992; Major et al., 1990;
Major, Zubek, Cooper, Cozzarelli, & Richards, 1997; Manne, Taylor,
Dougherty, & Kemeny, 1997; Rook, 1984; Vinokur & Van Ryn, 1993). More-
over, disclosure is not necessarily associated with positive outcomes for
the revealer (e.g., Cutrona, 1986). As Lepore, Ragan, and Jones (2000) put it,
"Disclosure seldom occurs in a social vacuum, and its effects may be
contingent on the social context" (p. 500).
Despite the health benefits of anonymous and confidential revealing
DILEMMAS TO REVEALING SECRETS 163

described in Chapter 4, some research has not supported the idea that
sharing emotions (i.e., joy, affection, sadness, fear, anger, and shame) with
others is helpful or curative (Rime et aI., 1991a). In that series of studies,
although an overwhelming majority of the participants reported being
highly motivated to share their emotions with others (especially their most
disruptive emotions), the amount of social sharing participants had en-
gaged in was not associated with their recovery over time from the disrup~
tion caused by the emotion (Rime et aI., 1991a). Moreover, as described in
Chapter 1, Finkenauer and Rime (1998a) found that shared and nonshared
events did not differ on ratings of recovery.
In this chapter, I describe the trade-offs associated with revealing
secrets to other people. It can feel good initially to unburden secrets, and
there is evidence that a revealer can become more physiologically relaxed
even when there are dire consequences to revealing, such as when a
criminal discloses crimes to a police officer (see Pennebaker, 1985). How-
ever, once the revelation has occurred, the revealer has to live with the
consequences. The following quote from the confession of the infamous
serial killer Edmund Emil Kemper illustrates this dilemma:
I went in to some detail today on these cases, and I wish I hadn't now.
(Referring to statements made earlier in the day during a telephone conver-
sation he had with police officers after surrendering.... While waiting for
units to arrive and take him into custody, Kemper spoke erratically, offering
to the officers fragmented details as evidence that he was, in fact, the "Co-Ed
Killer.") It's been bothering me more and more, just thinking about it, and
then talking about it today with someone else. It just really didn't have an
effect then. I told the officers that when I am talking about something like
that, from being in Atascadero for so long, and talking about very serious
things and very tender things, bothersome things, I get kind of calloused,
you know, where I don't show emotion. I just talk, getting the thing out, and
later on it hits me. I spent the whole afternoon in there trying to decide
whether I was gonna climb the bars and jump off or hang myself.... You
know, I was really very seriously depressed about the whole damn thing, so I
was hoping that-I suppose you're going to have something to go on prior to
going back and really getting something laid out. (Edmund Emil Kemper, III,
http://serial-killers.virtualave.net/kemper.htm. Retrieved November 21,
2000.)

I not only explore the negative and positive consequences of revealing

personal secrets in this chapter but also discuss how the confidant's quali-
ties and responses to a revelation can have a great impact on the revealer.
The chapter ends with a description of the characteristics of the ideal
confidant.
164 CHAPTER 8

NEGATIVE CONSEQUENCES OF REVEALING SECRETS

SECRET Is REPEATED TO OTHERS

One of the biggest problems associated with revealing personal se-

crets is that confidants often cannot be trusted to keep the secrets or to
protect the revealer's identity. In a recent pair of studies, college students
reported that when someone had disclosed an emotional event to them,
they in tum revealed the emotional disclosure to others in 66 to 78% of the
cases (Christophe & Rime, 1997). This high rate of revealing occurred
despite the fact that the participants were intimates of the original re-
vealers in 85% of the cases (Christophe & Rime, 1997). In addition, when
the original disclosure was of a high emotional intensity, as compared with
when it was of a low or moderate emotional intensity, the participants
indicated that they had shared it with others even more frequently and
had told more people. On the average, they told more than two other
people. In another study, Christophe and Di Giacomo (1995) found that in
78% of the cases in which the original event was disclosed to others, the
name of the original revealer was explicitly mentioned. These researchers
recommended that if people do not want others to learn about their
emotional experiences, then they should avoid sharing the experiences
with others altogether.
It is important to note, however, that a limitation to these studies is the
participants were asked about their having heard emotional episodes, as
opposed to secrets. The participants were not explicitly asked if they had
actually been sworn to secrecy when listening to the emotional episodes.
It is possible that explicitly telling a listener that confidentiality is expected
would discourage the person from repeating the secret. In addition, due to
the reciprocal nature of revealing personal information, it is likely that
intimates know personal information about each other that could be used
as a form of assurance that neither will disclose the others' secrets.
Petronio and Bantz (1991) conducted a study that more directly as-
sessed the lack of discretion surrounding the revelation of secrets per se.
They found that although the phrase" don't tell anybody this" is intended
to keep one's confidant from revealing one's secrets, that may not be the
case (Petronio & Bantz, 1991). A large group of undergraduates completed
questionnaires consisting of scenarios that varied in terms of how private
(high, moderate, or low) the information transmitted was and whether the
phrase "don't tell anybody this" was used in revealing the information
(Petronio & Bantz, 1991). A substantial percentage of both revealers and
confidants indicated that they expected the confidants to repeat that infor-
mation to others, which suggests that revealers are aware of potential
DILEMMAS TO REVEALING SECRETS 165

ramifications from disclosure. What was most interesting about these

findings was that whether or not the phrase" don't tell anybody this" was
used, confidants were more likely to repeat information than the revealers
expected when that information was either highly or moderately private.
Again, these findings reinforce the idea that revealing one's secrets may be
ill-advised because the confidant may repeat the secret to others, espe-
cially when the secret is "juicy."

REVEALER'S IDENTITY CAN BE UNDERMINED

The danger to having others know one's stigmatizing secrets is that
his or her very identity may be undermined by such exposure. For exam-
ple, if a man reveals an especially humiliating secret (or his confidant
reveals it) to his friends, then his friends become witnesses and reminders
of this humiliation. The following example is a true story, but the names
and some of the details have been changed to protect the identity of the
real-life "Jeff": At age 30, Jeff still struggles to get the love of his life out
of his mind 3 years after she abruptly abandoned him. The abandonment
was particularly painful, because it came suddenly only 21 days after she
moved into his apartment. He had returned home to find her clothes gone,
with no note of explanation. When he was later able to track her down, she
had moved to a new city and told him that she was not ready for a deep
commitment. She said that perhaps when they were both 40 there could be
some sort of rekindling of the relationship. Jeff still recounts those excru-
ciating details to his work associates, whom he considers to be close
friends. Those who met her agree with Jeff that she is one of the most
beautiful, intelligent, and mesmerizing women they ever met. One of the
friends, Alex, noted that after seeing that Jeff was capable of being in a
relationship with such an extraordinary woman, he became convinced
that Jeff was "capable of attracting even supermodels." When Jeff tells
his story to new close friends, he quotes Alex on this point.
Somehow these revelations to his friends do not seem to make Jeff
feel any better, even 3 years after the breakup. As compelling as it is to tell
others about personal struggles and emotional events in general, in this
case the problem has now become the revelations themselves. Because his
friends now know about this failure experience and because their opinions
matter a great deal to him, Jeff feels that he has to win her back to show
them that he is not a failure and that he truly is capable of attracting the
very most desirable women. If he had never admitted that he was so
devastated by that experience and had never mentioned what happened,
it is possible that he would have recovered by now. In essence, just the
sight of his friends reminds him of the rejection, and it is the friends' seeing
166 CHAPTER 8

him as a failure that is so troubling. As Jeff explained to me, even if he

acts as though he is interested in a new woman, he imagines they know
that he is still suffering.
Jeff recently made a new friend who has never met this woman and
who, despite hearing about the painful breakup, expresses his belief that
Jeff is masterful at wooing women and that Jeff actually is over the woman.
It is this friend toward whom Jeff gravitates for his salvation, because
this friend has managed to discount the revelation given his own lack of
belief in any such romantic things.
What Jeff's story illustrates is that a problem associated with telling
others especially humiliating events and receiving undesirable feedback
from them is that as described in the previous chapter, people may form
their identities through interacting with others (e.g., James, 1890; Cooley,
1902; Mead, 1934; Goffman, 1959). They incorporate their public self-
portrayals into their private self-beliefs in a process known as internaliza-
tion (Tice, 1992). Researchers have shown that public expressions can influ-
ence private self-beliefs regarding depression (Kelly et al., 1991), sociability
(McKillop, Berzonsky, & Schlenker, 1992), independence (Schlenker &
Trudeau, 1990), and global self-esteem Oones et al., 1981; Rhodewalt &
Agustsdottir, 1986).
Although social psychologists know that internalization occurs, they
have offered at least three different major explanations for why it occurs.
The explanation I emphasized in the previous chapter is what Schlenker
et al. (1994) referred to as public commitment to an identity. Specifically, after
people perform various self-presentations, they then incorporate either
real or imagined feedback from the audience (e.g., therapist or confidant)
to construct their self-images because publicly claiming to be particular
kinds of people obligates them to behave consistently with those identities
(e.g., Schlenker,1980; Schlenker et al., 1994). If people fail to be consistent in
their self-descriptions and actions, they may face the very negative conse-
quences of being seen as unreliable, hypocritical, or self-deluding, or as
liars (Kelly, 2000b). Along these lines, Schlenker et al. (1994) found that
participants who described themselves as sociable to one person actually
behaved more sociably with a new person. Schlenker and co-workers
argued that the new person would give feedback that is consistent with
more sociable self-descriptions, and thus the actor would become more
sociable to be consistent with the feedback.
A second explanation comes from self-perception theory (Bem,1972).
People look to their self-portrayals to infer their attitudes about them-
selves, especially when the self-portrayals are not incompatible with their
prior self-beliefs surrounding their self-portrayals or when the prior self-
beliefs are weak. A third explanation is the biased-scanning version of self-
DILEMMAS TO REVEALING SECRETS 167

perception theory. When people describe themselves in certain ways, they

search for information in their memories to support their freely chosen
self-descriptions aones et al., 1981; Rhodewalt & Agustsdottir, 1986). The
major distinction between these two explanations is that the former de-
scribes the creation of new self-beliefs, whereas the latter describes how
some existing self-beliefs are made more salient than others (Schlenker
et al., 1994).
I contend that the first explanation for internalization, the public
commitment view, is the best one so far because it seems to provide the
best fit with existing evidence. For example, after reviewing approx-
imately 50 naturalistic studies, Shrauger and Schoeneman (1979) con-
cluded that "people's self-perceptions agree substantially with the way
they perceive themselves as being viewed by others" (p. 549). Further-
more, researchers have shown that once people have committed them-
selves to particular identities, their peers tend to give them feedback that
constrains them to being consistent with those identities (e.g., see Swann,
1996).
Also supporting the public commitment view is evidence that the real
or imagined presence of an audience is a crucial element of internalizing
one's self-portrayals (Baumeister & Tice, 1984; Schlenker et al., 1994; Tice,
1992). Tice (1992) found that self-portrayals had a greater impact on the
participants' self-concepts when they were performed publicly rather than
privately. Although she suggested that the presence of an audience made
even more salient the participants' self-focus and corresponding memory
search for consistent information about themselves (Tice, 1992), Schlenker
et al. (1994) obtained evidence that directly contradicted this biased-
scanning view. Schlenker and coworkers asked participants to engage in
biased-scanning tasks in which they listed memories that were either
congruent or incongruent with their earlier self-portrayals. As it turned
out, the participants simply internalized their public self-portrayals, re-
gardless of the biased-scanning procedures.
It seems that the public commitment view offers the best-supported
explanation thus far for why people's identities may be undermined by
revealing too much to others. By revealing too much, they may constrain
themselves to being consistent with their earlier undesirable self-portrayals,
just as Jeff constrained himself to being the failure who could not hold on
to the love of his life.
These constraints associated with revealing too much may extend to
all domains of one's life, including therapy (as discussed in the previous
chapter) and the work setting. Revealing personal information in the work
setting may be threatening to one's identity or career if the revealed
information is contrary to how one wants to be seen at work. For example,
168 CHAPTER 8

if a woman makes the painful choice to announce to her co-workers that

she has been raped, then she may come to see herself as a victim because
she may perceive (either accurately or inaccurately) that her co-workers
see her as a victim. Or if she tells her colleagues that she has been punching
in early before her lunch hour is over, then the colleagues might report her
to the boss. As described earlier, most confidants probably cannot be
trusted to be completely discreet, especially about highly loaded or emo-
tional events.
Not only is this kind of exposure hard on most people, but also
rejection-sensitive individuals, or those who are attuned to and hurt by
negative evaluations from others, may be especially vulnerable to the nega-
tive effects of such exposure. Support for this idea comes from the same
researchers (Cole et a1., 1996a) who showed that being out of the closet was
associated with reduced risk for cancers and infectious diseases among a
sample of gay men (see Chapter 2). When these researchers studied a
sample of rejection-sensitive gay men in particular, they actually found the
opposite pattern. Specifically, among initially healthy HIV-positive gay
men, those who were rejection-sensitive experienced a significant acceler-
ation in how quickly they developed a critically low CD4 T-lymphocyte
level, an AIDS diagnosis, and died from the HIV infection (Cole, Kemeny,
& Taylor, 1997). However, the accelerated HIV progression was not ob-
served in the rejection-sensitive men who concealed their homosexual
identity, suggesting that being in the closet actually may protect such
individuals from negative health effects (Cole et a1., 1997). What is espe-
cially interesting about these findings is that the study was conducted in
Los Angeles, a place known for its large gay population and where gay
men may be accepted to a greater extent than they are in more socially
conservative cities. It could be that this buffering effect of staying in the
closet may be even more pronounced among rejection-sensitive men who
live in cities with a population that has greater anti-gay sentiments.

LISTENER FEELS BURDENED

Although the undermining of one's identity seems to be at the heart of

the problems associated with revealing secrets, a primary reason people
give for not sharing their traumatic or negative secrets is that they are
concerned that they will upset others if they do reveal their secrets to them
(Pennebaker, 1993; Pennebaker et a1., 1989). Coyne et a1. (1987) found that
people who lived with a depressed person reported that they were upset
by the depressed person's complaints of worthlessness and expressions of
worry. Furthermore, when people observe the distress of others, they often
respond with increased physiological and psychological arousal (Lazarus,
DILEMMAS TO REVEALING SECRETS 169

Speisman, Mordkof, & Davison, 1962; Leamer, 1980), sharp changes in

mood (Tannenbaum & Gaer, 1965), and unpleasant ruminations related to
the distress (Horowitz, 1975; Horowitz & Wilner, 1976; Wilner & Horowitz,
1975).
It seems that when telling traumatic secrets, the revealer becomes
more relaxed, whereas the listener becomes more physically aroused or
distressed (Pennebaker et al., 1989; Shortt & Pennebaker, 1992). A group of
Holocaust survivors who revealed their traumatic secrets experienced
decreased skin conductance levels (SCLs) as they talked, whereas the
undergraduates who were listening experienced increased SCLs (Penne-
baker et al., 1989; Shortt & Pennebaker, 1992). Harber and Pennebaker
(1992) argued that learning about a terrible trauma is difficult for the
listener because it can threaten the listener's assumptions about the order-
liness of the world. Listeners often react to such threats by attributing
personal responsibility for a trauma to the victim, or blaming the victim, to
lend credence to their own beliefs that the world is fair and just (Coates,
Wortman, & Abbey, 1979).
Unfortunately, the people who most desperately need supportive
feedback, such as those who are extremely depressed or who have suf-
fered a major loss, are the least likely to receive the support (Silver, Wort-
man, & Crofton, 1990). If victims of negative life events "maximize their
chances for personal adjustment by openly expressing their distress, they
may risk alienating their social network" (Silver et al., 1990, p. 401). Those
who express their struggles actually elicit more rejection from others than
do people who act as if they are coping quite well (Coates et al., 1979), and
people respond negatively to depressed individuals (see Gurtman, 1986,
for a review). For instance, in one experiment, depressed people elicited
depression, anxiety, hostility, and rejection from others with whom they
interacted for only 15 minutes (Strack & Coyne, 1983).
Most important, people tend to be avoided by confidants altogether
after revealing traumatic secrets to them (e.g., Coates et al., 1979; Lazarus,
1985). For example, cancer patients live with constant fear, but they do not
share their fear with family, friends, and health care staff because these
individuals do not respond well to such revelations (Spiegel, 1992). The
patients therefore end up withdrawing from others and feeling isolated
(Spiegel, 1992). This isolation is potentially extremely problematic for the
victim: Weaker social support has been found to be associated with less
protection from stress and from its related physical problems (see Broad-
head et al., 1983; Kessler & McLeod, 1985; Kessler, Price, & Wortman, 1985;
Levy, 1983; Mueller, 1980; Turner, 1983; Wortman, 1984; Wortman & Con-
way, 1985, for reviews). Moreover, as mentioned in Chapter 1, the need to
belong seems to be a fundamental human motivation (see Baumeister &
170 CHAPTERS

Leary, 1995). Any severe threat to one's network of social attachments is

likely to be both physically and psychologically detrimental to him or her.
In addition to fearing rejection, people also anticipate that others will
give unhelpful responses to their revelations, such as unwanted advice or
comments to the effect that the listener knows how the victim feels (Leh-
man, Wortman, & Williams, 1987; Pennebaker, 1993; Pennebaker et al.,
1989; Wortman & Lehman, 1985). A number of researchers have observed
that when people do disclose private information surrounding a trauma to
others, they tend to receive unhelpful comments (Davidowitz & Myrick,
1984; DiMatteo & Hays, 1981; Helmrath & Steinitz, 1978; House, 1981;
Lehman, Ellard, & Wortman, 1986; Lehman & Hemphill, 1990; Maddison &
Walker, 1967; Peters-Colden, 1982; Thoits, 1982; Wortman, 1984). Potential
confidants interrupt victims' disclosures and switch the topic of conversa-
tion to something other than the trauma. They also tend to impose upon
the victim their own interpretation of the trauma (Coates et al., 1979). In
one survey, people who experienced a death in their immediate families
were asked to report the kinds of responses they received from others
while they were grieving (Davidowitz & Myrick, 1984). Eighty percent of
the responses they reported were ones that they considered unhelpful;
the responses included such statements as, "You shouldn't question God's
will," and "Be thankful you have another son."
These types of responses are intended to discourage open discussion
and to encourage recovery, yet they actually isolate the victim, dismiss the
victim's feelings as being insignificant, and imply that the victim should be
getting over the trauma more quickly than the victim is (Lehman et al.,
1986). By encouraging someone in distress to look on the bright side, a
confidant may be conveying to the person that the person's feelings and
behaviors are not appropriate (Kessler et al., 1985). Moreover, when confi-
dants offer advice, they may be implying to the victims that the victims
are incapable of helping themselves (Brickman et al., 1982).
Despite the fact that confidants make such poor attempts at providing
support, people do know what responses would be helpful to hypothetical
victims (Lehman et al., 1986). However, they respond to the victim in ways
that dismiss the severity of the victim's distress in order to diminish
their own stress levels that have been generated by the victim's troubles
(Lehman et al., 1986). Cialdini et al. (1987) showed that even when under-
graduate participants were made to feel heightened empathy toward a
victim who was videotaped receiving electric shocks, the participants
apparently were selfishly motivated to reduce the distress that was associ-
ated with their feelings of empathy rather than to reduce the suffering of
the victim.
One limitation to the suggestion that it may not be wise to burden a
DILEMMAS TO REVEALING SECRETS 171

listener with one's secrets is that most of the studies cited in the previous
paragraphs have explored situations in which the confidant or listener
was someone who did not know the victim/secret keeper well. In Strack
and Coyne's (1983) depression research and in Cialdini and coworkers'
(1987) empathy experiments, for example, participants rejected strangers
whom they were told were depressed or victimized. In cases where a
confidant knows a victim well, or in cases where the confidant is highly
trained such as in a counseling context, the confidant may offer more
supportive feedback. In such cases, the confidant typically knows positive
aspects of the secret keeper to offset the negative secret or trauma, which
may help the confidant avoid rejecting the secret keeper. For instance, in
Coyne and colleagues' (1987) study, people who lived with a depressed
person felt distressed, but they continued to take care of the depressed
person.

REVEALING INFORMATION MAY MAKE IT MORE "REAL"

Another problem associated with revealing secrets is that people often

are not able to articulate their deep-seated feelings and motivations accu-
rately (see Nisbett & Wilson, 1977). Although people tend to know how
they feel, they frequently do not know why they have these feelings (Wil-
son, Lisle, & Schooler, 1988). When they do attempt to reveal their feelings
or problems, they may use cognitive explanations to describe these deep-
seated emotions and risk presenting distorted images of the feelings or
problems (Wilson, Dunn, Kraft, & Lisle, 1989). As such, their confidants are
not able to understand the concerns or offer appropriate responses follow-
ing the revelation. For example, a woman might tell her new boyfriend
that it is fine for him to see other women, just as she intends to see other
men. She may say that her reasoning is that she believes it is important for
people to enjoy life fully, whereas her real motivation may be that she is
worried that he will become bored with her sexually if he is constrained to
monogamy. However, because he does not understand her motivation, he
cannot respond appropriately and says "I wasn't raised that way and can't
be in relationship with a woman who isn't faithful."
Support for the notion that attempts to explain deep-seated feelings
can go awry comes from a study of dating relationships. Wilson and Kraft
(1993) found that students in a dating relationship who were asked why
their relationship was going either well or poorly described reasons that
were inconsistent with their actual degree of happiness in the relationship.
These students later changed their attitudes in the direction of their rea-
sons. There also is evidence that if people reveal their troubling symptoms
to others, those symptoms may become even more troubling and real
172 CHAPTERS

(Cioffi, 1996). Cioffi (1996) investigated the cognitive, behavioral, and so-
cial effects of having verbally expressed a somatic state and found that,
depending on how the symptom report is elicited and the context in which
it is made, the expression of a somatic state may change how that individ-
ual thinks about and remembers that state. Moreover, it affects how the
individual views himself or herself and how the individual is viewed by
others. In sum, people may need to be careful in sharing their private
concerns about a relationship as well as their somatic complaints, because
these revelations may make the negative emotional and physical experi-
ences more real.

SECRECY MAINTAINS PERSONAL BOUNDARIES

Some researchers have argued that secret keeping is actually healthy

and is an important component in the development of one's ego bound-
aries, or sense of identity (Hoyt, 1978; Margolis, 1966; Tausk, 1933). Learn-
ing about society's taboos (e.g., not to masturbate in public) and learning
to keep such information to oneself are thought to be central aspects of this
healthy development (Szajnberg, 1988). As mentioned in Chapter 1, Peskin
(1992) observed that as children mature, they learn to conceal information
from others and to use such information to influence or manipulate them.
Just as concealing may help people develop personal boundaries,
telling others personal information can lead people to lose their indepen-
dence and can leave them vulnerable to possible exploitation (Henley,
1973). In relationships, higher-status individuals know more about lower-
status others than lower-status others know about them (Henley, 1973).
Powerless groups, such as welfare recipients, mental patients, and pris-
oners, tend to have very little opportunity for privacy and often are targets
of inquiry (Derlega, 1988).
In romantic relationships, when both partners maintain clear personal
boundaries and keep some secrets from each other, they may create a
heightened sense of mystery and intrigue in the relationship. Olson, Bare-
foot, and Strickland (1976) demonstrated that, at least initially, keeping a
secret from a person enhanced attraction for that person. Participants
followed an opposite sex individual around and kept the person under
surveillance. Those participants who thought that the surveillance was not
known to the target, as compared with participants who thought that their
surveillance was known to the target, reported a higher attraction toward
the target (Olson et al., 1976). Wegner et al. (1994) also demonstrated that
keeping secrets about one's romantic relationship from other people can
enhance attraction to the romantic partner. These researchers showed that
mixed-sex couples who played a game of footsie with each other that they
DILEMMAS TO REVEALING SECRETS 173

kept secret from other couples, as compared with couples who played the
same game but who did not keep it a secret, reported a greater attraction
for their partners (Wegner et al., 1994). However, this experiment was not
designed to explore what happens to couples who keep secrets from each
other, nor was it designed to explore the long-term effects of secret keeping
on attraction. Although secret keeping may enhance attraction early on
(Olson et al., 1976), there is the possibility that secrets could undermine the
relationship in the long run by reducing the partners' trust in each other.
In a recent correlational study, Finkenauer and Hazam (2000) sur-
veyed married couples to find out how secrecy relates to marital satisfac-
tion. They asked married people to indicate to what extent they avoided
difficult topics in their marriages, or engaged in what the researchers
called contextual secrecy, and to what extent they suspected that their
spouses were keeping information from them. One of the items assessing
contextual secrecy was, "I avoid criticizing the way my partner treats me. II

The researchers found that contextual secrecy was positively related to

one's own marital satisfaction, whereas suspecting one's spouse of secret
keeping was negatively related to one's marital satisfaction. These find-
ings converge with those from Vangelisti's (1994) study, in which she
found that perceiving one's family to be high in secrecy, without actually
being higher in secrecy, was associated with lower satisfaction with one's
family.

SUMMARY

There are a number of negative consequences associated with reveal-

ing personal secrets. One of these consequences is that confidants may
repeat the secret to others, even if they are asked not to do so. Moreover,
victims of trauma fear that they will burden the listener, and in fact, they
are likely to receive unsatisfactory responses when they do relate their
traumatic experiences to others. This rejection and negative feedback
could lead people to construct negative identities for themselves, and
seeing the others in whom they confided may serve as a reminder of these
negative identities. In addition, when secret keepers do try to reveal their
secrets, they may not be able to articulate them in such a way that the
confidant will be able to understand the secret keeper or offer helpful
feedback. Finally, keeping secrets from others may help maintain healthy
personal boundaries and a sense of privacy that are associated with having
status or power in relation to others. One drawback to this research is that
most of it has explored rejecting feedback that comes from confidants who
are strangers. In the next section, I describe what happens when people
reveal their secrets to appropriate, helpful confidants.
174 CHAPTER 8

POSITIVE CONSEQUENCES OF REVEALING SECRETS

ALLOWS MEANINGFUL RESOLUTION OF SECRETS

As described in Chapter 5, if people share their secrets with others,

they may gain insights regarding the meaning of those secrets and may
develop a sense of control over their lives (Pennebaker, 1989, 1990; Penne-
baker & Hoover, 1985; Tait & Silver, 1989). They may find meaning in the
experiences by reframing them and assimilating them into their world-
views (Horowitz, 1975; Meichenbaum, 1977; Pennebaker et a1., 1988; Silver
et a1., 1983). For example, imagine a woman who feels enraged at her
husband because he has filed for divorce. She may decide to share those
feelings with an appropriate confidant, who responds in a helpful way by
reframing the anger as "a useful asset in making sure you will protect your
interests in the divorce settlement." Part of the trouble with hiding infor-
mation from others is that the secret keeper does not get the chance to hear
another person's perspective on an issue (Pennebaker, 1990). The secret
keeper may have a distorted perception of his or her problems and may
benefit from someone else's challenge to that perception (Pennebaker,
1990).
As described in Chapter 4, there is empirical support for the idea that
sharing secrets or traumas is beneficial in that it helps a person understand
the trauma (Foa et a1., 1995; Pennebaker et a1., 1990; Silver et a1., 1983).
However, revealing secrets does not just offer cognitive benefits, it also can
reduce shame and guilt or states of negative emotional arousal (Derlega,
1993; Stice, 1992). People who have experienced stigmatizing traumatic
events, such as rape and incest, feel shame, and thus they choose to conceal
these events from others (Derlega, 1993; Pennebaker, 1985, 1989; Silver et
a1., 1983). Unfortunately, as Pennebaker (1985) suggested, "the act of not
discussing or confiding the event with another may be more damaging
than having experienced the event per se" (p. 82). The victims may tell
themselves that because of the fact that they have hidden the experience
from others, the event must indeed be very negative or shameful (Derlega,
1993). As described in Chapter 3, it is possible that through this self-
perception process (Bern, 1967, 1972), people may develop feelings of
lowered self-worth (Derlega, 1993). It seems that not only having a painful
secret but also working to hide the secret can make people feel bad about
themselves.
In cases in which someone feels guilty about having betrayed some-
one else, such as when one marital partner has had a hidden affair, the
person may feel a tremendous sense of relief upon finally sharing the
betrayal with his or her spouse. Theoretically, this relief may occur even
DILEMMAS TO REVEALING SECRETS 175

if the revelation results in the breakup of the marriage. In one study, when
people who felt guilty about their actions were given the opportunity to
talk about these behaviors, they reported experiencing relief from their
guilt feelings (Regan, 1968). Similarly, in another study, when people were
permitted to confess their guilt-inducing behaviors, they subsequently
exhibited fewer remorseful behaviors (Carlsmith, Ellsworth, & Whiteside,
1968). However, these studies have not examined the long-term effects of
revealing guilty secrets, and they leave open the possibility that feelings
of regret could emerge at some time after revealing the secret.

AVOIDS SUSPICIONS OF RECIPROCAL SECRECY

Social psychologists have long documented the phenomenon wherein

people see others as being more similar to them than they actually are and
called it the false consensus effect (see Ross, Greene, & House, 1977). Borrow-
ing from Sagarin, Rhoads, and Cialdini's (1998) theorizing about the nega-
tive relational effects of lying, I suggest that keeping a secret from someone
may damage the relationship even if the secret is never discovered, be-
cause the secret keeper may assume that person is keeping a secret in
return.
Support for this idea can be gleaned from Sagarin and coworkers'
(1998) recent study in which they induced some undergraduates to lie to a
partner about how they were able to solve a difficult puzzle, whereas other
undergraduates simply told the partner the truth about their puzzle-
solving strategies. In addition to this manipulation, half the participants
in the deception condition were led to believe that their lies hurt their
partners by preventing them from receiving extra credit, whereas the other
participants were told that their partners received the extra credit. All the
participants later were asked to rate the partners and the typical student
on various traits including honesty. The participants who had lied to their
partners, as compared with participants who had not lied, later rated the
partners as less honest, particularly when the lie damaged the partners.
These deceptive participants also rated the typical student as less honest.
The authors explained the denigration of the partners by indicating that
the participants in the lying condition were justifying their actions in a
self-protective version of the false consensus effect, wherein they justified
their dishonesty by seeing others as being similarly dishonest. However,
as the researchers themselves pointed out, this study involved newly
formed dyads. In long-term relationships, people who lie to their partners
may continue to see their partners as honest both because they care about
them and because they have had many opportunities to observe how
honest their partners are. Plus, romantic couples tend to assume that their
176 CHAPTER 8

partners are honest (Derlega et al., 1993), unless they are given evidence to
the contrary (see Cole, 2001).

MAKES SECRETS SEEM LESS HARMFUL

A benefit of revealing secrets is that they may seem less negative than
if they are first concealed and then later discovered. As mentioned in
Chapter 3, Fishbein and Laird (1979) showed that participants who re-
vealed ambiguous information, as compared with those who concealed it,
viewed that information as less negative. Moreover, people generally view
openness as a sign of closeness (see Derlega et al., 1993) and may get angry
at others for keeping a secret from them. Imagine the circumstance in
which a woman named Elizabeth conceals the fact that she is involved
romantically with her friend Mary's ex-boyfriend. Elizabeth conceals this
information from Mary, even though she knows that Mary is no longer
interested in the ex-boyfriend and would not mind if Elizabeth dated
him. Elizabeth also realizes that there is a high probability that Mary
already knows about her involvement with the ex-boyfriend because they
have many mutual friends. Concealing that information is likely to be a
signal to Mary that Elizabeth is not close to her, which may be more
upsetting than the fact that Elizabeth is seeing her ex-boyfriend.
This phenomenon wherein personal information seems less negative
when it is out in the open may help explain why people sometimes are
willing to say astonishingly undesirable things about themselves, even
though they are motivated to be seen in desirable ways. For example, years
ago, I knew a married man who openly told his friends that he wears
women's undergarments because they feel good against skin and arouse
him. Even though he seemed to be giving the impression that he was
impervious to what his friends thought, at some level he may have viewed
his revelation as increasing the likelihood that he would be seen in a
desirable way, as an open person who was so confident that he did not care
what others thought about him. Indeed, in his immediate circle, his open-
ness seemed to make the behavior acceptable, and his friends viewed him
as a quirky, likable person. However, when his friends described his
behavior to people outside his circle, they simply would describe his cross-
dressing behaviors and omit the fact he himself was the relaxed and
confident source of the information. The people outside his circle simply
carne to see him as "that pervert who wears women's lingerie," rather than
as he wanted to be seen (Le., as someone who is open, confident, and
comfortable with his sexuality). The point I am making here is that people
may find the process of revealing information to be reinforcing in the
DILEMMAS TO REVEALING SECRETS 177

immediate context, but as time goes on and more people know about the
information, its meaning may change from desirable to undesirable.
In sum, given Fishbein and Laird's (1979) finding that concealed
information seems more negative than revealed information, perhaps peo-
ple should be open about their transgressions early in relationships, so
that the transgressions are never a secret. However, people need to know
their confidants somewhat well before taking that risk because the confi-
dants might reject them outright as a result of the revelation.

MAY ENHANCE LIKING

In the self-disclosure literature, meta-analyses have uncovered a sta-

tistically significant disclosure-liking effect: People who disclose personal
or intimate information tend to be more liked than people who disclose at
lower level (Collins & Miller, 1994). This effect is qualified by the content
and appropriateness of the disclosure, such that disclosures that are too
intimate and offered too soon may lead to perceptions of the discloser as
maladjusted and less likable (Altman & Taylor, 1973; Collins & Miller,
1994). In general, there seem to be fairly strict norms about what is appro-
priate to reveal in various contexts (Derlega & Grzelak, 1979). It seems
that disclosure will not lead to liking at extreme levels of intimacy (Archer
& Berg, 1978; Brewer & Mittelman, 1980; Cozby, 1972). When disclosures
are too intimate and offered too soon, people may feel obligated or pres-
sured to reveal at an equally intimate level. Therefore, Cozby (1972) sug-
gested that there is a curvilinear relationship between disclosure and
liking, such that moderate levels of disclosure should bring the greatest
liking. Along these lines, in their meta-analyses, Collins and Miller (1994)
found that the strongest links between self-disclosure and liking occurred
in correlational or naturalistic studies, as opposed to experimental studies.
They proposed that this was because lithe former were almost always
concerned with disclosure processes in established relationships, whereas
experimental studies were primarily concerned with disclosure and liking
between strangers" (p. 470). It seems that high levels of disclosure to
strangers backfired because too much information was revealed too soon.
Ajzen (1977) suggested that the reason disclosure is positively related
to liking is that liking is determined by holding positive beliefs about
another person. People who disclose more intimately, as compared with
those who disclose less, may be viewed by others as more trusting,
friendly, and warm. As logic would dictate, the more positive the beliefs,
the greater the liking (Ajzen, 1977; Dalto, Ajzen, & Kaplan, 1979). Re-
searchers have shown in a number of experiments and correlational
178 CHAPTER 8

studies that people form more favorable impressions of others who are
willing to reveal personal information about themselves (e.g., Davis &
Sloan, 1974; Jones & Archer, 1976; Kleinke & Kahn, 1980). For example, in
one study, group therapy members who self-disclosed more, as compared
with those who disclosed less, were rated as more likable by both the
therapists and other group members (Weigel, Dinges, Dyer, & Straum-
fjord, 1972).

SUMMARY

Some of the main benefits of revealing secrets are that it can alleviate
the stress and guilt associated with hiding information from others, pro-
vide the opportunity to receive new insights or perspectives on the secrets,
make the secrets seem less harmful, and make the revealer more likable.
These positive consequences help explain why people so often reveal
undesirable hidden information about themselves.

CONSEQUENCES OF REVEALING
MAY DEPEND ON THE CONFIDANT

Do the benefits of revealing secrets outweigh the costs of revealing in

most circumstances? I suggest that such outcomes should hinge on the
feedback from and qualities of the confidant. Consistent with this idea,
Macdonald and Morley (2001) found that the anticipation of negative
responses to disclosure, particularly labeling and judging responses, was
associated with not disclOSing emotional events to other people in a diary
study involving psychotherapy clients. In this section, I describe the re-
search on how the confidant may influence the outcomes of revealing
potentially stigmatizing secrets, beginning with the example of sexual
orientation.
Gay men who live around rejecting people or in a stigmatizing culture
may not have as positive self-perceptions as they would otherwise. Frable,
Wortman, and Joseph (1997) asked a large sample of predominantly white,
young, educated, middle-class gay and bisexual men living in the Chicago
and surrounding areas to complete a 90-minute self-administered ques-
tionnaire that included self-esteem, well-being, and symptomatology mea-
sures. The men answered questions about their experiences with gay
stigma, visibility as gay men, involvement in the gay community, and
commitment to a positive gay identity. The gay men in this sample, as
compared with nonstigmatized samples in general, were neither partic-
ularly low in global self-esteem nor high in psycholOgical distress on the
DILEMMAS TO REVEALING SECRETS 179

average. However, the ones who lived around people who stigmatized
gay men, as compared with those who did not, had more negative self-
perceptions.
Along those same lines, Frable, Platt, and Hoey (1998) asked 86 Har-
vard undergraduates to keep a diary over 11 days and rate their momen-
tary self-esteem and affect. Those participants with concealable stigmas
(students who indicated that they were gay, that they were bulimic, or that
their family earned less than $20,000 each year), as compared with both
those whose stigmas were visible and those without stigmatizing charac-
teristics, reported lower self-esteem and more negative affect. However,
the students with concealable stigmas who had contact with similar others
were the least likely to experience low self-esteem and depressed mood.
Thus, Frable et a1. (1998) concluded that contact with similar others may
protect individuals with hidden stigmas from negative cultural messages.
When comparing 150 lung cancer patients with matched control pa-
tients from chest disorder wards, Kissen (1966) found that the cancer pa-
tients showed a significantly diminished outlet for emotional discharge or
had fewer opportunities for discussing their feelings with others. Kissen
(1966) concluded that among heavy cigarette smokers, a poor outlet for emo-
tional discharge was as important for cancer development as urban resi-
dence and even more important than working in an air-polluted setting.
However, having an outlet for emotional discharge in response to a
stressor may be helpful only when those people offering such an outlet are
supportive. Major et a1. (1990) interviewed women prior to their having a
first trimester abortion and assessed their perceptions of social support
from their partner, family, and friends. They also measured the women's
depression, mood, physical complaints, and anticipation of negative con-
sequences just after a 30-minute postabortion recovery period. They found
that the women who had told close others of their abortion but perceived
them as less than completely supportive, as compared with women who
had not told or who had told and perceived complete support, had poorer
postabortion psychological adjustment.
Bolger, Foster, Vinokur, and Ng (1996) also challenged the commonly
held belief that when life crises occur, significant others help to alleviate
distress and resolve practical problems. They interviewed 102 breast can-
cer patients and their significant others at 4 and 10 months after the breast
cancer diagnosis. It seemed that this life crisis may have overwhelmed the
significant others and detracted from their ability to provide effective
support to these women. Although the significant others provided sup-
port in response to the women's physical impairment, they withdrew
support in response to the women's emotional distress. The result of this
pattern of support from significant others was that it did not alleviate the
180 CHAPTERS

women's distress nor did it enhance their physical recovery. The re-
searchers concluded that in times of extreme stress there are limits to the
effectiveness of close relationships.
Researchers also have studied how social constraints on discussion
of a traumatic experience can interfere with cognitive processing of and
recovery from loss (Lepore, Silver, Wortman, & Wayment, 1996). Mothers
who had recently lost an infant to sudden infant death syndrome (SIDS)
were interviewed at 3 weeks, 3 months, and 18 months after their infants'
death. Part of the interviews involved questions about the mothers' experi-
encing intrusive thoughts surrounding the death. Lepore et al. (1996)
defined intrusive thoughts as recurring, unwanted memories, thoughts,
and images of a stressor, and they conceptualized such thoughts as a sign
that people are trying to make sense of the stressor. The mothers who had
experienced social constraints to discussing the death (i.e., felt that others
were uncomfortable, not fully supportive, or not willing to listen when
they talked about the loss) were compared in their emotional adjustment
to the women who had not experienced such social constraints. Inter-
estingly enough, among socially constrained mothers, the more the women
had experienced intrusive thoughts at 3 weeks after the loss, the less they
talked about the infant's death at 3 months and 18 months after the loss.
The reverse associations were found among unconstrained mothers.
Moreover, when the researchers conducted analyses that statistically con-
trolled for the women's initial levels of distress, there was a positive
relation between intrusive thoughts at 3 weeks and depressive symptoms
over time among socially constrained mothers. However, higher levels of
intrusive thoughts at 3 weeks were associated with a decrease in depressive
symptoms at 18 months among mothers with unconstrained social rela-
tionships. It seems that the cognitive processing these women experienced
through intrusive thoughts about the death was only helpful when the
women had supportive people to talk to about the death (Lepore et al.,
1996).
This same pattern was found in studies on women with breast or colon
cancer (Lepore, 1997a), men with prostate cancer (Lepore & Helgeson,
1998), and children exposed to inner-city violence (Kliewer, Lepore, Oskin,
& Johnson, 1998). Lepore (1997b) suggested that, given these findings,
expressing one's emotions surrounding stressful events may facilitate cog-
nitive processing and emotional adjustment only when this expression
occurs in a safe or supportive environment.
In an even more recent study, Lepore, Ragan, and Jones (2000) used an
experimental design to assess the influence of talking and the social con-
text of the talking on cognitive-emotional processes of adjustment to
stressors. Two hundred fifty-six undergraduates (half men and half women)
DILEMMAS TO REVEALING SECRETS 181

viewed a stressful 14-minute visual presentation on the Nazi Holocaust

and then were assigned to one of four conditions. These were a no-talk
control condition or one of three talk conditions: talk alone, talk to a
validating confederate, or talk to an invalidating confederate. In the talk
to a validating confederate condition, the confederate nodded, maintained
eye contact, and smiled approvingly while the participant disclosed feel-
ings and reactions to the film. The confederate also reciprocated disclosure
and identified and agreed with the reactions of the participant. In contrast,
in the invalidating confederate condition, the confederate avoided eye
contact while the participant disclosed and when disclosing, disagreed
with several thoughts and feelings expressed by the participant. Two days
later, the participants were reexposed to the stressful film. As it turned out,
participants in the talk alone and validate conditions, as compared with
those in the no-talk condition, had a lower level of intrusive thoughts in
the 2-day interim and they reported experiencing less stress when re-
exposed to the stressor. Lepore et al. (2000) observed that it seemed to be
through experiencing fewer intrusive thoughts of the stressor that talking
and validation lowered participants' perceived stress. At the same time,
the participants in the invalidating confederate condition did not differ
from those in the other three groups on the intrusive thoughts or distress
measures. The researchers suggested that the reason this last finding
seems to contradict those of Major et al. (1990) is that the confederate in this
study was a peer and the participants may have dismissed the invalidating
feedback, whereas the people providing support in the Major and co-
workers' study were significant others, and thus their unsupportive feed-
back may have been more damaging. Lepore et al. (2000) concluded that
talking about acute stressors can facilitate adjustment to the stressors
through gaining cognitive resolution on the stressor, a process discussed in
detail in Chapter 4.
In sum, the benefits associated with trying to make sense of a private
disturbing event, such as the death of one's child or the diagnosis of cancer,
seem to hinge on supportiveness of one's confidants. Next, I illustrate
that many of the dilemmas to revealing secrets described in this chapter
may be resolved by carefully choosing an appropriate confidant.

FEATURES OF HELPFUL CONFIDANTS

What makes an appropriate confidant or one to whom secret keepers

would benefit from revealing? I suggest that if a troubled secret keeper
has a confidant (1) who is discreet and can be trusted not to reveal a secret,
(2) who is perceived by the secret keeper to be nonjudgmental, and (3) who
182 CHAPTER 8

is able to offer new insights into the secret, then the secret keeper should
reveal to that person (see Kelly, 1999; Kelly & McKillop, 1996). The rationale
for this recommendation is that having such a confidant would allow the
revealer to gain potentially invaluable insights into the secret as well as
avoid the negative consequences of revealing secrets described earlier. My
rationale for depicting each of these characteristics as essential in a confi-
dant is described in more detail in the following paragraphs.

DISCREET

Researchers have demonstrated that assuring people of the confiden-

tiality of their disclosures, as compared with not offering such assurances,
generally leads to greater disclosure of private information (Corcoran,
1988; Kobocow, McGuire, & Blau, 1983; Woods & McNamara, 1980). Trust-
ing the confidant not to repeat the secret to others may be the most
important factor in determining whether someone should reveal to that
person. When a person can be assured that only the confidant will know a
secret, then that person can assess how the confidant is reacting and can
make a decision about whether to continue the relationship with the
confidant. In contrast, in situations in which confidentiality is not guaran-
teed, the person loses this sense of control and may be victimized by others
who are privy to the secrets (Henley, 1973). In addition, the person may
begin to construct a negative identity for himself or herself by imagining,
or actually hearing, the judgments others will make upon learning about
the secret (see Schlenker, 1986). Given these negative possibilities, I recom-
mend that a secret keeper avoid revealing secrets to any confidants who
cannot or will not refrain from transmitting the secrets to others.

NONJUDGMENTAL

Pennebaker (1990) has stated that people are most likely to disclose
their deepest secrets if they think that others will not judge them and will
accept them no matter what they say: "Pennebaker's research on self-
disclosure highlights the importance of a confidant who will listen without
judging or withdrawing love and support" (Kennedy et al., 1990, p. 262).
The idea that it is important for a confidant to be nonjudgmental is backed
by Carl Rogers's (1951, 1957) theoretical work on the benefits of having an
accepting therapist or close friend with whom one can sort out personal
problems. Rogers (1951, 1957) proposed that the primary role of the thera-
pist is to provide clients with a safe environment (i.e., one free from
conditions of worth or judgments from the therapist) so that the clients can
learn to trust their inner experiencing. He believed that if the therapist
DILEMMAS TO REVEALING SECRETS 183

is genuinely empathic (expresses understanding) toward the client and

offers the client unconditional positive regard (warmth and acceptance),
then the client will develop his or her self-actualizing or growth potential.
Truax and Carkhuff (1967) and Truax and Mitchell (1971) conducted
reviews on studies assessing Rogers's theoretical ideas about the thera-
pists's role and concluded that there was strong evidence that therapist
empathy, warmth, and genuineness were important in inducing client
improvement. However, Lambert, Dejulio, and Stein (1978) subsequently
pointed out that there was no empirical evidence that the facilitative
features of the therapist actually caused client change. In addition, more
recent work suggests that these features of a therapist may be more
relevant for mildly disturbed individuals than for severely disturbed pa-
tients (Lambert & Bergin, 1994). Despite the fact that the debate concerning
whether Rogers's proposed facilitative conditions do induce client change
remains unresolved, Rogers's theoretical ideas have had a tremendous
impact on the counseling field (Hill & Corbett, 1993). For example, most
counseling psychology graduate programs now train their students to
use the basic empathy skills (e.g., reflection of feelings, restatement of
content) proposed by Rogers (Hill & Corbett, 1993).
The reason that having a nonjudgmental confidant is so helpful in the
context of revealing secrets is that such a confidant provides all the benefits
associated with unburdening described above (e.g., the health benefits,
new insights about the secret) but none of the ramifications associated
with receiving negative feedback (e.g., being rejected by and becoming
isolated from others). In sharing with nonjudgmental others, the secret
keeper can feel understood and accepted and can feel comforted by the
fact that he or she is no longer carrying the secret burden alone (Yalom,
1985). For instance, if a young man tells his therapist that he accidentally
killed a pedestrian while driving drunk and if she conveys that she accepts
him despite his actions, he can feel relieved from his guilt. He can continue
to construct a positive self-image by seeing himself through her eyes,
rather than through the critical eyes of imagined others.

ABLE TO OFFER NEW INSIGHTS

One problem with telling a secret down a well as John Steinbeck

described is that the well does not offer new helpful insights on the
troubling secrets. I suggest that a key factor in making a wise decision to
reveal to a confidant is whether that confidant is able to offer new insights
or perspectives on a problematic secret. As described in the previous
chapter, my students and I (Kelly et al., 2001, Study 1) examined under-
graduates' reports of what they had gained from revealing their most
184 CHAPTER 8

private secrets to their confidants in the past. Gaining new perspectives on

their secrets was associated with feeling better about the secrets now,
whereas experiencing catharsis actually was negatively associated with
recovery from the secrets. In addition, participants' ratings of the expert-
ness of their confidants (i.e., their preparedness and their ability to help)
were predictive of the participants' gaining insights into their secrets. In a
second study, we found that participants who wrote about their secrets
while trying to gain new insights into them, as compared with participants
who tried to gain catharsis during their writing or who wrote about the
details of their day, felt much better about their secrets after 1 week (Kelly
et al., 2001, Study 2). Thus, I suggest that if the secret keeper has a confidant
who is particularly insightful (such as highly skilled and well-trained
counselor) and is able to provide new perspectives on the secret, then
the person should reveal to that confidant.

EVIDENCE FOR THESE THREE FEATURES

Further support for the argument that these three qualities are impor-
tant in a confidant comes from that same research I conducted with my
students (Kelly et al., 2001, Study 1). The undergraduate participants were
provided with brief deSCriptions of six features of a hypothetical confidant
(i.e., someone to whom they would be willing to share their most personal
secrets) and asked them to rank these features in the order of importance.
Specifically, participants were asked to rank six qualities according to
"how important they are in determining whether or not you would tell
someone your secrets in general." The qualities were: "won't judge me,"
"understands me," "is able to help me," "will keep my secret," "is similar
to me in personal characteristics," and "has had an experience that was
similar to my secret experience." This list was not meant to be exhaustive
of all the desired qualities in a confidant but rather to include a number of
the key confidant qualities described or alluded to in the social support
literature (Barak & LaCrosse, 1975; Davidowitz & Myrick, 1984; DiMatteo
& Hays, 1981; Helmrath & Steinitz, 1978; House, 1981; Kessler et al., 1985;
Lehman et al., 1986; Lehman & Hemphill, 1990; Maddison & Walker, 1967;
Peters-Golden, 1982; Thoits, 1982; Pennebaker, 1993; Pennebaker et al.,
1989; Wortman & Lehman, 1985). Participants rated "will keep my secret"
and "understands me" as the most important features of a confidant. The
third most important feature of a confidant was "will not judge me,"
followed by "is able to help me," "is similar to me in personal characteris-
tics," and "has had similar experiences." In essence, having a discreet
confidant and being understood (and not judged) were deemed as more
important than having a similar confidant.
DILEMMAS TO REVEALING SECRETS 185

Those same undergraduates were asked to provide descriptions of

whether they gained catharsis or new insights from revealing to their
actual confidants in the past and to describe how trustworthy, expert, and
socially attractive (i.e., warm and likable) these confidants were. They also
were asked to indicate among these features of expertness, trustworthi-
ness, and social attractiveness, which was the most important to them in
deciding whether to reveal secrets to a hypothetical confidant. We picked
these three features because counseling and clinical psychologists have
long identified these as being important qualities in a therapist in terms of
getting psychotherapy clients to reveal themselves (see Strong, 1968). As it
turns out, trustworthiness was rated as the most important feature of a
hypothetical confidant. However, participants' ratings of the expertness,
not the trustworthiness or attractiveness, of their actual confidants signifi-
cantly predicted participants' gaining new insights into their secrets. As
such, the findings indicate that there may be a discrepancy between what
people say they want in a confidant and what is most helpful to them.
Specifically, people are apt to value trustworthiness in a confidant, but a
confidant who is expert (i.e., prepared and able to help) may be more likely
to offer new insights into a secret, which may be most critical in recovering
from the secret. As described earlier, it is through gaining new insights that
one might expect the expertness of the confidant to be associated with
recovery from secrets.

OPTIMAL NUMBER OF CONFIDANTS

Now that I have described the characteristics of the ideal confidant,
the following question remains: How many confidants are appropriate?
Stokes (1983) discovered that seven confidants may be the optimal number.
He asked undergraduates in an interviewing course to solicit four partici-
pants each and to give them assessments of the extensiveness of their
social networks and their satisfaction with these networks. Stokes found
that satisfaction with the social networks increased up to having seven
confidants, but after that having more confidants was not associated with
increased satisfaction. He speculated that the added support of having
more confidants may not be worth the costs of maintaining the relation-
ships and offering social support in return.

CONCLUSION
There are a number of positive and negative consequences associated
with revealing stigmatizing secrets. The negative consequences include
186 CHAPTER 8

damaging one's social reputation, being rejected by the listener, and form-
ing a negative opinion of oneself by knowing that others are aware of the
stigmatizing information. The positive consequences include hearing
other people's insights on the secret, becoming more well liked, and
relieving oneself of the guilty burden of secrecy. I have suggested that the
outcomes of revealing stigmatizing personal secrets hinge mainly on the
responses of the confidant, although researchers only recently have begun
to examine the role of the confidant and more evidence is needed to
support this idea. The research that is available indicates that when reveal-
ing a highly emotional or private secret, people do not adequately keep in
mind that others often cannot be trusted to refrain from repeating the
secret. Moreover, given that what others think matters a great deal to
people and that there seems to be a fundamental need to belong, revealers
may pay a price in terms of their identity and well-being. As a way of
dealing with the dilemmas to revealing, I have suggested that secret
keepers carefully select their confidants and search for one who is discreet,
nonjudgmental, and able to offer new insights into the secrets. In the next
chapter, I offer an even more complex analysis of when to reveal secrets
that includes a discussion of whether the confidant expects access to the
hidden information and whether he or she will discover the information
anyway.
CHAPTER 9

WHEN TO REVEAL PERSONAL

SECRETS IN A PARTICULAR
RELATIONSHIP

This final chapter opens with one more movie example that I think may
have sent a counterproductive message to the public about how protecting
one's privacy relates to one's integrity. In the 2000 movie, "The Contender,"
the vice president dies, and the president appoints a replacement, Laine
Hanson (played by Joan Allen). But a political opponent in charge of
approving the new appointee dredges up a history of sexual misconduct
on her part, threatening to destroy her political career and tarnish the repu-
tation of the current administration. Andrew Manning wrote in his October
2000 review that the ending is" completely unrealistic-it involves a politi-
cian being so honest that they're willing to ruin themselves for their ideals,
and I find that highly unbelievable in a field where deception and pandering
are some of the cornerstones of the career" (http://www.rottentomatoes.
com/ click/ movie-l100998/ reviews.php ?critic=approved&sortby=default
&page=1O&rid=156099. Retrieved 3/16/01).
What Manning is describing here is that Hanson never denies that she
is the woman in the pictures (taken when she was in college) having sex
with two men as the centerpiece of a fraternity party, even though she can
easily disprove the accusation by demonstrating that birth marks on her
leg do not match the pictures of the woman. Her rationale (revealed to the
president at the end of the movie) is that she should not even be asked such
questions because they are a matter of privacy. My suggestion, however,
would be to deny such rumors promptly and then become outraged at
the inappropriateness of such questions. The same message of personal
187
188 CHAPTER 9

integrity can be delivered without letting people draw highly undesirable,

false conclusions. I say this because just thinking that everyone is making
negative judgments about one can undermine one's very identity.
Of course, that was a very straightforward, easily handled example of
what to do with a troubling secret. My aim in this chapter is to offer sug-
gestions for dealing with the sometimes very complicated situations in-
volving secrecy that many people confront in their various relationships
on a daily basis. My aim is to help people lead more physically and
psychologically healthy lives; that is how central secrecy is to their exis-
tence. I encourage people to conceal and reveal information judiciously.
In the previous chapters, I described circumstances in which reveal-
ing personal secrets can lead to benefits and circumstances in which it can
lead to detriments for the secret keeper. I also pointed out that the fact that
high self-concealers as a group are sicker than low self-concealers may not
be the result of their keeping a particular secret but rather may be the result
of their being born with a genetic predisposition to be vulnerable to
various illnesses. The way I see it, although self-concealers are at a dis-
advantage to start with, both high and low self-concealers may reap great
benefits from judicious revelation. I base this proposition on the assump-
tions that the self-concept has an interpersonal element and that it can
change over time (Baumeister, 1998). Telling someone a very stigmatizing
personal secret, as compared with keeping the secret, offers greater poten-
tial for negative repercussions for one's identity and health. In the follow-
ing pages, I embark on addressing the one major question on secrecy that
remains unanswered: When should people reveal their personal secrets in
particular relationships and take the risk of being stigmatized or rejected?
This chapter takes the discussion from the previous chapter a step further
by adding the context of a specific relationship to the complex decision to
reveal.
An acquaintance of mine named Gary (not his real name) asked me
for insights on an event that upset him years earlier and still perplexed
him. He went to a wedding reception with his then girlfriend (who,
incidentally, had a doctoral degree in psychology). When she happened to
see a former lover at the reception, she whispered, "I had sex with that guy
in college." Gary told me that his night was ruined by that one remark,
which he viewed as either unthinkably cruel or just plain stupid. Shortly
thereafter, he broke off the relationship with her because he perceived her
as an insensitive person.
Just as in the example described in Chapter 1, in which Diane Zamora's
confessions to her Naval Academy roommates seemed unbelievably fool-
ish or not well thought out, this revelation too seems puzzling. Perhaps
part of what the example of Gary's girlfriend boils down to is that people
WHEN TO REVEAL PERSONAL SECRETS 189

are motivated to protect their reputations, and a strategy for doing so is

to give others one's own account of events that may make them look bad
(Emler, 1990). Although social psychologists (see Emler, 1990) have argued
that people normally underestimate the interconnectedness of their social
networks, in this situation Gary's girlfriend may have overestimated the
interconnectedness of her network and revealed something that Gary was
not likely to discover on his own. She may have feared that if he did
discover the information, he would feel betrayed by an omission of infor-
mation that seemed relevant to their relationship. Did she really make
the wrong decision as Gary's recounting of events implies? In this chapter,
I offer a model for when to reveal or conceal hidden information (such as
previous sexual experiences, masturbatory habits, extrarelational affairs,
or hygiene rituals) in a particular relationship, and in so doing I address
this and a variety of related questions that are pivotal to maintaining
relationships as well as a sense of personal well-being.
The decision to reveal personal secrets within a particular relationship
may most aptly be described as a trade-off between unburdening oneself
of the stress and guilt of keeping troubling secrets from the relational
partner on the one hand and constructing desirable images of oneself as
well as avoiding rejection on the other hand. As discussed in the previous
chapters, the essence of the problem with revealing personal, undesirable
information is that revealers may come to see themselves in undesirable
ways if other people, especially those in their immediate social networks,
know their stigmatizing secrets.
The primary focus of the model for when to reveal hidden informa-
tion is to help individuals construct favorable self-images, and thus en-
hance their well-being. The model also is intended to help them become
more aware of their decision-making processes and recognize that they
can control the degree of personal access they let other people have,
without deception. With every step of the model, an important option is to
keep the potentially stigmatizing information hidden to avoid the risks to
their self-images associated with revealing such information (see Figure 1).
Certainly, individuals may have additional motives besides constructing
favorable views of themselves when they decide to reveal hidden informa-
tion, such as deliberately wanting to get fired, become divorced, or make
someone else feel superior. The model does not address these motives.
Furthermore, the model does not necessarily address the broader moral
implications of keeping or revealing a secret (see Bok, 1982, for a discussion
of the moral implications). At the same time, however, the model may be
considered morally sound because potential secret keepers and their rela-
tional partners are both likely to benefit from it. It encourages them to
avoid getting into situations that will lead to the development of troubling
190 CHAPTER 9

Is the hidden information

secret (vs. private)?
Other person expects
access

Yes
...-- --"NO

+ +
Is the other person an Keep hidden
appropriate confidant?
discreet
from other person
nonjudgmental
will not reject revealer

Yes No
+ +
Reveal secret Is the other person
to other person likely to discover the
secret?

Yes

+
Reveal secret (focusing Is the secret troubling in
on themes) OR the relational context?
End relationship ruminations OR
anxiety, depression, ulcers,
headaches, back pain

~
Yes No

Reveal secret (focusing

+ +
Keep the secret
on themes) OR from other person
Write it down OR
End relationship
FIGURE 1. When to reveal personal secrets in a particular relationship.

secrets by making them think about the relational consequences of their

secretive actions. It also encourages them to avoid revealing information
that their partners cannot handle emotionally or do not need to know.
Even though some of the steps in the model may seem obvious to the
reader, it is my contention that many people in American society have lost
WHEN TO REVEAL PERSONAL SECRETS 191

sight of the importance of personal boundaries and have been lured into
revealing too much information to their partners in an effort to be open
and honest when building their relationships. By offering this model, I
remind people of the dangers of too much revelation and provide appro-
priate options to complete openness.

PROPOSED MODEL FOR WHEN TO REVEAL

The first step of the model for when to reveal secrets involves deciding
whether the stigmatizing hidden information is secret versus private. This
decision hinges on whether the relational partner expects access to the
hidden information. If the partner does not expect access to that informa-
tion, then it is private and may be kept hidden. If the partner does expect
access, however, then it is a secret, and the second step is to assess carefully
whether he or she is an appropriate confidant. As described in the previous
chapters, there is empirical support for the idea that if a troubled secret
keeper has a confidant who can be trusted not to reveal a secret, is perceived
as nonjudgmental, and is able to offer the secret keeper new insights into
the secret, then the secret keeper should reveal to that person (see Kelly &
McKillop, 1996). However, in the context of a particular relationship, these
characteristics of an appropriate confidant need to be modified. Specifi-
cally, the insightfulness of the partner may not be as crucial as whether he
or she will reject the revealer and end the relationship. My rationale for this
claim is that the revealer can get new insights into the secret from other
confidants and may not need them from a particular partner. Thus, if the
partner is discreet, nonjudgmental, and not rejecting of the secret keeper,
then the secret keeper should reveal the secret to him or her. The third step
of the model is to consider whether the relational partner is likely to dis-
cover the secret on his or her own. In cases where the probability of discov-
ery is high, it may benefit the secret keeper to disclose the secret, focusing
on the themes as opposed to the details (see Kelly, 2000b). This focus on
themes can help the revealer be honest while avoiding the stigma and
humiliation that often accompany the revelation of details. If the partner is
not likely to discover the secret, the fourth and final step is to determine
whether the information is troubling. Because of the risks to one's identity
associated with revealing stigmatizing personal secrets to inappropriate
confidants, I content that people only should contemplate revealing such
information if keeping the secret seems to be causing them internal stress
and negative effects described by various researchers; in other words, if
the secret is troubling (e.g., Lane & Wegner, 1995; Larson & Chastain, 1990;
Pennebaker, 1989, 1990; Wegner, 1989, 1992, 1994). Implied in this last step is
my contention that secret keeping per se is not necessarily problematic.
192 CHAPTER 9

Note that this model concerning when to reveal secrets is different

from the one described in Kelly and McKillop (1996), in which we offered
suggestions for when to reveal personal secrets, such as a childhood
molestation experience, that a relational partner mayor may not expect to
be told. This model, in contrast, includes the specific relational context of
the secret information, and thus provides a more precise and comprehen-
sive analysis of when to reveal personal secrets. In particular, it includes an
analysis of whether the partner expects access to the hidden information
and how likely the partner is to discover the information on his or her own,
whereas the previous model did not. The term "relational partner" in this
context refers to any other person with whom one has a relationship,
including one's boss, spouse, or friends. In the following paragraphs, I
elaborate on each of the steps to the proposed model. Each of the next four
headings corresponds to the four boxes in Fig. 1, starting with the box at
the top of the figure.

PRIVATE VERSUS SECRET INFORMATION

Derlega et a1. (1993) have argued that in the best interests of a relation-
ship, on the one hand, a person may want to disclose information that re-
lates to issues of trust, trustworthiness, and caring for one's partner. An
example might include having gambled away some of the couple's earnings
that week or having had an extramarital affair. On the other hand, Derlega
and coworkers argued, information that does not have important implica-
tions for the relationship may not be necessary to reveal (see also, Karpel,
1980). Examples of this type include secrets about one's past, such as a
sexual molestation that has been reasonably resolved and does not seem to
have significant effects on how the person behaves in the relationship.
According to Derlega et a1. (1993), the person may want to reveal the infor-
mation as a means of showing trust in the partner, as well as wanting to be
seen accurately by the partner, but need not feel an obligation to do so.
However, this distinction between secrets that have bearing on the
relationship and those that do not is a very difficult one to make. Take the
following as an example: An acquaintance, whom I will call Nancy, offered
me a detailed account of the revelation of her most troubling secret.
She told me that through her experiences in therapy, she finally got the
courage 3 months before her wedding to tell her fiance that she had
received an abortion as a teenager. She felt that the information was
relevant to their upcoming marriage and plans to have children, especially
since they are both Catholic. She also felt that because it was relevant, not
telling him would create mounting stress for her. Thus, she gave him the
full disclosure of what was to her a traumatic experience and secretive
WHEN TO REVEAL PERSONAL SECRETS 193

burden. Although he seemed to respond well to the revelation initially, 2

weeks later he called off the wedding and asked her to move out of their
new house. She was so shocked and distraught over this rejection that she
took sedatives for several months. After she recovered, she said that she
was relieved to discover sooner rather than later what an empty and
unsupportive person he was. Although just months earlier she had felt
that he was the love of her life and was so happy to be marrying him, now
she believed that it was better to get out of this relationship than to live
with the stress of that secret.
Was the abortion a relevant secret that had bearing on issues of trust in
the relationship? Or was it a matter of privacy and something she would
have benefitted from keeping to herself? Part of the reader's interpretation
of the wisdom of Nancy's decision to reveal her abortion secret is likely to
reflect the reader's core assumptions about the amount of stress that is
associated with secret keeping and the extent to which suppressing infor-
mation can undermine or sabotage a relationship. Classically trained psy-
choanalysts, for example, are likely to believe that virtually all major secrets
are relevant to adult functioning and interpersonal relationships and that
keeping secrets could allow the secrets to manifest themselves in recurring
nightmares and other symptoms. Viewing secrets in this manner would
seem to lead to the conclusion that Nancy made a wise decision to reveal
the abortion secret. The reader should keep in mind, however, that psycho-
analysts also might interpret Nancy's revealing her secret as an effort to
sabotage the relationship and avoid dealing with her own feelings of
intimacy and commitment. They might argue that revealing the secret in
the therapy session is appropriate and necessary, whereas revealing to the
partner is not.
The research that has the most direct bearing on Nancy's situation
was the study by Major and Grarnzow (1999) described in Chapter 3. They
found no relationship between keeping an abortion a secret and feeling
distressed (after they statistically controlled for other, more relevant fac-
tors), which suggests that keeping an abortion a secret might not be as
distressing as one might think. However, even if one does assume that
secret keeping is stressful and can sabotage a particular relationship, I
contend that the relevant-secret versus not-relevant-secret distinction pro-
posed by Derlega et al. (1993) does not offer a completely satisfying guide
for when one should reveal a personal secret in part because it is extremely
difficult to determine which secrets are relevant to issues of trust in a
particular relationship. In addition, sometimes a partner will directly ask a
question about a particular secret and say, "Have you ever had an abor-
tion?" Derlega and coworkers' recommendation does not address how
one should respond in such cases.
194 CHAPTER 9

Perhaps a better guide for making the decision to reveal personal

hidden information gets back to my definition of secrecy versus privacy
described in Chapter 1. When making the decision to reveal hidden infor-
mation, a person must first decide whether that information is indeed
secret-as opposed to private-information. Specifically, the individual
must ask himself or herself, "Does the other person expect to have access
to this kind of information" (Le., expect to be told the hidden information)?
If the answer is no, then any fretting over a decision to reveal is moot. The
person can write off that information as private and keep it hidden. In
Nancy's case, her therapist encouraged her to reveal the abortion secret to
her fiance. Thus, it is possible that he reframed that once-private informa-
tion as secret information, even though in actuality Nancy's fiance may not
have expected access to this information and may not have wanted her to
reveal it (Le., the information may have been private in the context of that
relationship).
Such distinctions between secrecy and privacy also are essential in
professional settings, where much personal information is private and is
expected to be hidden. Personal revelations made at work in the name of
openness and honesty often are considered inappropriate. Just as Peskin
(1992) observed that the ability to conceal information to get what one de-
sires is a sign of maturation in children, so too I argue that using discretion
in the workplace is a sign of maturity and professionalism. However,
people often reveal personal information at work that they probably
should not reveal. In my own experiences as a professor, for example, I
have been struck by the fact that students sometimes tell me that the
reason they missed a class or test was that they were hung over and were
throwing up! Such disclosures put me in an awkward position: Knowing
that they induced their own illness makes me reluctant to give them a
makeup exam because it does not seem fair to the other students who
studied the night before the exam and took it on time. It is much better
when they offer a simpler explanation such as, "I was sick and had to
miss the test." In another example, a master's-level student once asked
me to be on his thesis committee, even though I did not know the research
area of his thesis very well. When I formally met with him and asked why
he had picked me for his committee, he said bluntly and honestly, "I know
that you are flexible when it comes to students who have already collected
their data before the proposal meeting." That was the entire explanation.
There was nothing substantive mentioned about why I might be an appro-
priate member of his committee, just the simple truth. In yet another
example, a student in my undergraduate seminar told me in a private
meeting that she had written the paper (for which she got an A) in one
night. In all these cases, it would seem as though the students simply
WHEN TO REVEAL PERSONAL SECRETS 195

slipped up on their impression management tactics and just wanted to be

honest. However, given the data presented in Chapters 7 and 8, it seems
that at some level people always are trying to construct desirable identi-
ties, or at least avoid undesirable ones. In these cases, the students proba-
bly thought that being seen as honest was good and desirable in this
professional context, just as it typically is in their personal contexts. In
truth, though, I was startled by the blunt honesty and realized that I have
come to expect private information to remain hidden in the workplace as
part of professionalism. I viewed their disclosures as private information
to which I did not expect to have access. The students may have believed
that they would be viewed favorably for being so honest, but actually they
more aptly might be described as being somewhat naive or lacking matu-
rity in this instance. Thus, my recommendation to the reader is to keep in
mind what the relational partner expects to have access to and what the
person expects to be kept private, and to weigh that information heavily in
a decision to reveal or conceal the information. This recommendation is
especially salient at work where there is so much at stake and where
keeping private motives hidden is often the norm. Revealing too much
information can be seen as naive rather than as honest.
However, if the answer to the question of whether the other person
expects access to hidden information is yes, then the information is a
secret. For instance, some professors actually may expect students to give
them detailed explanations for why they miss class and reward them for
admitting to the complete truths. As the reader can see, such expectations
are defined from the perspective of the potential confidant in a particular
relationship, rather than defined according to some general rule about
various social contexts. Even therapists who generally report wanting a
great deal of openness from their clients do vary on what they expect their
clients to reveal (Kelly, 1998, raw data). If one is uncertain what the other
person's expectations about revelations are, he or she can ask about the
partner's feelings about the general topic. Some couples, for example,
agree that it is acceptable to engage in extrarelational sexual activities but
not to tell each other about them. I contend that ascertaining whether the
partner expects access to certain kinds of hidden information typically is
easier than trying to figure out whether the hidden information has bear-
ing on issues of trust or trustworthiness in a relationship because the
former requires less insight from the partner. Also, as mentioned earlier,
new relational partners often ask direct questions such as, "How many
sexual partners have you had" or "Have you had an HlV test" which
suggests that they expect to be told that information. If the hidden infor-
mation is indeed a secret, as defined by the expectations of the relational
partner, then the secret keeper must make some additional decisions
196 CHAPTER 9

before revealing the secret. These decisions regarding the revealing of

secrets, as opposed to private information, are outlined in the following
paragraphs.

DETERMINING WHETHER A PARTNER

Is AN ApPROPRIATE CONFIDANT

Once a person has decided that the hidden information is a secret in

the context of a particular relationship, he or she must try to gauge
whether the relational partner is an appropriate confidant (i.e., who will be
discreet and nonjudgmental and will not reject the secret keeper). How-
ever, anticipating whether a person qualifies as an appropriate confidant
presents a challenge because it is tough to know how the person will react
until after the secret has been revealed. If the secret keeper knows the
partner well and has known this person for a substantial period of time,
then the secret keeper will have a wealth of prior experience on which to
base a prediction regarding the likelihood that the partner's response to a
revelation will be positive. The secret keeper may reflect on times when he
or she or others have revealed personal information to the partner and
may recall whether the partner reacted in a supportive fashion as well as
refrained from revealing the information to other people. The secret
keeper may also recall the partner's stated opinions on topics related to the
secret and use this information to predict the partner's likely response. For
example, a formerly bald man would probably avoid revealing the fact
that he has had hair implants to a partner who has repeatedly expressed
disdain for those who have had such procedures.
In addition to relying on previous experience, individuals may test
their partners' responses to a disclosure by jokingly or seemingly inadver-
tently introducing the topic and gauging their partners' reactions (Duck,
1988). Miell and Duck (1986) have shown that, in much the same way as
politicians do, people float trial balloons before making serious and poten-
tially damaging disclosures. For example, a husband might recount to his
wife the tale of a friend's marital infidelity (taking great care not to arouse
suspicion) and carefully study her reaction before deciding whether to
reveal his own transgression. Or a child might ask a parent, quite transpar-
ently, what could happen to "someone" who admitted to stealing candy
from the checkout line, before revealing that he or she had actually stolen
the candy. Other ways of anticipating whether the relational partner will
be an appropriate confidant might include paying attention to whether
that person reveals personal information about other people and whether
others consider this person to be trustworthy and nonjudgmental. If the
partner passes these "tests," then the secret keeper should reveal to that
WHEN TO REVEAL PERSONAL SECRETS 197

person, knowing that there is still some risk or uncertainty in sharing with
another.
Typically, such evaluations of the partner as a potential confidant take
time. Timing also is important in the sense that the partner is likely to
expect different levels of revelation at different times in the relationship.
Models of self-disclosure indicate that the association between stage of
relationship and the effects of self-disclosure is curvilinear (see Collins &
Miller, 1994; Derlega, 1988). Specifically, revealing highly personal infor-
mation upon meeting someone may backfire, whereas making such dis-
closures after getting acquainted with someone can be a way of develop-
ing intimate bonds with that person (Altman & Taylor, 1973; Wortman,
Adesman, Herman, & Greenberg, 1976). As mentioned in the previous
paragraphs, knowing a partner well allows the secret keeper to have
enough information to determine whether the partner will keep the secret
and will not judge the secret keeper for its content. However, after know-
ing someone a long time, revealing secrets again may backfire, because the
partner may perceive that the secrets have been inappropriately withheld
and may feel betrayed by the secret keeper (see Derlega, 1988). For exam-
ple, if a patient who is receiving treatment to control her drinking con-
fesses to her therapist during their last scheduled session that she has
continued drinking throughout the treatment, the patient will not have the
opportunity to sort out the implications of the revelation with the thera-
pist. Moreover, the therapist may be astounded by the fact that she did not
trust him enough to tell him this relevant information earlier in their
sessions. I recommend that people share their secrets with relational part-
ners they know well but not so well that it is too late to reveal to those
partners.
An important paradox emerges when considering what stage of a
relationship is the appropriate time for a revelation. Although people in
relationships tend to trust each other more as time passes, their self-
disclosure becomes more judgmental as the relationship develops (Der-
lega et al., 1993). Thus, each partner might be tempted to reveal very
personal information during the initial, relatively nonjudgmental phase of
the relationship as a way of getting closer to the other person. However,
that same information that was once viewed without negative judgment
may come to haunt the revealer later when his or her partner looks back on
those disclosures with disdain. For example, a woman tells her new atten-
tive boyfriend that she experimented with lesbian relationships in college
but that she is decidedly heterosexual now. He seems at ease with the
information and even thinks that the revelation makes her more intrigu-
ing, until a year later when she begins to lose interest in sex with him
because she is so buried in work. He becomes threatened by her earlier
198 CHAPTER 9

disclosures and "accuses" her of being a lesbian. The point of this example
is to encourage a secret keeper to use discretion, especially at the begin-
ning of a relationship when the partner seems most receptive and non-
judgmental. My rationale for this recommendation is that the secret keeper
has not yet had enough contact with the person to be able to evaluate how
he or she might respond to various revelations later.
In addition to being sensitive to the stage of the relationship when
deciding to reveal to a relational partner, the secret keeper should make
sure that the partner knows that the information should be kept hidden
from others. Sometimes partners have a different philosophy of openness
and inadvertently tell others' secrets because they themselves would not
have kept such secrets in their own relationships. For example, 2 years ago,
a friend told me a "secret" about his participation in a dangerous team
athletic event. He had to climb a telephone pole as part of a team-building
exercise with his colleagues. I thought that event actually made him look
daring and brave, but he asked me not to tell anyone. Some months later,
I stumbled and told the story to another friend right in front of him! He
was irritated with me and I was shocked at my own slip, given how much I
care about being a good confidant. I realized later that it was hard for me to
remember that the story was supposed to be kept secret because it did not
seem inherently like "secret material." I believed that at some level he
wanted me to tell others, but I was wrong. Conversely, sometimes people
tell secrets that they expect (and hope) will get spread around. One time in
a professional setting a senior colleague told me in confidence that she was
going to resign. I did not tell anyone that information, and she was later
shocked to discover that none of my peers knew about her imminent
resignation. In that case, she herself would have told others that "secret,"
and thus expected me to do the same. My point here is that if one wants to
tell his partner a secret, then he should offer the rationale for why it should
be kept secret. If the other does not seem to understand the rationale or
agree that the information should be kept secret, then the revealer should
expect that the other person might reveal it later. At the same time, if one is
the recipient of such information that he does not agree should be kept
secret, then he should say that he cannot be expected to remember in the
years to come to keep the secret.
Another occurrence that enhances the chances that a relational part-
ner will remain an appropriate confidant and continue to be discreet long
after the revelation is that the partner usually discloses very personal
information in return. Self-disclosure researchers often have discussed the
norm of reciprocity: When someone reveals personal information to an-
other person, it is expected that the other person will share some personal
information in return or will act interested in the revelation and show
WHEN TO REVEAL PERSONAL SECRETS 199

understanding (Berg, 1987; Berg & Archer, 1980; Chaikin & Derlega, 1974).
If a secret keeper learns equally personal and potentially stigmatizing
information about the partner, then the partner might fear retaliation
enough that he or she would never reveal the secret even if the two were
no longer in a relationship. Part of the trouble with the modem, mobile
society with a very high rate of divorce is that husbands and wives tell
each other very personal secrets, which still are protected in courts of law
under "spousal privilege," and then often end up in new cities with new
spouses. What they have divulged to their previous spouse is no longer as
protected as they once had thought because there is less accountability
now that they share different networks of friends. What this means is that
people who have new networks do not necessarily feel the need to keep a
former partner's secrets because they do not have to worry about losing
that bond (i.e., it is already lost). At the same time, because of the "small-
world problem" (see below) and increasing technology that is intensifying
that problem, there is a greater risk than ever that one's reputation can be
damaged by a former spouse's revelations. In a nutshell, revealing one's
personal secrets to a relational partner puts the revealer in a vulnerable
position. This vulnerability often is diminished because the revealer learns
some equally stigmatizing information about the partner. A way of pro-
tecting oneself is to show the other that there is still some accountability
and relational connection even after a breakup.
In sum, if a secret keeper believes that his partner is an appropriate
confidant who will not tell others the secret, will not judge him negatively,
and will not reject him, then he should reveal the secret to her. He needs to
wait until the timing is right both to decide whether she has these qualities
and to match her expectations of their intimacy level for the time that they
have known each other. Also, he should tell her why the information
should be kept secret, so that she can remember to do so. The fact that she
usually will share secrets in return may enhance her level of continued
discretion, even if the relationship ends. If he decides that she is not an
appropriate confidant, however, then he next needs to consider how likely
she is to discover the secret on her own before deciding whether to reveal
the secret to her.

PROBABILITY OF BEING DISCOVERED

People typically expect to have to protect their social reputations,

especially when they think that their actions will come to the attention of
their family and friends from outside sources (Emler, 1990). Assuming that
there are benefits associated with revealing hidden negative personal
information oneself-so that one can put his or her own spin on the
200 CHAPTER 9

information and gamer the support of intimates (Emler, 1990)-it would

be helpful to be able to gauge the probability of a secret's being discovered
when making the decision to reveal the secret. As yet, there is no objective
way of knowing what these probabilities are. In the following paragraphs,
I offer competing arguments concerning whether people are likely to
underestimate or overestimate the likelihood of their secrets being dis-
covered.
Emler (1990) proposed that because people rarely commit shameful or
delinquent acts alone, these acts are not as concealable as they might think.
Supporting this idea are the findings (described in Chapter 8) that confi-
dants frequently reveal personal information about other people (e.g.,
Christophe & Rime, 1997). Moreover, researchers have demonstrated that
when social reputations are measured, typically there is a consensus
among knowledgeable informants (i.e., those who share community mem-
bership or having extensively overlapping personal networks with the
target person) (Emler, 1990) about the attributes of the targets (e.g., Cheek,
1982; Moskowitz & Schwartz, 1982). Even though these reputational mea-
sures have been criticized for measuring more about the perceivers than
the perceived (e.g., Cronbach, 1955), they do suggest that people's social
networks may be more interconnected than they think. Emler (1990) noted
that this is especially the case for people with high levels of social partici-
pation, such as actors, teachers, and politicians. Likewise, Milgram (1967)
showed that people's social networks are surprisingly interconnected,
calling this phenomenon the "small-world problem. Specifically, he ob-
II

served in 1967 that starting with any two people, the probability that they
will know each other is 1 chance in 200,000 for the American population.
However, there is a 50/50 chance of linking any two people up with only
two intermediate acquaintances! He demonstrated that the median inter-
connection between any two Americans chosen randomly is 5, with a
range from 2 to 10. In addition, Emler and Fisher (1981) found that in
informal conversations of students and teachers, the most common topic
of conversation after self-disclosure was named acquaintances and the
doings of these acquaintances. Thus, it seems both that people are more
likely to be linked to others than they think and that these others are likely
to be gossiping about common acquaintances.
I myself witnessed the "small-world" phenomenon firsthand at a
recent hair-cutting appointment. My hairdresser told me that while his
good-looking male friend was on vacation, the friend had sex with a
beautiful and very famous young actress. My hairdresser did not tell me
the name of his friend but certainly did tell me the name of the actress. This
example fits Emler's point that a person's level of social participation
affects how much people are aware of that person's social reputation. In
WHEN TO REVEAL PERSONAL SECRETS 201

this case, the actress has a much higher rate of social participation as
compared with my hairdresser's friend, and thus needs to be much more
aware of what people know about her if she is to manage her reputation.
At the same time, however, because the actress is so famous, the listener is
probably more likely to assume that the story was made up as a form of
bragging.
In contrast to Emler's (1990) position that people tend to overestimate
how concealable their stigmatizing actions are, I suggest that it also is
possible that people might be too paranoid about how much people are
likely to discover about them. Tversky and Kahneman (1973) demon-
strated that people tend to overestimate the probability of the occurrence
of events that are salient or vivid to them (i.e., those that come to mind
more easily). These researchers referred to this phenomenon as the "avail-
ability bias." Moreover, as described in the previous chapter, Ross et al.
(1977) demonstrated that people tend to see others as more similar to
themselves than they actually are. Applying these well-documented prin-
ciples to the model, I suggest that because people's own secrets are likely
to be more vivid or salient to them than they are to other people, and
because people tend to assume that others are more similar to themselves
than they actually are, people may overestimate the probability that others
are discussing or thinking about their secrets. They may draw this conclu-
sion simply because they themselves are thinking about their secrets.
I suggest that a general guide for how likely it is that one's relational
partner will discover a secret is whether there have been witnesses to this
secret event; if there are witnesses and those witnesses have either direct or
indirect contact with the partner, then one may assume that there is a
reasonable chance that the partner will discover the secret. (Unless the
witness is particularly discreet or is legally and ethically bound to confi-
dentiality, such as one's doctor, therapist, or lawyer.) In cases where there
are witnesses, the secret keeper and all the witnesses must keep track of
what is secret information, and this is not an easy task. Wegner et al. (1994)
observed that under conditions of high cognitive load, people are not very
good at suppressing secret information. High cognitive load refers to
diminished cognitive capacity, which can result from drinking alcohol,
being very tired, or being nervous or distracted. This difficulty is illus-
trated by what happened to a couple (whom I will call Jim and Cindy) who
dated for 2 years, broke up, and then each had a brief sexual relationship
with another person. Their different sets of friends knew about their
respective brief relationships. Then, when the couple got back together
and their friends were sitting around having a drink, one of Cindy's
friends mentioned her dislike of the man whom Cindy had dated briefly.
Cindy had not yet told her boyfriend that she had another relationship
202 CHAPTER 9

when they broke up, so Jim became very embarrassed. His friends could
see that he was not aware of the information and that he was stunned to
be told under those circumstances.
If others know the secret and they have any contact with the partner,
then the secret keeper should consider revealing the secret to the partner.
This recommendation especially holds in clear cases in which the discov-
ery of the secret is imminent, say, for example, a former lover threatens to
tell one's husband about an affair. The secret keeper should explain her
perspective on the once-hidden event. Even then, though, trying to put a
spin on the revelation might not be that helpful if the husband believes that
she was forced to reveal the secret. Thus, it is important to reveal the secret,
focusing on the theme and not the details (if possible) at a time when the
partner can believe that the decision to reveal it was initiated by the secret
keeper herself.
If only the secret keeper knows the secret, however, then it is possible
that he or she will be able to keep the secret quite well after a period of
getting used to the secret. Moreover, the secret keeper should be better at
keeping it than other people because he or she is more motivated to keep
track of what should be kept secret. In the example of Nancy who had an
abortion, it seems as though since no one knew about it except for her (and
her doctor) from the beginning, she perhaps eventually could get used to
the idea that it is private information that no one but her needs to know
about. If it is not likely that the partner will discover the secret because
only the secret keeper knows about the hidden information, then the secret
keeper will need to move to the next step of the model in deciding whether
to reveal the secret to the partner.

TROUBLING SECRETS

If the secret keeper determines that she does not think that the partner
will discover the secret, she still needs to ask herself whether the secret is
troubling in the context of the relationship. It may seem obvious that one
only should reveal a secret if the secret is troubling. However, as described
in the previous chapters, this idea contradicts the long-held supposition by
psychoanalysts dating back to Freud (see Freud, 1958) that secret keepers
ultimately will develop symptoms if they continue to squelch their secrets.
Although there is a fair amount of recent correlational evidence that self-
reports of secret keeping and symptoms are linked (e.g., Ichiyama et al.,
1993; Larson & Chastain, 1990), there is no direct experimental evidence
that people who choose to keep secrets actually develop symptoms as a
result (see Chapter 3). It is possible that with practice, people can learn to
WHEN TO REVEAL PERSONAL SECRETS 203

suppress their unwanted thoughts with very little effort (Kelly & Kahn,
1994). In cases in which keeping secrets from the partner is relatively
effortless for the secret keeper, I argue that the secret keeper should avoid
sharing secrets to sidestep the rejecting feedback and other negative conse-
quences that can be associated with sharing (e.g., see Lehman et al., 1986).
It is the people who are suffering from carrying personal secrets whom I
encourage to reveal their secrets.
How does one know if a secret is troubling in the context of a relation-
ship? Sometimes the link between a secret and problems is apparent to the
secret keeper because he or she is ruminating over the secret and is upset
by such repetitious and intrusive thoughts. But at other times the connec-
tion between secrets and symptoms is less clear. I suggest that if a person is
keeping a secret and is experiencing symptoms, such as depression, ulcers,
and headaches, that have been found to be associated with secret keeping
(e.g., Larson & Chastain, 1990), then the person should consider the possi-
bility that the symptoms are the result of keeping the secret or that the
secret is troubling. Also, if the person is worried that the partner would
feel betrayed by the secrecy or that the partner will find out about the
secret, then the person should view the secret as troubling.
I suggest that people who are troubled by their secrets in the context
of particular relationships should talk about them with their relational
partners. This decision to reveal is risky, however, because in reaching this
last step in the model, the secret keeper already has assessed that the
partner may not be an appropriate confidant (i.e., is not discreet, or is
judgmental, or will reject the secret keeper). In the following paragraphs, I
offer options to revealing troubling secrets to a relational partner who may
not be an appropriate confidant. My emphasis is on avoiding the dangers
associated with revealing to such a person who could not only end the
relationship but also damage the secret keeper's reputation afterward by
telling others about the secret.

OPTIONS TO REVEALING THE COMPLETE TRUTH

REVEALING THEMES AS OPPOSED TO DETAILS

As noted in the previous paragraphs, there are two points in the

model where the secret keepers are encouraged to focus on revealing the
themes to their secrets rather than the details of them. These are when the
partner is not perceived as an appropriate confidant and either (1) the
partner is likely to discover the secret on his or her own or (2) the secret is
troubling in the context of the relationship. For example, imagine that a
204 CHAPTER 9

young, conservative man asks his new girlfriend how many sexual part-
ners she has had. Imagine, too, that she estimates that she has had 60
partners and fears that when they visit her hometown at Christmastime,
he will discover that she had a reputation in high school for promiscuity.
Because she is worried that he will reject her if he knows how many
partners she has had, she can focus (truthfully) on the theme by saying, "I
think it's crude to discuss numbers, but I can tell you that I used to be
pretty sexually active when I was younger. Another example that is much
II

more disturbing involves a psychotherapy client who has burned his

children with lit cigarettes as a punishment for their misbehavior. The
client may imagine or accurately perceive that his therapist will judge him
negatively for this behavior, and the therapist is legally required to report
such a revelation to the authorities. Yet he is very troubled by the secret, so
he says, "I have trouble managing my anger with my children and want to
learn what I need to do when my anger gets out of control. This way he
II

can receive help for his abuse of the children and avoid losing his custody
of them and being put in prison.
Even though I am offering this as a potential solution to the dilemma
of wanting to be open and yet avoid negative judgments or rejection,
Arkin and Hermann (2000) argued that themes, as compared with specific
behaviors, have greater implications for one's identity (Vallacher &
Wegner, 1987). In particular, they asserted that a focus on themes in the
context of psychotherapy could encourage clients to experience a gener-
alized sense of shame as opposed to a feeling of guilt, and that guilt can be
adaptive in inducing clients to change their behaviors, whereas shame can
prompt a maladaptive effort to change the self. As a solution to the
dilemma surrounding negative disclosures, Arkin and Hermann (2000)
indicated that
it may be better to describe the event fully, but in a neutral, factual manner.
This would allow the opportunity for client and therapist, when necessary, to
reframe the client's understanding of specific events and help negate the
tendency to move to broader, self-oriented levels of thinking and characteriz-
ing of the self. (p. 503)

Their suggestion certainly is reasonable for clients who can imagine that
their therapists would view them favorably after a description of negative
events. Moreover, it captures what many therapists, including psycho-
analysts and cognitive-behavioral therapists, already do.
However, my recommendation targets those times when clients and
secret keepers in general cannot imagine that their confidants would view
them favorably after a revelation. I agree with Arkin and Hermann (2000)
that it can be beneficial for a client to reveal behaviors like, "I sometimes
WHEN TO REVEAL PERSONAL SECRETS 205

slam the door, or stomp out" (p. 503). But these behaviors almost certainly
would not be in the category of acts that most clients would fear revealing
to their therapists. In Kelly (2000a), I used the example of the man who hit
his wife with a bat one time because when people do such terrible things,
those few acts often corne to define them. The same is likely to be true for
very humiliating details, such as those in the infamous Starr Report (1998),
which was criticized for its gratuitous provision of sexual details: "Accord-
ing to Ms. Lewinsky, she performed oral sex on the President on nine
occasions ... on one occasion, the President inserted a cigar into her
vagina" (p. 5). I speculate that a client who feels burdened by similarly
humiliating details could indicate something like "we had sexual contact"
and could discuss her feelings about the events to let the therapist know
what impact those events had on her. This option might be preferable to
revealing the details and then imaging the undesirable themes that the
therapist is developing about her. Of course, one must keep in mind that it
is possible for a person to want others to hear such sexual details, because
she anticipates that she will be seen in a desirable way (e.g., as a playful or
adventurous woman). Also, what is considered to be undesirable by most
people in one culture at one period of time may not be undesirable in
another culture or at another time. My recommendation to focus on
themes as opposed to details applies only to cases where the revealer
expects to be viewed in an undesirable way if the details are revealed.
Support for my argument that a few noteworthy acts can be linked
directly to their broader undesirable themes comes from evidence (see
Mischel & Peake, 1982) that people judge others to have consistent traits,
such as self-centeredness or conscientiousness, by stringing together their
highly prototypical behaviors (i.e., behaviors that are representative of
their broader categories). If a therapist hears that a client has beaten his
wife on the back with a bat, the client might imagine that the therapist will
weigh that detail heavily in evaluating that client's overall personality.
This perception is likely to be accurate, too, given that Regan and Hill
(1992) found that therapists formed very negative clinical conjectures
about their clients, which they then hid from the clients. In addition,
research has shown that therapists' perceptions of a target person are
consistently less favorable than laypersons' perceptions, whether that
target person is a client or nonclient (see Wills, 1978). It should not be
surprising that clinicians form such negative opinions, because they are
trained to use the Diagnostic and Statistical Manual of Mental Disorders
(DSM) (see American Psychiatric Association, 1994), which lists behaviors
that qualify their patients to have various disorders.
My suggestion about revealers' generating their own themes involves
their creating relatively desirable themes, as opposed to the undesirable
206 CHAPTER 9

themes (e.g., having feelings of being self-centered) described by Arkin

and Hermann (2000). For example, the man who hit his wife might say, "I
am a person who is committed to dealing with my issues of rage" as
opposed to, "I am a despicable wife-beater who has no place in society." If
the client reveals the detail of hitting his wife, he might imagine getting
feedback from the therapist that supports the more undesirable wife-
beater label. Moreover, therapists probably are more apt to empathize with
and like their clients if the clients can focus on their feelings of rage,
frustration, hurt, and helplessness, rather than if the clients describe the
heinous acts in a detached way. Empathy has been found to play an
important role in clients' improvement (see Beutler et al., 1994), and clients
who are more well liked by their therapists tend to show more progress in
therapy (e.g., Stoler, 1963).
In the case of revealing to relational partners who are not ethically or
legally bound to keep the secret information confidential the way thera-
pists are, an important added benefit of revealing themes (which are
phrased in ways that are desirable to the revealer) is that even if the
partners tum out to be indiscreet, they still will be less able to use that
information to damage the revealer's reputation. Themes, as compared
with specific details, are less vivid and less interesting, and therefore are
less likely to be fodder for gossip. For example, saying "I used to be more
sexually active" is less noteworthy than "I had 60 sexual partners."
In sum, an alternative to complete revelation is to focus on the themes
(which are phrased in desirable ways) as opposed to the details of the
secrets. The problem with revealing details is that they may be readily
linked to broader negative judgments, and relational partners may tell
others these details and damage the revealer's reputation. Even in therapy,
clients might consider revealing the themes to their problems when the
details are particularly heinous and they would not be able to imagine that
the therapist would view them favorably after hearing the details.

ESTABLISHING WHAT Is PRIVATE INFORMATION

As mentioned earlier, the first steps in the model are to determine

whether the relational partner expects access to the information and
whether the partner is an appropriate confidant. I suggested that if the
answer to both of those questions is yes, then the secret keeper should
reveal the secret. However, it is still up to the secret keeper to decide
whether he or she really wants to reveal that information. For example,
even if the partner is an appropriate confidant who expects to be told
certain information, such as whether the secret keeper has ever been
pregnant, the secret keeper still may choose to keep this information to
WHEN TO REVEAL PERSONAL SECRETS 207

herself because she views it as something that he should not expect access
to or because she would like to keep her personal boundaries firmly
established in the relationship. Relationships are dynamic, and part of
developing intimacy involves learning what level of revelation the two
partners can expect from each other. If the partners have different expecta-
tions, they need to identify this difference early on in the relationship to
avoid overstepping their boundaries and offending each other. They can
set up rules such as "Don't ask questions when you can only accept one
answer" or "Let me offer information about my sexual past if I wish-do
not ask me specific questions."
An interesting paradox emerges when people respond to personal or
embarrassing questions: They may view a completely factual and truthful
set of responses as less representative of themselves than a more distorted,
favorable set of disclosures (see Kelly et al., 1996; Schlenker, 1986). This
distortion seems to be a normal part of everyday living and has even been
construed as a sign a positive mental health (see Taylor & Brown, 1988).
Imagine that someone who sees herself as a very moral and good person is
cornered into a factual revelation about her past and in particular is asked
whether she has ever had an affair with a married man. To herself, she
acknowledges that she has had such an affair but is committed to never
doing that again. As such, she does not see the whole, blunt truth as being
representative of herself as she is now. So paradoxically she may be
representing more honestly how she sees herself by omitting this previous
bad act, especially if she feels that the listener will draw the wrong conclu-
sions about her if she reveals this information. Even if she is factually
truthful in her disclosures and then tries to explain the circumstances in
which she had this affair, the chances are that the listener will not weigh
that disqualifying information very heavily in judging her character. After
all, researchers have demonstrated in a number of studies over the years
that although people weigh situational causes for their own bad behaviors
very heavily, they tend to attribute bad acts to other people's character
(see, for example, Silvia & Duvak, 2001).
My suggestion for addressing this dilemma of wanting to be truthful
but also to avoid presenting oneself in an undesirable way is to say
something like, "As a matter of principle, I consider my previous sexual
experiences-or lack thereof-to be private matters. If I answer this ques-
tion, there may be more to follow like it, and I want to be clear that I would
rather volunteer any such information than to be asked about it." If instead
the question was how many sexual partners has a person had, the partner
might have a legitimate concern for his safety as her new lover. In that case,
she can use condoms and get tested for sexually transmitted diseases to
address any potential risks that getting involved with her poses to his
208 CHAPTER 9

health. A revelation of the details of her sexual past is not required to

address his concerns. This recommendation to declare what she finds to be
private early in the relationship is consistent with the empirical findings
that partners in successful relationships negotiate what they will talk
about with each other (e.g., Coupland, Coupland, Giles, & Wieman, 1988).
Such negotiations can avoid some of the complications involving the
dynamics between people. For example, there are times when a person
perceives that the partner expects access to the hidden information when
the partner really does not expect access. There also are times when the
partner already knows about the hidden information and considers it
private yet still gets upset because the partner believes that the person is
keeping it secret. Because the process of keeping a secret from one's part-
ner can be disconcerting, it may be especially useful to identify what is
private information early in the relationship.

REVEALING TO SoMEONE ELSE

Imagine that a secret keeper has gotten through each of the steps of
the model and has decided that a secret is troubling in the context of the
relationship and that the partner is not an appropriate confidant and is
unlikely to discover the secret on his own. In such a case, the secret keeper
is in a bit of a dilemma. If the secret keeper could conquer her feelings
of being troubled about the secret, then her relationship could remain
undamaged by the secret. The example of Nancy, who had the abortion,
comes to mind. She talked to her therapist about the abortion, and he
encouraged her to reveal it to her fiance. However, it might have been
enough for her to work through her feelings about the abortion with the
therapist, who could have helped her see the abortion as a private matter
that could rest within her with no negative consequences. As described in
Chapter 5, a possible route to feeling better about one's secrets is to gain
new insights into them. Therapists are trained to offer interpretations to
clients' problems, and secret keepers might benefit from sorting out their
troubling secrets with the therapists instead of with relational partners in
their immediate social context.

WRITING IT DOWN

A drawback to this last suggestion stems from the finding in my

(Kelly, 1998) study, which showed that even with trained therapists who
presumably fit all the positive qualities of confidants described in the
previous chapter, clients who were keeping a relevant secret experienced
greater symptom reduction than those who were not keeping one. The
seeming contradiction between this finding and the model can be resolved
WHEN TO REVEAL PERSONAL SECRETS 209

by the fact that the model stresses that the secret keeper's perception of the
relational partner as nonjudgmental is critical in his or her decision to
reveal a personal secret. It is likely that at times secret keepers may
imagine, or accurately perceive, that their relational partners are judgmen-
tal, particularly when the secret keepers have committed unusually hei-
nous acts. Moreover, in some cases, even if a relational partner is com-
pletely trustworthy, the secret keeper may not perceive the partner's
trustworthiness or may simply never feel comfortable telling anyone
about an appalling or extremely embarrassing secret. In these cases, the
secret keeper may benefit from privately writing down the secret and not
sharing it with the partner (see Pennebaker, 1990). It is possible that he or
she may benefit from writing about the secrets in an effort to gain new
insights into them (see Kelly et al., 2001). However, this option is only
reasonable in cases in which the partner who is an inappropriate confidant
is unlikely to discover the information. If the partner is likely to discover
the secret, then a more drastic step may be required, as described next.

ENDING THE RELATIONSHIP

Sometimes it is better to end the relationship with a partner who is not

perceived as an appropriate confidant and who nonetheless demands
access to a particular secret, such as one's having herpes, than to reveal the
secret. If the partner learns of the herpes secret, he or she might end the
relationship anyway and then tell others about this potentially embarrass-
ing and reputation-damaging secret. Since self-disclosure is such a crucial
element of forging relationships (see Derlega et al., 1993), perceiving one's
partner as an inappropriate confidant may be sufficient grounds for end-
ing a relationship anyway.

LIMITATIONS
ENCOURAGING THOUGHT SUPPRESSION?

Elsewhere, social psychologists Arkin and Hermann (2000) criticized

my recommendation for psychotherapy clients to reveal themes as op-
posed to details of really heinous acts to the therapists (see Kelly, 2000a,b)
and argued that efforts to withhold information may have the ironic effect
of enhancing the impact of that information on one's self-concept. They
cited research on the negative effects of thought suppression (see Wegner,
1994) and suggested that encouraging clients to withhold details might
make those thoughts even more vivid, salient, and central. Hill et al. (2000),
too, suggested that clients' withholding information could cause them to
miss out on the benefits of therapy, and they cited research on the negative
210 CHAPTER 9

effects of inhibiting information to support their claim (see Pennebaker,

1997a).
I have two sets of responses to this point. First, I am advocating an
overhauling of current expectations of very high levels of revelation in
therapy. If clients can come to believe that it is not a requirement of therapy
for them to reveal unseemly details about themselves, then there would be
no need for them to suppress such details. As described in Chapter 1, I
make the following distinction between secrecy and privacy: Whereas
privacy connotes the expectation of being free from unsanctioned intru-
sion, secrecy does not. Secrecy involves active attempts to prevent such
intrusion or leaks, and the secret keeper exerts this energy, in part, because
he or she perceives that other people may have some claim to the hidden
information. It follows that if clients do not feel pressed for disclosure, then
they could view these details as private, rather than as secrets that need to
be suppressed.
Second, even if clients did choose to suppress some details, the evi-
dence that those details would ironically become more Significant is far
from clear. Wegner and his colleagues have demonstrated that when peo-
ple are given thoughts to suppress, they become more preoccupied with
those thoughts than if they had not suppressed them (e.g., Wegner et aI.,
1987). However, one pair of experiments (Kelly & Kahn, 1994) showed that
when participants were asked to suppress their own unwanted intrusive
thoughts, they actually either became less preoccupied with the thoughts
(Exp. 1) or did not experience any change in the frequency of the thoughts
(Exp. 2). Also, the results from Pennebaker's (1997b) studies may not be
relevant to the present discussion, because those studies focused on reve-
lations in confidential, anonymous settings (i.e., they were not designed to
look at the effects of revealing to an important audience). The crux of my
argument for why it is acceptable for clients to censor some particularly
heinous details is that the perceived feedback from an important audience
(Le., their therapists) could have negative implications for their self-
concepts.

TYPE OF SECRET

As mentioned in Chapter 1, to date, the research comparing types of

secrets has focused primarily on traumatic, negative secret keeping. This
research has yet to support the idea that the type of negative or traumatic
secret one is keeping (e.g., about having experienced incest vs. having
been divorced several times) plays a role in whether one would benefit
from revealing that secret to others. For example, when Pennebaker and
Q'Heeron (1984) surveyed spouses of people who had committed suicide
or died in a car crash, they expected that the spouses of suicide victims
WHEN TO REVEAL PERSONAL SECRETS 211

would talk less about the trauma with others than would the spouses of
the car crash victims due to the stigma associated with suicide. They also
expected that the spouses of suicide victims would have poorer psycho-
logical functioning as a result of not discussing the trauma with others.
However, whether the victims committed suicide or died in a car crash
was not predictive of spouses' talking about the trauma or of their psycho-
logical functioning.
One reason for the lack of support for the idea that the type of trau-
matic secret is predictive of benefits from revealing is that there is a
tremendous amount of variation in response to life stressors (e.g., Penne-
baker, 1990; Silver et al., 1983; Silver & Wortman, 1980; Wortman & Silver,
1987,1989). For instance, when my students and I (Kelly, Coenen, & John-
ston, 1995) asked undergraduates to identify and evaluate their most
traumatic life events, some individuals reported the breakup of a romantic
relationship as their most traumatic life experience and rated it as "ex-
tremely disturbing," whereas others listed having been sexually molested
by a close relative or having been violently raped as their most traumatic
life event and rated it as somewhat less disturbing. Moreover, Silver et al.
(1983) found that among a group of survivors of incest, although the
majority reported a strong desire to make meaning of the incest experi-
ence, a small subset of them reported that they felt no need to make
meaning out of the experience. Tait and Silver (1989) have argued that the
type of major negative life event a person experiences is not what causes
adjustment difficulties. What seems to be critical in determining how one
responds to traumatic life events is how one interprets the experience (e.g.,
Epstein, 1985; Kelly, 1955).
Further research is necessary to compare the consequences of reveal-
ing various types of secrets: ones that are perceived to be negative or
traumatic compared with ones that are perceived to be pleasant, such as
having a high opinion of oneself. Researchers could explore how the
controllability of the secret (from the potential listener's perspective),
responsibility that the secret keeper has for the secret, number of people
who are affected by the secret, and type of emotion associated with the
secret may influence the outcome of revealing. In this section, I detail how
the type of secret adds to the complexity of the model for when to reveal
secrets in a particular relationship.

Controllability
Researchers studying social support have depicted controllability of
stressful life events as the central organizing dimension for understanding
what types of social support (e.g., instrumental vs. emotional support)
victims need from their confidants (Cutrona, 1990; Cutrona & Russell, 1990;
212 CHAPTER 9

Lazarus & Folkman, 1984). In a similar vein, I content that the controllabil-
ity of a secret may be a central dimension in predicting whether a secret
keeper will benefit from sharing a secret. As described in the previous
chapter, there is evidence that depressed people tend to get rejected when
they appear to be suffering or coping poorly with their problems (e.g.,
Silver & Wortman, 1980). However, if the secret seems controllable to the
relational partner, then the partner most likely will not be burdened or
overly distressed by the revelation and will be unlikely to reject the secret
keeper. The secret keeper may wish to present her secret to a relational
partner in such a way that the partner perceives that he can help with the
secret disclosed. For instance, a rape victim may manage to tell her partner
that what she really needs is for him to listen to her story and to accept and
support her.

Responsibility
Whether a secret keeper is perceived as personally responsible for a
problematic secret complicates the timing of the revelation. For example, if
a student has cheated on a test and the teacher is likely to discover this
cheating before the end of the semester, the student would probably get a
more sympathetic response by admitting to the cheating sooner rather
than later in the semester. Evidence for this claim comes from a study by
Jones and Gordon (1972). They asked undergraduates to listen to an osten-
sible interview between a student and his academic advisor. There were
two manipulations: The student on the tape had missed a semester of high
school either because the student had engaged in cheating and plagiarism
(personal responsibility condition) or because of the litigation surrounding
his parents' divorce (no responsibility condition). The other manipulation
was that the student revealed the negative information very early in the
conversation or during a later segment when he was asked directly why he
had transferred high schools. As it turned out, the participants liked the
student who was responsible for the negative event (i.e., cheating) more
when he disclosed that information early in the conversation rather than
later when he was forced to "give it up." At the same time, the participants
liked the student who was not responsible for the negative event more
when he revealed it later rather than sooner. These findings were inter-
preted to mean that in the former condition, the student was seen as demon-
strating trustworthiness by admitting to the act early on, whereas the
student in the latter condition was seen as trying to obtain pity by reveal-
ing an event for which he was not responsible early in the dialogue. Thus,
the findings provide support for the notion that revealing a secret for which
one has personal responsibility could backfire if it is revealed too late.
WHEN TO REVEAL PERSONAL SECRETS 213

Communality
Secrets often have adverse implications for other people besides the
secret keeper. In deciding whether to reveal such secrets, secret keepers
must take into account the collective consequences of such a decision. For
example, if a woman is having an affair with her husband's best friend and
business partner and she reveals this information to her husband, then her
lover is likely to lose both his best friend and his job as a result. Many cases
of physical and sexual abuse, criminal activity, cheating, lying, sexual
indiscretion, and drug abuse either tacitly or explicitly involve other indi-
viduals who would be affected by the keeper's decision to reveal the
secret. There also are times when the relational partners are likely to be
hurt by a revelation, either because of their emotional ties to the secret
keeper or because the secret involves them personally. For instance, if a
man tells his wife that he is secretly lusting after another woman, his wife
is likely to become jealous and depressed, and he may suffer as a result of
seeing her this way. As such, the decision to reveal secrets that involve
others is substantially more complex than the decision to reveal those that
do not.

Emotion
There seem to be contradictions in the literature concerning whether
disclosing secrets is associated with immediate or delayed relief for the
secret keeper (e.g., see Pennebaker, 1985, 1990; Pennebaker & Beall, 1986).
I speculate that if the secret involves guilt, typically around something the
secret keeper has done (see Baumeister, Stillwell, & Heatherton,1994), then
the secret keeper may experience immediate benefits from disclosing.
However, if the secret involves humiliation or anguish, typically around
something that was done to the secret keeper, then the secret keeper may
experience a delay before benefiting from revealing, assuming that the
confidant is an appropriate one.
Clinical researchers have suggested that if clients are to benefit from
therapy, they must first work through painful experiences (e.g., Courtois,
1992; Horowitz, 1986; Rando, 1993; Reichert, 1994). For example, studies of
the treatment of anxiety disorders have shown that confronting or expos-
ing oneself to feared stimuli as curative (Foa & Kozak, 1986; Foa et al., 1991;
Kozak et al., 1988; Rachman, 1980). Similarly, research has shown that
people feel worse after writing about traumas before they feel better
(Pennebaker, 1990; Pennebaker & Beall, 1986). This pattern may be the
result of their rehashing feelings of pain, humiliation, and anguish that
prior to their writing were not directly influencing them. To be reminded
214 CHAPTER 9

of pain is painful initially, but the writing may then lead to the lasting relief
associated with gaining new insights into the secret (see Chapter 5). In the
case of the suspected criminals (see Pennebaker, 1985, 1990), an explana-
tion for why they felt better immediately upon confessing is that they
could relieve their guilt and could stop expending their cognitive re-
sources to protect their lies. However, it remains to be seen whether this
kind of relief lasts, especially once the negative implications (e.g., prison
time) of their revelations materialize.

SUMMARY

The proposed model for when people should reveal hidden informa-
tion in a particular relationship hinges on whether their relational partners
(1) expect access to the information, (2) are appropriate confidants, and
(3) would discover the information anyway. The model is intended to
make people more aware of their ability to control how much access to
themselves they allow others to have. Based on the fact that the majority of
the previous work that has linked secret keeping with problems is only
correlational, I encourage people to consider the likelihood that there will
be times when keeping a secret is not harmful and may even be beneficial.
In particular, I hope that they will make revelations with discretion and
will evaluate closely whether their relational partners are discreet, non-
judgmental, and not rejecting of them. In cases where the relational part-
ners do not have these fine qualities, I have offered alternatives to com-
plete revelation, such as telling the partner that certain hidden information
is private, revealing the theme as opposed to the details of the secret, and
revealing the secret to another, more appropriate confidant. The decision
to reveal a secret is complicated further by a consideration of the types of
secrets people keep. For example, someone might decide that even if she
personally will benefit from revealing, she may not want to reveal because
revealing might hurt someone else too much. I have described several
factors that add to the complexity of the model, which are how controllable
the secret is (from the perspective of the relational partner), how respon-
sible the secret keeper is for the secret, how many people the secret
involves, and the kind of emotion that is associated with the secret.

TESTING THE MODEL WITH SAMPLE SECRETS

The model may sound reasonable to the reader so far, but how well
does it hold up when applied to some of the most common, vexing secrets
WHEN TO REVEAL PERSONAL SECRETS 215

that people are likely to have in their close relationships? I am going to use
sexual examples to "test" the model, because these have been the most
frequently reported kinds of secrets (see Chapter 1).

HERPES

An example of a common stigmatizing sexual secret that comes to

mind immediately is having genital herpes, which affects roughly 22% of
adult Americans. Imagine that a man with genital herpes is entering a new
sexual relationship with a woman who knows many of his friends and
colleagues. Should he tell her about the herpes? The first step of the model
is to ask himself whether she expects access to this information. The
chances are that she does, especially since sexual contact with him puts her
at risk for contracting the disease. Many people today even ask their new
partners if they have any sexually transmitted diseases. Assuming that she
does expect access to this knowledge, he next needs to ask himself if she is
an appropriate confidant. Because she has previously expressed great fear
of contracting a sexually transmitted disease, he fears that she might reject
him for the disclosure. Also, she is closer to one of his female colleagues
than she is to him, so he wonders if she would tell that colleague, especially
if the relationship breaks up. To be safe, he concludes that she may not be
an appropriate confidant. Next, he must ask himself whether she is likely
to find out about the herpes, and the chances are good that she will, either
because she might see his herpes suppression medication or sees the sores
on his genitals, or worse yet develops sores for the first time herself.
Normally, the model is helpful in offering the suggestion to reveal the
secret by focusing on themes in such cases when the partner is likely to
discover the secret on her own. However, in this case, the secret is quite
specific. It is difficult to come up with any theme that represents herpes
without saying what it is. My suggestion is either to abstain from sex or to
wear condoms and avoid haVing sex when the sores are present while
getting to know the partner well enough to decide whether she is an
appropriate confidant and then either telling the secret if she is appropriate
or ending the relationship.

NUMBER OF SEXUAL PARTNERS

The female protagonist in the movie, "Four Weddings and a Funeral,"

had 33 sexual partners. She simply tells the Hugh Grant character about
the 33 relationships, because he asked her how many partners she had and
does not reject her for the revelation. However, in many relationships, such
a revelation might lead to rejection. If a young man asks his new girlfriend
216 CHAPTER 9

(who has had 33 partners) how many sexual partners she has had, he is
making it obvious that he expects access to this information-the first
step of the model. If he has expressed disdain for women he calls "sluts"
in the past, then she might assume that he is not an appropriate confidant
because he could reject her for the revelation-the second step of the
model. Also, because she had a number of sexual partners in high school,
he might discover this information when they visit her hometown during
the holidays-the third step of the model. My suggestion is that because
he may discover this information on his own, she should tell him the secret
by focusing on the theme and not offering the specifics. She could say,
"Even though I think it is dehumanizing to disclose specific numbers, I can
tell you that I have been sexually active since high school. I have always
used condoms and have been faithful when my boyfriends and I have
agreed to be, as you and I have." This expresses that idea that there has
been more than one lover and addresses the fears that may underlie the
questions.

CONCLUSION

Secrets are a part of virtually every adult's life. Psychologists long

have believed that keeping secrets is stressful and that it undermines
mental and physical health. Even though they have not been able to show
that keeping secrets causes problems, they recently have offered convinc-
ing experimental evidence that there are health benefits to revealing se-
crets in confidential and anonymous settings. Revealing in a private diary
or journal, especially with an emphasis on gaining new meanings or new
perspectives on the secrets, is likely ultimately to make secret keepers feel
much better on the average.
However, once personal secrets are understood in their relational
context, a number of potential drawbacks to revealing them become evi-
dent, including being rejected by the relational partner and receiving
damage to one's reputation if the partner tells others the secret. It seems
that making the decision to reveal personal secrets to others involves a
trade-off. On the one hand, a secret keeper can feel better by revealing the
secrets and no longer having to feel deceptive toward the partner. On the
other hand, the secret keeper can avoid looking bad and risking rejection
from the partner by not revealing to him or her. The key to making a wise
decision to reveal one's personal secrets is to determine whether the
relational partner is an appropriate confidant: someone who is discreet,
who is perceived by the secret keeper to be nonjudgmental, and who will
not reject the secret keeper. One also must take into account the probability
WHEN TO REVEAL PERSONAL SECRETS 217

of whether the secret will be discovered anyway when deciding to reveal

to a partner, because there are potential benefits associated with putting
one's own spin on undesirable events rather than letting the partner
discover them on his or her own (see Emler, 1990).
In bringing this volume to a close, I wish to highlight several other
points made earlier. One is that high self-concealers are sicker than low
self-concealers. I have tried to explain this observation by offering a pre-
dispositional model of concealment, wherein high self-concealers might be
born with the predisposition to be vulnerable to physical and psychologi-
cal problems. This idea is offered very cautiously and will need to be tested
in future research. An important area of inquiry will be to see whether high
self-concealers are especially vulnerable to specific illnesses the way re-
pressors seem to be vulnerable to cancer. Also interesting will be to see
how one's level of concealment interacts with a decision to reveal or
conceal a secret in influencing the outcomes of that decision. In addition,
independent of a person's self-concealment level, it will be important to
find an ethical way of experimentally testing whether keeping a particular
secret from a relational partner does make a person sick. Scientists need to
determine once and for all whether persons' fears about the dangers of
secrecy are warranted. It may seem odd that I am expecting that self-
concealment (the dispositional quality) and secret keeping (the process
variable) will predict illness in different directions, with keeping a particu-
lar secret possibly serving as a buffer against illness. However, this is
where the discussion of the role of the confidant becomes so crucial. Telling
others something really horrible about oneself offers so much more poten-
tial for damage than keeping that same horrible information a secret,
especially since so many personal secrets actually may more appropriately
be labeled private information anyway (i.e., their relational partners do not
expect access to them). One of the solutions I have offered for times when
some degree of revealing surrounding a heinous and stigmatizing per-
sonal problem is needed, such as in psychotherapy, is to reveal the themes
instead of the really bad details. Even in therapy, people rightfully are
concerned with presenting desirable impressions of themselves to their
therapists, given that there is an interpersonal element to the self-concept,
which can become more negative in the face of real or imagined undesir-
able feedback. It would be nice if people could not care what others think,
the way some autonomy-oriented humanists advocated in the 1960s. How-
ever, given that what others think does seem to playa crucial role in the
way people see themselves, I suggest that it is time for people to acknowl-
edge that process and reveal personal secrets judiciously.
REFERENCES

Abbey, A, Abramis, D. J., & Caplan, R. D. (1985). Effects of different sources of

social support and social conflict on emotional well-being. Basic and Applied
Social Psychology, 6,111-129.
Abraham, N., Torok, M., & Rand, N. (1994). The shell and the kernel: Renewals of
psychoanalysis, Vol. 1. Chicago: University of Chicago Press.
Adams, J. E (1993). The utilization of family secrets as constructive resources in
strategic therapy. Journal of Family Psychotherapy, 4, 19-33.
Ajzen, I. (1977). Information processing approaches to interpersonal attraction. In S.
W. Duck (Ed.), Theory and practice in interpersonal attraction (pp. 51-77). San
Diego, CA: Academic Press.
Alloy, 1. (1988). Cognitive processes in depression. New York: Guilford Press.
Altman, I., & Taylor, D. (1973). Social penetration: The development of interpersonal
relationships. New York: Hold, Rinehart, & Winston.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: Author.
Amini, E, Burke, E. 1., & Edgerton, R. (1978). Social structure of a psychiatric ward
for adolescents and the therapeutic implications of patient staff and intra staff
conflicts. Adolescence, 13, 411-418.
Anderson, C. D. (1981). Expression of affect and physiological response in psycho-
somatic patients. Journal of Psychosomatic Research, 25, 143-149.
Andreasen, N. J. c., & Norris, A S. (1972). Long-term adjustment and adaptation
mechanisms in severely burned adults. Journal of Nervous and Mental Disease,
154, 352-362.
Antonovsky, A (1990). Pathways leading to successful coping and health. In M.
Rosenbaum (Ed.), Learned resourcefulness: On coping skills, self-control, and adap-
tive behavior (Vol. 24, pp. 31-63). New York: Springer.
Archer, R. 1., & Berg, J. H. (1978). Disclosure reciprocity and its limits: A reactance
analysis. Journal of Experimental Social Psychology, 14, 527-540.
Arkin, R. M., & Hermann, A D. (2000). Constructing desirable identities-Self-

219
220 REFERENCES

presentation in psychotherapy and daily life: Comment on Kelly (2000). Psy-

chological Bulletin, 126, 501-504.
Arnow, B. (1996). Cognitive-behavioral therapy for bulimia nervosa. In J. Werne &
I. Yalom (Eds.), Treating eating disorders (pp. 101-142). San Francisco: Jossey-
Bass.
Asner, J. (1990). Reworking the myth of personal incompetence: Group psycho-
therapy for bulimia nervosa. Psychiatric Annals, 20, 395-397.
Avery, N. (1982). Family secrets. Psychoanalytic Review, 69,471-486.
Baer, R, Hinkle, S., Smith, K., & Fenton, M. (1980). Reactance as a function of actual
versus projected autonomy. Journal of Personality and Social Psychology, 38,
416-422.
Baldwin, M. W. (1992). Relational schemas and the processing of information.
Psychological Bulletin, 112, 461-484.
Bandura, A., Lipsher, D. H., & Miller, P. E. (1960). Psychotherapists' approach-
avoidance reactions to patients' expressions of hostility. Journal of Consulting
Psychology, 24, 1-8.
Barak, A., & laCross, M. B. (1975). Multidimensional perception of counselor
behavior. Journal of Counseling Psychology, 22, 471-476.
Barger, S. D., Kircher, J. c., & Croyle, R T. (1997). The effects of social context and
defensiveness on the physiological responses of repressive copers. Journal of
Personality and Social Psychology, 73,1118-1128.
Barron, J. W, Beaumont, R, Goldsmith, G., & Good, M. I. (1991). Sigmund Freud:
The secrets of nature and the nature of secrets. International Review of Psycho-
Analysis, 18, 143-163.
Baumeister, R E (1982). A self-presentational view of social phenomena. Psychologi-
cal Bulletin, 91, 3-26.
Baumeister, R E (1990). Suicide as escape from self. Psychological Review, 97, 90-113.
Baumeister, R E (1991). Escaping the self. New York: Basic Books.
Baumeister, R E (1998). The self. In D. T. Gilbert, S. T. Fiske, & G. Lindzay (Eds.),
The handbook of social psychology (Vol. 2,4th ed., pp. 680-740). Boston: McGraw-
Hill.
Baumeister, R E, & Leary, M. R (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117,
497-529.
Baumeister, R E, & Tice, D. M. (1984). Role of self-presentation and choice in
cognitive dissonance under forced compliance: Necessary or sufficient causes?
Journal of Personality and Social Psychology, 46, 5-13.
Baumeister, R E, Stillwell, A. M., & Heatherton, T. E (1994). Guilt: An interpersonal
approach. Psychological Bulletin, 115, 243-267.
Baxter, L. A., & Wilmot, W W (1985). Taboo topics in close relationships. Journal of
Social and Personal Relationships, 2, 253-269.
Bellman, B. L. (1981). The paradox of secrecy. Human Studies, 4, 1-24.
Bern, D. J. (1967). Self-perception: An alternative interpretation of cognitive disso-
nance phenomena. Psychological Review, 74,183-200.
Bern, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in experi-
mental social psychology (Vol. 6, pp. 1-62). San Diego, CA: Academic Press.
REFERENCES 221

Berg, J. H. (1987). Responsiveness and self-disclosure. In V. J. Derlega & J. H. Berg

(Eds.), Self-disclosure: Theory, research, and therapy (pp.101-130). New York: Plenum.
Berg, J. H., & Archer, R L. (1980). Disclosure or concern: A second look at liking for
the norm-breaker. Journal of Personality, 48,245-257.
Bergmann, M. V. (1992). An infantile trauma, a trauma during analysis, and their
psychic connexion. International Journal of Psycho-Analysis, 73, 447-454.
Bernfeld, S. (1941). The facts of observation in psychoanalysis. The Journal of Psychol-
ogy, 12,289-305.
Berry, D. S., & Pennebaker, J. W. (1993). Nonverbal and verbal emotional expression
and health. Psychotherapy and Psychosomatics, 59, 11-19.
Beutler, L. E., Machado, P. P. P., & Neufeldt, S. A (1994). Therapist variables. In A E.
Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th
ed., pp. 229-269). New York: John Wiley & Sons.
Biederman, J., Rosenbaum, J. E, Hirshfeld, D. R, Faraone, S. V. (1990). Psychiatric
correlates of behavioral inhibition in young children of parents with and
without psychiatric disorders. Archives of General Psychiatry, 47, 21-26.
Binder, R (1981). Why women don't report sexual assault. Journal of Clinical Psychia-
try, 42, 437-438.
Bingham, A, & Bargar, J. (1985). Children of alcoholic families: A group treatment
approach for latency age children. Journal of Psychosocial Nursing and Mental
Health Services, 23, 13-15.
Black, C. (1981). Innocent bystanders at risk: The children of alcoholics. Alcoholism,
22-26.
Bloch, S., & Reibstein, J. (1980). Perceptions by patients and therapists of therapeu-
tic factors in group psychotherapy. British Journal of Psychiatry, 137, 274-278.
Bohart, A C. (1980). Toward a cognitive theory of catharsis. Psychotherapy: Theory,
Research & Practice, 17, 192-201.
Bohart, A, Allen, E., Jackson, M., & Freyer, R (1976). Understanding-sharing versus
discharge in interpersonal counseling. Paper presented at the 56th annual conven-
tion of the Western Psychological Association, Los Angeles, 1976. Accepted for
inclusion in the ERIC/CAPS system.
Bok, S. (1982). Secrets: On the ethics of concealment and revelation. New York: Pantheon
Books.
Bolger, N., Foster, M., Vinokur, A D., & Ng, R (1996). Close relationships and
adjustments to a life crisis: The case of breast cancer. Journal of Personality and
Social Psychology, 70,283-294.
Booth, R J., Petrie, K. J., & Pennebaker, J. W. (1997). Changes in circulating lympho-
cyte numbers following emotional disclosure: Evidence of buffering? Stress
Medicine, 13, 23-29.
Bootzin, R R (1997). Examining the theory and clinical utility of writing about
emotional experiences. Psychological Science, 8, 167-169.
Bor, R, Miller, R, Scher, I., & Salt, H. (1991). The practice of counseling HIV / AIDS
clients. British Journal of Guidance and Counseling, 19, 129-138.
Braginsky, B. M., & Braginsky, D. D. (1967). Schizophrenic patients in the psychi-
atric interview: An experimental study of the manipulative tactics of mental
patients. Journal of Consulting Psychology, 31, 543-547.
222 REFERENCES

Braginsky, B. M., Grosse, M., & Ring, K. (1966). Controlling outcomes through
impression management: An experimental study of the manipulative tactics of
mental patients. Journal of Consulting Psychology, 30, 295-300.
Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance: A theory of freedom and
control. New York: Academic Press.
Breuer, J., & Freud, S. (1975). Studies in hysteria. New York: Basic Books. (Original
work published 1893-1895)
Brewer, M. D., & Mittelman, J. (1980). Effects of normative control of self-disclosure
on reciprocity. Journal of Personality, 48, 89-102.
Brickman, P., Rabinowitz, V, Karuza, J., Coates, D., Cohn, E., & Kidder, 1. (1982).
Models of helping and coping. American Psychologist, 37, 368-384.
Broadhead, W. E., Kaplan, B. H., James, S. A, Wagner, E. H., Schoenbach, V J.,
Grimson, R., Heyden, S., Tibblin, G., & Gehlbach, S. H. (1983). The epidemi-
ologic evidence for a relationship between social support and health. American
Journal of Epidemiology, 117, 521-537.
Brown, E. (1991). Dealing with secret affairs in psychotherapy. In E. Brown (Ed.),
From patterns of infidelity and their treatment (pp. 53-73). New York: Brunner-
Mazel.
Buchele, B. J. (1993). Group psychotherapy for persons with multiple personality
and dissociative disorders. Bulletin of the Menninger Clinic, 57, 362-370.
Buck, R. W. (1984). The communication of emotion. New York: Guilford.
Burger, J. M. (1997). Personality (4th ed.). Pacific Grove, CA: Brooks/Cole.
Burgess, A W., & Holmstrom, 1. 1. (1974). Rape trauma syndrome. American
Journal of Psychiatry, 131, 981-986.
Bush,1. K., Barr, C. 1., McHugo, G. J., & Lanzetta, J. T. (1989). The effects of facial
control and facial mimicry on subjective reactions to comedy routines. Motiva-
tion and Emotion, 13, 31-52.
Bushman, B. J., Baumeister, R. E, & Stack, A D. (1999). Catharsis, aggression, and
persuasive influence: Self-fulfilling or self-defeating prophecies? Journal of
Personality and Social Psychology, 76,367-376.
Buss, A H., & Briggs, S. R. (1984). Drama and the self in social interaction. Journal of
Personality and Social Psychology, 47, 1310-1324.
Butler, S. (1978). Conspiracy of silence: The trauma of incest. San Francisco: New Glide.
Caracena, P. E, & Victory, J. R. (1969). Correlates of the phenomological and judged
empathy. Journal of Counseling Psychology, 16, 510-515.
Carlsmith, J. M., Ellsworth, P., & Whiteside, J. 1. (1968). Guilt confession and
compliance. Unpublished manuscript, as cited in Freedman, J. 1., Carlsmith, J.
M., & Sears, D. O. (1970). Social psychology. Englewood Cliffs, NJ: Prentice Hall.
Carver, C. S., & Scheier, M. E (1985). Aspects of self and the control of behavior. In B.
R. Schlenker (Ed.), The self and social life (pp.146-174). New York: McGraw-Hill.
Casriel, D. (1972). A scream away from happiness. New York: Grosset & Dunlap.
Cass, V (1979). Homosexual identity formation: A theoretical model. Journal of
Homosexuality, 4, 219-235.
Castets, B. (1988). The place of secrecy in analysis. Psychiatric Francaise, 19, 15-18.
Cepeda-Benito, A, & Short, P. (1998). Self-concealment, avoidance of psychological
services, and perceived likelihood of seeking professional help. Journal of
Counseling Psychology, 45, 58-64.
REFERENCES 223

Chafetez, J., Sampson, P., Beck, P., & West, J. (1974). A study of homosexual women.
Social Work, 19, 714-723.
Chaikin, A. L., & Derlega, V. J. (1974). Liking for the norm-breaker in self-disclosure.
Journal of Personality, 42, 117-129.
Champion, C. D. (2001). Effects of deception on the perceptions of those deceived.
Unpublished manuscript. South Bend, IN: University of Notre Dame.
Cheek, J. M. (1982). Aggregation, moderator variables, and the validity of person-
ality tests: A peer-rating study. Journal of Personality and Social Psychology, 43,
1254-1269.
Chevalier, A. J. (1995). On the client's path: A manual for the practice of solution-focused
therapy. Oakland, CA: New Harbinger Publications.
Chodoff, P., Friedman, S. B., & Hamburg, D. A. (1964). Stress, defenses, and coping
behavior: Observations in parents of children with malignant disease. Ameri-
can Journal of Psychiatry, 120, 743-749.
Christophe, B., & Rime, B. (1997). Exposure to the social sharing of emotion:
Emotional impact, listener responses and secondary social sharing. European
Journal of Social Psychology, 27, 37-54.
Christophe, v., & Di Giacomo, J. P. (1995). Contenu du partage social secondaire suite a
un episode emotionnel negatif ou positif. Unpublished manuscript.
Cialdini, R B., Schaller, M., Houlihan, D., Arps, K., Fultz, J., & Beaman, A. L. (1987).
Empathy-based helping: Is it selflessly or selfishly based? Journal of Personality
and Social Psychology, 52, 749-758.
Cioffi, D. (1996). Making public the private: Possible effects of expressing somatic
experience. Psychology and Health, 11, 203-222.
Cioffi, D., & Holloway, J. (1993). Delayed costs of suppressed pain. Journal of
Personality and Social Psychology, 64, 274-282.
Cline, R J. W., & McKenzie, N. J. (2000). Dilemmas of disclosure in the age of HIV /
AIDS: Balancing privacy and protection in the health care context. In S. Pe-
tronio (Ed.), Balancing the secrets of private disclosure (pp. 71-82). Mahwah, NJ:
Lawrence Erlbaum.
Coates, D., Wortman, C. B., & Abbey, A. (1979). Reactions to victims. In I. H. Frieze,
D. Bar-Tal, & J. S. Carroll (Eds.), New approaches to social problems (pp. 21-52).
San Francisco: Jossey-Bass.
Cochrane, N., & Neilson, M. (1977). DepreSSive illness: The role of aggression
further considered. Psychological Medicine, 7, 283-288.
Colby, C. Z., Lanzetta, J. T., & Kleck, R E. (1977). Effects of the expression of pain on
autonomic and pain tolerance responses to subject-controlled pain. Psycho-
physiology, 14, 537-540.
Cole, S. W., Kemeny, M. E., Taylor, S. E., & VISscher, B. R (1996a). Elevated physical
health risk among gay men who conceal their homosexual identity. Health
Psychology, 15, 243-251.
Cole, S. w., Kemeny, M. E., Taylor, S. E., Visscher, B. R, & Fahey, J. L. (1996b).
Accelerated course of human immunodeficiency virus infection in gay men
who conceal their homosexual identity. Psychosomatic Medicine, 58, 219-231.
Cole, S. W., Kemeny, M. E., & Taylor, S. E. (1997). Social identity and physical health:
Accelerated HIV progression in rejection-sensitive gay men. Journal of Person-
ality and Social Psychology, 72, 320-335.
224 REFERENCES

Cole, T. (2001). Lying to the one you love: The use of deception in romantic
relationships. Journal of Social and Personal Relationships, 18, 107-129.
Collins, N. L., & Miller, L. C. (1994). Self-disclosure and liking: A meta-analytic
review. Psychological Bulletin, 116, 457-475.
Consumer Reports. (1995). Mental health: Does therapy help? November, pp. 734-739.
Cooley, C. H. (1902). Human nature and the social order. New York: Charles Scribner's
Sons.
Coons, P. M. (1986). Treatment progress in 20 patients with multiple personality
disorder. The Journal of Neroous and Mental Disease, 174, 715-721.
Corcoran, K. J. (1988). The Journal of Psychology, 122, 193-195.
Cornwell, J., Nurcombe, B., & Stevens, L. (1977). Family response to loss of a child
by sudden infant death syndrome. The Medical Journal of Australia, 1, 656-658.
Coupland, J., Coupland, H., Giles, H., & Wieman, J. (1988). My life in your hands:
Processes of self-disclosure intergenerational talk. In N. Coupland (Ed.), Styles
of discourse (pp. 201-253). London: Croon Helm.
Courtois, C. (1992). The memory retrieval process in incest survivor therapy. Journal
of Child Sexual Abuse, 1, 15-31.
Cox, T., & McCay, C. (1982). Psychosocial factors and psychophysiological mecha-
nisms in the aetiology and development of cancers. Social Science and Medicine,
16, 381-396.
Coyne, J. c., Kessler, R. c., Tal, M., Turnbull, J., Wortman, C. B., & Greden, J. F.
(1987). Living with a depressed person. Journal of Consulting and Clinical Psy-
chology, 55, 347-352.
Cozby, P. C. (1972) Self-disclosure, reciprocity, and liking. Sociometry, 35, 151-160.
Cramer, K. M., & Barry, J. E. (1999). Psychometric properties and confirmatory
factor analysis of the Self-Concealment Scale. Personality and Individual Differ-
ences, 27, 629-637.
Cramer, K. M., & Lake, R. P. (1998). The Preference for Solitude Scale: Psychometric
properties and factor structure. Personality and Individual Differences, 24,
193-199.
Cronbach, L. (1955). Processes affecting scores on "understanding of others" and
"assumed similarity." Psychological Bulletin, 52, 177-193.
Cutrona, C. E. (1986). Behavioral manifestations of social support: A microanalytic
investigation. Journal of Personality and Social Psychology, 51, 201-208.
Cutrona, C. E. (1990). Stress and social support-in search of optimal matching.
Journal of Social and Clinical Psychology, 59, 3-14.
Cutrona, C. E., & Russell, D. (1990). Type of social support and specific stress:
Toward a theory of optimal matching. In I. G. Sarason, B. R. Sarason, & G. R.
Pierce (Eds.), Social support: An interactional view (pp. 319-336). New York:
Wiley.
Dalto, C. A., Ajzen, I., & Kaplan, K. J. (1979). Self-disclosure and attraction: Effects
of intimacy and desirability on beliefs and attitudes. Journal of Research in
Personality, 13, 127-138.
Dattore, P. J., Shontz, F. c., & Coyne, L. (1980). Premorbid personality differentia-
tion of cancer and noncancer groups: A test of the hypothesis of cancer
proneness. Journal of Consulting and Clinical Psychology, 48, 388-394.
REFERENCES 225

Davanioo, H. (1980). A method of short-term dynamic psychotherapy. In H.

Davanloo (Ed.), Short-term dynamic psychotherapy (pp. 32-64). New York: Jason
Aronson.
Davidowitz, M., & Myriak, R D. (1984). Responding to the bereaved: An analysis of
"helping" statements. Research Record, 1, 35-42.
Davis, C. G., Nolen-Hoeksema, S., & Larson, J. (1998). Making sense of loss and
benefiting from the experience: Two construals of meaning. Journal of Person-
ality and Social Psychology, 75,561-574.
Davis, J. D., & Sloan, M. L. (1974). The basis of interviewee matching and inter-
viewer self-disclosure. British Journal of Social and Clinical Psychology, 13, 359-367.
Davis, K. L., & Meara, N. M. (1982). So you think it is a secret. Journal for Specialists in
Group Work, 7,149-153.
Davis, P. J., & Schwartz, G. E. (1987). Repression and the inaccessibility of emotional
memories. Journal of Personality and Social Psychology, 52, 155-162.
Denollet, J. (1991). Negative affectivity and repressive coping: Pervasive influence
on self-reported mood, health, and coronary-prone behavior. Psychosomatic
Medicine, 53, 538-556.
Denollet, J. (1993). Biobehavioral research on coronary heart disease: Where is the
person? Journal of Behavioural Medicine, 16, 115-141.
DePaulo, B. M., Kashy, D. A, Kirkendol, S. E., Wyer, M. M., & Epstein, J. A (1996).
Lying in everyday life. Journal of Personality and Social Psychology, 70, 979-995.
Derlega, V. J. (1988). Self-disclosure: Inside or outside the mainstream of social
psychological research? In M. R Leary (Ed.), The state of social psychology.
Journal of Social Behavior and Personality, 3, 27-34.
Derlega, V. J., & Grzelak, J. (1979). Appropriateness of self-disclosure. In G. J.
Chelune (Ed.), Self-disclosure: Origins, patterns, and implications of openness in
interpersonal relationships (pp. 151-176). San Francisco: Jossey-Bass.
Derlega, V. J., Metts, S., Petronio, S., & Margulis, S. T. (1993). Self-disclosure. Newb-
ury Park, CA: Sage.
Derogatis, L. R (1993). Brief Symptom Inventory (BSI): Administration, scoring, and
procedures manual (3rd ed.). Minneapolis, MN: National Computer Systems.
Derogatis, L. R, Abeloff, M. D., & Melisaratos, N. (1979). Psychological coping
mechanisms and survival time in metastatic breast cancer. Journal of the Ameri-
can Medical Association, 242, 1504-1508.
DiMatteo, M., & Hays, R (1981). Social support and serious illness. In B. Gottlieb
(Ed.), Social networks and social support (pp. 117-148). Beverly Hills, CA: Sage.
Doka, K. J., & Schwartz, E. (1978). Assigning blame: The restoration of the "senti-
mental order" follOWing an accidental death. Omega, 9,279-285.
Donnelly, D. A, & Murray, E. J. (1991). Cognitive and emotional changes in written
essays and therapy interviews. Journal of Social and Clinical Psychology, 10,
334-350.
Dowd, E. T., & Milne, C. R (1986). Paradoxical interventions in counseling psychol-
ogy. Counseling Psychologist, 14, 237-282.
Dowd, E. T., Milne, C. R, & Wise, S. L. (1991). The Therapeutic Reactance Scale: A
measure of psychological reactance. Journal of Counseling and Development, 69,
541-545.
226 REFERENCES

Duck, S. W. (1988). Relating to others. Pacific Grove, CA: Brooks/Cole.

Dushman, RD., & Bressler, M. J. (1991). Psychodrama in an adolescent chemical
dependency treatment program. Individual Psychology, 47, 515-520.
Eaker, B. (1986). Unlocking the family secret in family play therapy. Child and
Adolescent Social Work, 3, 235-253.
Ebbesen, E. B., Duncan, B., & Konecni, V. J. (1975). Effects of content of verbal
aggression on future verbal aggression: A field experiment. Journal of Experi-
mental Social Psychology, 11, 192-204.
Ehrlich, H. J., & Bauer, M. L. (1967). Therapists' feelings toward patients and patient
treatment outcome. Social Science and Medicine, I, 283-292.
Ekman, P. (1991). Telling lies: Clues to deceit in the marketplace, politics, and marriage.
New York: W. W. Norton.
Ekman, P., & O'Sullivan, M. (1991). Who can catch a liar? American Psychologist, 46,
913-920.
Elliott, R (1985). Helpful and nonhelpful events in brief counseling interviews: An
empirical taxonomy. Journal of Counseling Psychology, 32, 307-322.
Emler, N. (1990). A social psychology of reputation. European Review of Social
Psychology, I, 171-193.
Engebretson, T. 0., Matthews, K. A, & Scheier, M. F. (1989). Relations between
anger expression and cardiovascular reactivity: Reconciling inconsistent find-
ings through a matching hypothesis. Journal of Personality and Social Psychology,
57, 513-521.
Epstein, S. (1985). The implications of cognitive-experiential self-theory for re-
search in social psychology and personality (special issue). Journal for the
Theory of Social Behavior, IS, 283-310.
Esterling, B. A, Antoni, M. H., Kumar, M., & Schneiderman, N. (1990). Emotional
repression, stress disclosure responses, and Epstein-Barr viral capsid antigen
titers. Psychosomatic Medicine, 52, 397-410.
Esterling, B. A, Antoni, M. H., Kumar, M., & Schneiderman, N. (1993). Defensive-
ness, trait anxiety, and Epstein-Barr viral capsid antigen antibody titers in
healthy college students. Health Psychology, 12, 132-139.
Esterling, B. A, Antoni, M. H., Fletcher, M. A, Marguiles, 5., & Schneiderman, N.
(1994). Emotional disclosure through writing or speaking modulates latent
Epstein-Barr virus antibody titers. Journal of Consulting and Clinical Psychology,
62, 130-140.
Esterling, B. A, L' Abate, L., Murray, E. J., & Pennebaker, J. W. (1999). Empirical
foundations for writing in prevention and psychotherapy: Mental and physi-
cal health outcomes. Clinical Psychology Review, 19,79-96.
Eugster, S. L., & Wampold, B. E. (1996). Systematic effects of participant role on
evaluation of the psychotherapy session. Journal of Consulting and Clinical
Psychology, 64, 1020-1028.
Evans, N. (1976). Mourning as a family secret. Journal of the American Academy of
Child Psychiatry, IS, 502- 509.
Fawzy, F. I., Fawzy, N. W., Hyun, C. 5., Elashoff, R, Guthrie, D., Fahey, J. L., &
Morton, D. L. (1993). Malignant melanoma: Effects of an early structured
REFERENCES 227

psychiatric intervention, coping, and affective state on recurrence and survival

6 years later. Archives of General Psychiatry, 50, 681-689.
Fazio, R H., Effrein, E. A, & Falender, V. J. (1981). Self-perceptions following social
interaction. Journal of Personality and Social Psychology, 41, 232-242.
Feldman, D. A, Strong, S. R, & Danser, D. B. (1982). A comparison of paradoxical
and nonparadoxical interpretations and directives. Journal of Counseling Psy-
chology, 29, 572-579.
Finkenauer, c., & Hazam, H. (2000). Disclosure and secrecy in marriage: Do both
contribute to marital satisfaction? Journal of Social and Personal Relationships, 17,
245-263.
Finkenauer, c., & Rime, B. (1998a). Socially shared emotional experiences vs.
emotional experiences kept secret: Differential characteristics and conse-
quences. Journal of Social and Clinical Psychology, 17, 295-318.
Finkenauer, c., & Rime, B. (1998b). Keeping emotional memories secret: Health and
subjective well-being when emotions are not shard. Journal of Health Psychol-
ogy, 3, 47-58.
Fish, J. M. (1970). Empathy and the reported emotional experiences of beginning
psychotherapists. Journal of Consulting and Clinical Psychology, 35, 64-69.
Fish, J. M. (1996). Prevention, solution-focused therapy, and the illusion of mental
disorders. Applied and Preventive Psychology,S, 37-40.
Fish, V., & Scott, C. G. (1999). Childhood abuse recollections in a nonclinical
population: Forgetting and secrecy. Child Abuse and Neglect, 23, 791-802.
Fishbein, M. J., & Laird, J. D. (1979). Concealment and disclosure: Some effects of
information control on the person who controls. Journal of Experimental Social
Psychology, 15, 114-121.
Flowers, J. v., Armentrout, S. A, Booraem, C. D., Kraft, M. E., Maddi, S. R, &
Wadhwa, P. D. (1995). The association between psychosocial factors and the
incidence of breast cancer. Annals of Behavioral Medicine, 17, S081.
Foa, E. B., & Kozak, M. J. (1986). Emotional processing of fear: Exposure to correc-
tive information. Psychological Bulletin, 99, 20-35.
Foa, E. B., & Rothbaum, B. O. (1989). Behavioural psychotherapy for posttraumatic
stress disorder. International Review of Psychiatry, 1, 219-226.
Foa, E. B., Rothbaum, B. 0., Riggs, D. S., & Murdock, T. B. (1991). Treatment of
posttraumatic stress disorder in rape victims: A comparison between
cognitive-behavioral procedures and counseling. Journal of Consulting and
Clinical Psychology, 59,715-723.
Foa, E. B., Molnar, c., & Cashman, L. (1995). Changes in rape narratives during
exposure therapy for posttraumatic stress disorder. Journal of Traumatic Stress,
8,675-690.
Fong, M. L., & Cox, B. G. (1983). Trust as an underlying dynamic in the counseling
process: How clients test trust. The Personnel and Guidance Journal, 63, 163-166.
Frable, D. E. S., Wortman, c., & Joseph, J. (1997). Predicting self-esteem, well-being,
and distress in a cohort of gay men: The importance of cultural stigma,
personal Visibility, community networks, and positive identity. Journal of Per-
sonality, 65, 599-624.
228 REFERENCES

Frable, D. E. S., Platt L., & Hoey, S. (1998). Concealable stigmas and positive self-
perceptions: Feeling better around similar others. Journal of Personality and
Social Psychology, 74,909-922.
Frank, J. D. (1974). Therapeutic components of psychotherapy. A 25-year progress
report of research. The Journal of Nervous and Mental Disease, 159, 325-342.
Frankl, V. E. (1976/1959). Man's search for meaning. New York: Pocket.
Frederikson, M., & Engel, B. T. (1985). Cardiovascular and electrodermal adjust-
ments during a vigilance task in patients with borderline and established
hypertension. Journal of Psychosomatic Research, 29, 235-246.
Freud, S. (1958). On the beginning of treatment: Further recommendations on the tech-
niques of psychoanalysis. London: Hogarth Press. (Original work published 1913)
Friedlander, M. L., & Schwartz, G. S. (1985). Toward a theory of strategic self-
presentation in counseling and psychotherapy. Journal of Counseling Psychol-
ogy, 32, 483-501.
Friedman, H. S., Hall, J. A, & Harris, M. J. (1985). Type A behavior, nonverbal
expressive style and health. Journal of Personality and Social Psychology, 48,
1299-1315.
Fromm-Reichmann, E (1950). Principles of intensive psychotherapy. Chicago: Univer-
sity of Chicago Press.
Fuller, E, & Hill, c. E. (1985). Counselor and helpee perceptions of counselor
intentions in relation to outcome in a single counseling session. Journal of
Counseling Psychology, 32, 329-338.
Garfield, S. L. (1980). Psychotherapy: An eclectic approach. New York: Wiley.
Garfield, S. L. (1991). Common and specific factors in psychotherapy. Journal of
Integrative and Eclectic Psychotherapy, 10, 5-13.
Garfield, S. L. (1994). Research on client variables in psychotherapy. In A E. Bergin
& S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th ed.,
pp. 190-228). New York: Wiley.
Garron, D. c., & Leavitt, E (1979). Demographic and affective covariates of pain.
Psychosomatic Medicine, 41, 525-534.
Geiser, R. (1979). Hidden victims: The sexual abuse of children. Boston: Beacon.
Gelso, C. J., & Carter, J. A (1994). Components of the psychotherapy relationship:
Their interaction and unfolding during treatment. Journal of Counseling Psy-
chology, 41, 296-306.
Gergen, K. J. (1965). Effects of interaction goals and personality feedback on the
presentation of self. Journal of Personality and Social Psychology, 1, 413-424.
Gesell, S. B. (1999). The roles of personality and cognitive processing in secret
keeping (anxiety). Dissertation Abstracts International: Section B: The Sciences and
Engineering, 60(6-B): 2971.
Gillman, R. D. (1992). Rescue fantasies and the secret benefactor. Psychoanalytic
Study of the Child, 47, 279-298.
Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY: Double-
day Anchor.
Goldstein, D. A, & Antoni, M. H. (1989). The distribution of repressive coping
styles among non-metastatic and metastatic breast cancer patients as com-
pared to non-cancer patients. Psychology and Health, 3, 245-258.
REFERENCES 229

Good, G. E., Dell, D. M., & Mintz, L. B. (1989). Male role and gender role conflict:
Relations to help-seeking in men. Journal of Counseling Psychology, 36, 295-300.
Goodwin, S. (1982). Sexual abuse: Incest victims and their families. Boston: John Wright.
Greenberg, M. A, & Stone, A A (1992). Emotional disclosure about traumas and its
relation to health: Effects of previous disclosure and trauma severity. Journal of
Personality and Social Psychology, 63, 75-84.
Greenberg, M. A, Wortman, C. B., & Stone, A A (1996). Emotional expression and
physical health: Revising traumatic memories or fostering self-regulation?
Journal of Personality and Social Psychology, 71, 588-602.
Greer, S. (1983). Cancer and the mind. British Journal of Psychiatry, 143, 535-543.
Greif, G., & Porembski, E. (1988). Implications for therapy with significant others of
persons with AIDS. Journal of Gay and Lesbian Psychotherapy, 1, 79-86.
Grencavage, L. M., & Norcross, J. C. (1990). Where are the commonalities among
the therapeutic common factors? Professional Psychology: Research and Practice,
21, 372-378.
Grolnick, L. (1983). Ibsen's truth, family secrets, and family therapy. Family Process,
22,275-288.
Gross, J. J., & Levenson, R. W. (1993). Emotional suppression: Physiology, self-
report, and expressive behavior. Journal of Personality and Social Psychology, 64,
970-986.
Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: The acute effects of inhibiting
negative and positive emotion. Journal of Abnormal Psychology, 106, 95-103.
Gurtman, M. B. (1986). Depression and the response of others: Re-evaluating the re-
evaluation. Journal of Abnormal Psychology, 95, 99-101.
Gutheil, T., & Avery, N. (1977). Multiple overt incest as family defense against loss.
Family Process, 16, 106-116.
Haley, J. (1963). Strategies of psychotherapy. New York: Grune & Stratton.
Halpern, T. P. (1977). Degree of client disclosure as a function of past disclosure,
counselor disclosure, and counselor facilitativeness. Journal of Counseling Psy-
chology, 24, 41-47.
Handelman, S. (1981). Interpretation as devotion: Freud's relation to Rabbinic
henneneutics. The Psychoanalytic Review, 68, 201-218.
Hansen, J. c., Moore, G. D., & Carkhuff, R. R. (1968). The differential relationships
of objective and client perceptions of counseling. Journal of Clinical Psychology,
24, 244-246.
Harber, K. D., & Pennebaker, J. W. (1992). Overcoming traumatic memories. In S. A
Christianson (Ed.), The handbook of emotion and memory (pp. 359-387). Hills-
dale, NJ: Erlbaum.
Hartley, D., Roback, H. B., & Abramowitz, S. I. (1976). Deterioration effects in
encounter groups. American Psychologist, 247-255.
Hays, K. F. (1987). The conspiracy of silence revisited: Group therapy with adult
survivors of incest. Journal of Group Psychotherapy, Psychodrama, and Sociometry,
39,143-156.
Heilman, M. E., & Garner, K. A (1975). Counteracting the boomerang: The effects
of choice on compliance to threats and promises. Journal of Personality and
Social Psychology, 31, 911-917.
230 REFERENCES

Heilman, M. E., & Toffler, B. L. (1976). Reacting to reactance: An interpersonal

interpretation of the need for freedom. Journal of Experimental Social Psychol-
ogy, 12, 519-529.
Helmrath, T. A., & Steinitz, E. M. (1978). Death of an infant: Parental grieving and
the failure of social support. The Journal of Family Practice, 6, 785-790.
Hencken, J., & O'Dowd, W. T. (1977). Coming out as an aspect of identity formation.
Gay Academic Union Journal: Gai Saber, I, 18-22.
Henley, N. M. (1973). Power, sex, and nonverbal communication. Berkeley Journal of
Sociology, 18, 1-26.
Herman, J. (1981). Father-daughter incest. Cambridge: Harvard University.
Herman, J., & Schatzow, E. (1984). Time-limited group therapy for women with a
history of incest. International Journal of Group Psychotherapy, 34, 605-616.
Hesselman, S. (1983). Elective mutism in children 1877-1981: A literary review. Acta
Paedopsychiatrica, 49, 297-310.
Heyn, D. (1992). The erotic silence of the American wife. New York: Turtle Bay (Ran-
dom House).
Hill, C. E. (1974). A comparison of the perceptions of a therapy session by clients,
therapists, and objective judges. Journal Supplements Abstract Service, 4(564}.
Hill, C. E. (1982). Counseling process research: Philosophical and methodological
dilemmas. Counseling Psychologist, 10, 7-19.
Hill, C. E. (1992). Research on therapist techniques in brief individual therapy:
Implications for practitioners. Counseling Psychologist, 20, 689-711.
Hill, C. E., & Corbett, M. M. (1993). A perspective on the history of process and
outcome research in counseling psychology. Journal of Counseling Psychology,
40,3-24.
intentions illustrated in a case study and with therapists of varying theoretical
orientations. Journal of Counseling Psychology, 32, 3-22.
Hill, C. E., Helms, J. E., Spiegel, S. B., & Tichenor, V. (1988a). Development of a
system for categorizing client reactions to therapist interventions. Journal of
Counseling Psychology, 35, 27-36.
Hill, C. E., Helms, J. E., Tichenor, v., Spiegel, S. B., O'Grady, K. E., & Perry, E. S.
(1988b). The effects of therapist response modes in brief psychotherapy. Journal
of Counseling Psychology, 35, 222-233.
Hill, C. E., Thompson, B. J., & Corbett, M. M. (1992). The impact of therapist ability
to perceive displayed and hidden client reactions on immediate outcome in
first sessions of brief therapy. Psychotherapy Research, 2, 148-160.
Hill, C. E., Thompson, B. J., Cogar, M. c., & Denman, D. W. (1993). Beneath the
surface of long-term therapy: Therapist and client report of their own and each
other's covert processes. Journal of Counseling Psychology, 40, 278-287.
Hill, C. E., Gelso, C. J., & Mohr, J. J. (2000). Client concealment and self-presentation
in therapy: Comment on Kelly (2000). Psychological Bulletin, 127, 495-500.
Hoffman, 1. Z. (1983). The patient as interpreter of the analyst's experience. Contem-
porary Psychoanalysis, 19, 389-422.
Hokanson, J. E., & Burgess, M. (1962). The effects of three types of aggression on
vascular processes. Journal of Abnormal and Social Psychology, 64, 446-449.
REFERENCES 231

Hokanson, J. E., & Shetler, S. (1961). The effect of overt aggression on physiological
arousal level. Journal of Abnormal and Social Psychology, 63, 446-448.
Holahan, C. J., Moos, R H., Holahan, C. K, & Brennan, P. L. (1997). Social context,
coping strategies, and depressive symptoms: An expanded model with car-
diac patients. Journal of Personality and Social Psychology, 72, 918-928.
Hoorwitz, A (1983, November). Guidelines for treating father-daughter incest.
Social Casework: The Journal of Contemporary Social Work, 515-524.
Horowitz, M. J. (1975). Intrusive and repetitive thoughts after experimental stress:
A summary. Archives of General Psychiatry, 32, 1457-1463.
Horowitz, M. J. (1986). Stress response syndromes (2nd ed.). Northvale, NJ: Jason-
Aronson.
Horowitz, M. J., & Wilner, N. (1976). Stress films, emotion, and cognitive response.
Archives of General Psychiatry, 33, 1339-1344.
Horvath, A 0., & Symonds, B. D. (1991). Relation between working alliance and
outcome in psychotherapy: A meta-analysis. Journal of Counseling Psychology,
38,139-149.
Hough, G. (1992). When confidentiality mandates a secret be kept: A case report.
International Journal of Group Psychotherapy, 42, 105-115.
House, J. S. (1981). Work, stress, and social support. Reading, MA: Addison-Wesley.
Howard, K I., Orlinsky, D. E., & Hill, J. A (1970). Patients' satisfactions in psycho-
therapy as a function of patient-therapist pairing. Psychotherapy: Theory, Re-
search, and Practice, 7,130-134.
Hoyt, M. F. (1978). Secrets in psychotherapy: Theoretical and practical considera-
tions. International Review of Psycho-Analysis, 5, 231-241.
Hymer, S. (1982). The therapeutic nature of confessions. Journal of Contemporary
Psychotherapy, 13, U9-143.
Ichiyama, M. A, Colbert, D., Laramore, H., Heim, M., Carone, K, & Schmidt, J.
(1993). Self-concealment and correlates of adjustment in college students.
Journal of College Student Psychotherapy, 7, 55-68.
Imber-Black, E. (1993). Secrets in families and family therapy: An overview. In E.
Imber-Black (Ed.), Secrets in families and family therapy (pp. 3-28). New York:
W. W. Norton.
Jackins, H. (1962). Elementary counselor's manual. Seattle, WA: Rational Island.
James, W. (1890). The principles of psychology. New York: Henry Holt.
Jammer, L. D., & Leigh, H. (1999). Repressive/defensive coping, endogenous
opioids and health: How a life so perfect can make you sick. Psychiatry
Research, 85, 17-31.
Jang, K L., McCrae, R R, Angleitner, A, Riemann, R, & Livesley, W. J. (1998).
Heritability of facet-level traits in a cross-cultural twin sample: Support for a
hierarchical model of personality. Journal of Personality and Social Psychology,
74,1556-1565.
Janoff-Bulman, R (1972). Shattered assumptions: Towards a new psychology of trauma.
New York: The Free Press.
Janov, A (1970). The primal scream. New York: Dell.
Jay, K, & Young, A (1979). The gay report. New York: Summit Books.
232 REFERENCES

Jensen, M. R. (1987). Psychobiological factors predicting the course of breast cancer.

Journal of Personality, 58, 317-342.
Jones, E. E. (1990). Interpersonal perception. New York: W. H. Freeman.
Jones, E. E., & Archer, R. L. (1976). Are there special effects of personalistic self-
disclosure? Journal of Experimental Social Psychology, 12, 180-193.
Jones, E. E., & Gordon, E. M. (1972). Timing of self-disclosure and its effects on
personal attraction. Journal of Personality and Social Psychology, 24, 358-365.
Jones, E. E., & Pittman, T. S. (1982). Toward a general theory of strategic self-
presentation. In J. Suls (Ed.), Psychological perspectives on the self (Vol. 1, pp.
231-262). Hillsdale, NJ: Erlbaum.
Jones, E. E., Rhodewalt, F., Berglas, S., & Skelton, J. A. (1981). Effects of strategic self
presentation on subsequent self-esteem. Journal of Personality and Social Psy-
chology, 41, 407-421.
Jourard, S. M. (1963). Personal adjustment: An approach through the study of healthy
personality (2nd ed.). New York: Macmillan.
Jourard, S. M. (1971a). Self-disclosure: An experimental analysis of the transparent self.
New York: Wiley.
Jourard, S. M. (1971b). The transparent self. New York: Van Nostrand Reinhold.
Jung, c. G. (1933). Modern man in search of a soul. New York: Harvest Books.
Justice, B., & Justice, R. (1979). The broken taboo: Sex in the family. New York: Human
Sciences Press.
Kagan, J. (1994). Galen's prophecy: Temperament in human nature. New York: Basic
Books.
Kagan, J., Reznick, J. S., & Snidman, N. (1988, April 8). Biological bases of childhood
shyness. Science, 240, 167-171.
Kagan, J., Reznick, J. S., & Gibbons, J. (1989). Inhibited and uninhibited types of
children. Child Development, 60, 838-845.
Kahn, J. H., & Kelly, A. E. (1998). Dimensions of self-concealment. Unpublished
manuscript. Illinois State University.
Kahn, J. H., & Hessling, R. M. (2001). Measuring the tendency to conceal versus dis-
close psychological distress. Journal of Social and Clinical Psychology, 20, 41-65.
Kahn, J. H., Achter, J. A., & Shambaugh, E. J. (2001). Client distress disclosure,
characteristics at intake, and outcome in brief counseling. Journal of Counseling
Psychology, 48, 203-211.
Karpel, M. A. (1980). Family secrets. Family Process, 19, 295-306.
Kaslow, F. (1993). Attractions and affairs: Fabulous and fatal. Journal of Family
Psychotherapy, 4, 1-34.
Kaufman, I., Peck, A., & Tagiuri, C. (1954). The family constellation and overt
incestuous relations between father and daughter. American Journal of Ortho-
psychiatry, 24, 266-279.
Kaufman, Y. (1989). Analytical psychotherapy. In R. J. Corsini & D. Wedding (Eds.),
Current psychotherapies (pp. 119-154). Itasca, IL: F. E. Peacock.
Kazdin, A. E. (1986). Comparative outcome studies of psychotherapy: Meth-
odological issues and strategies. Journal of Consulting and Clinical Psychology,
54,95-105.
REFERENCES 233

Kelly, A. E. (1997). Views of gaining catharsis versus new insights for coping with
secrets. Unpublished raw data. South Bend, IN: University of Notre Dame.
Kelly, A E. (1998). Clients' secret keeping in outpatient therapy. Journal of Counseling
Psychology, 45, 50-57.
Kelly, A. E. (1999). Revealing personal secrets. Current Directions in Psychological
Science, 8, 105-109.
Kelly, A. E. (2000a). Helping construct desirable identities: A self-presentational
view of psychotherapy. Psychological Bulletin, 126, 475-494.
Kelly, A. E. (2000b). A self-presentational view of psychotherapy: Reply to Hill,
Gelso, and Mohr (2000) and to Arkin and Hermann (2000). Psychological Bulle-
tin, 126, 505-511.
Kelly, A. E., & Achter, J. A. (1995). Self-concealment and attitudes toward counsel-
ing in university students. Journal of Counseling Psychology, 42, 40-46.
Kelly, A. E., & Kahn, J. H. (1994). Effects of suppression of personal intrusive
thoughts. Journal of Personality and Social Psychology, 66,998-1006.
Kelly, A. E., & McKillop, K. J. (1996). Consequences of revealing personal secrets.
Psychological Bulletin, 120, 450-465.
Kelly, A. E., McKillop, K. J., & Neimeyer, G. J. (1991). Effects of counselor as audi-
ence on internalization of depressed and nondepressed self-presentations.
Journal of Counseling Psychology, 38, 126-132.
Kelly, A E., Coenen, M. E., & Johnston, B. L. (1995). Confidants' feedback and
traumatic life events. Journal of Traumatic Stress, 8, 161-169.
Kelly, A. E., Kahn, J. H., & Coulter, R. G. (1996). Client self-presentations at intake.
Journal of Counseling Psychology, 43, 300-309.
Kelly, A. E., Klusas, J. A., von Weiss, R. T., & Kenny, C. (2001). What is it about
revealing secrets that is beneficial? Personality and Social Psychology Bulletin, 27,
651-665.
Kelly, G. A. (1955). The psychology of personal constructs (Vols. 1 and 2). New York:
Norton.
Kennedy, S., Kiecolt-Glaser, J. K., & Glaser, R. (1990). Social support, stress, and the
immune system. In I. G. Sarason, B. R. Sarason, & G. R. Pierce (Eds.), Social
support: An interactional view (pp. 253-266). New York: Wiley.
Kessler, R. c., & McLeod, J. D. (1985). Social support and mental health in commu-
nity samples. In S. Cohen & S. L. Syme (Eds.), Social support and health (pp. 219-
240). Orlando, FL: Academic Press.
Kessler, R. c., Price, R. H., & Wortman, C. B. (1985). Social factors in psychopathol-
ogy: Stress, social support, and coping processes. Annual Review of Psychology,
36, 531-572.
Kiesler, D. J. (1981). Interpersonal theory for personality and psychotherapy. In J. C.
Anchin & D. J. Kiesler (Eds.), Handbook of interpersonal psychotherapy (pp. 3-24).
New York: Pergamon.
King, L. A., & Emmons, R. A. (1990). Conflict over emotional expression: Psycho-
logical and physical correlates. Journal of Personality and Social Psychology, 58,
864-877.
King, L. A., & Emmons, R. A. (1991). Psychological, physical, and interpersonal
234 REFERENCES

correlates of emotional expressiveness, conflict, and control. European Journal

of Personality,S, 131-150.
King, L. A, Emmons, R A, & Woodley, S. (1992). The structure of inhibition. Journal
of Research in Personality, 26, 85-102.
Kirman, W. J. (1991). Short-term modern analytic group therapy. Modern Psycho-
analysis, 16, 151-159.
Kissen, D. M. (1966). The significance of personality in lung cancer among men.
Annals of the New York Academy of Science, 125, 820-826.
Klein, J. W. (1976). Ethnotherapy with Jews. International Journal of Mental Health,S,
26-38.
Klein, M. H., Mathieu, P. L., Gendlin, E. T., & Kiesler, D. J. (1969). The experiencing
scale: Vol. 1. A research and training manual. Madison: Wisconsin Psychiatric
Institute.
Klein, M. H., Mathieu-Coughlan, P., & Kiesler, D. J. (1986). The experiencing scales.
In L. S. Greenberg & W. M. Pinsof (Eds.), The psychotherapeutic process: A
research handbook (pp. 21-71). New York: Guilford Press.
Kleinke, C. L., & Kahn, M. L. (1980). Perceptions of self disclosers: Effects of sex and
physical attractiveness. Journal of Personality, 48, 190-205.
Kliewer, W. L., Lepore, S. J., Oskin, D., & Johnson, P. D. (1998). The role of social and
cognitive processes in children's adjustment to community violence. Journal of
Consulting and Clinical Psychology, 66,199-209.
Kneier, A W., & Temoshok, L. (1984). Repressive coping reactions in patients with
malignant melanoma as compared to cardiovascular disease patients. Journal
of Psychosomatic Research, 28, 145-155.
Kobocow, B., McGuire, J. M., & Blau, B. J. (1983). The influence of confidentiality
conditions on self-disclosure of early adolescence. Professional Psychology, 4,
435-443.
Koutstaal, w., Schacter, D. L., Johnson, M. K., Angell, K. E., & Gross, M. S. (1998).
Post-event review in older and younger adults: Improving memory acces-
sibility of complex everyday events. Psychology and Aging, 13, 277-296.
Kozak, M. J., Foa, E. B., & Steketee, G. (1988). Process and outcome of exposure
treatment with obsessive-compulsives: Psychophysiological indicators of
emotional processing. Behavior Therapy, 19, 157-169.
Kremer, T. G., & Gesten, E. L. (1998). Confidentiality limits of managed care and
clients' willingness to self-disclose. Professional Psychology: Research and Prac-
tice, 29, 553-558.
Krestan, J., & Bepko, C. (1993). On lies, secrets, and silence: The multiple levels of
denial in addictive families. In E. Imber-Black (Ed.), Secrets in families and family
therapy (pp. 141-159). New York: W. W. Norton.
Kulik, J. A, Sledge, P., & Mahler, H. 1. (1986). Self-confirmatory attribution, egocen-
trism, and the perpetuation of self-beliefs. Journal of Personality and Social
Psychology, 50, 587-594.
Kurtz, R R, & Grummon, D. L. (1972). Different approaches to the measurement of
therapist empathy and their relations to therapy outcome. Journal of Consulting
and Clinical Psychology, 39, 106-115.
Kyrios, M. (1998). A cognitive-behavioral approach to the understanding and
REFERENCES 235

management of obsessive-compulsive personality disorder. In C. Perris & P.

D. McGorry (Eds.), Cognitive psychotherapy of psychotic and personality disorders:
Handbook of theory and practice (pp. 351-378). Chichester, England: American
Ethnological Press.
Labott, S. M., Ahleman, S., Wolever, M. E., & Martin, R B. (1990). The physiological
and psychological effects of the expression and inhibition of emotion. Behav-
ioral Medicine, 16, 182-189.
Lamb, S. (1991). Acts without agents: An analysis of linguistic avoidance in journal
articles on men who batter women. American Journal of Orthopsychiatry, 61,
250-257.
Lambert, M. J., & Bergin, A E. (1994). The effectiveness of psychotherapy. In A E.
Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th
ed., pp. 143-189). New York: Wiley.
Lambert, M. J., & Hill, c. E. (1994). Assessing psychotherapy outcomes and pro-
cesses. In A E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and
behavior change (4th ed., pp. 72-113). New York: Wiley.
Lambert, M. J., Dejulio, S. 5., & Stein, D. M. (1978). Therapist interpersonal skills:
Process, outcome, methodological considerations, and recommendations for
future reference. Psychological Bulletin, 85,467-489.
Lambert, M. J., Shapiro, D. A, & Bergin, A E. (1986). The effectiveness of psycho-
therapy. In S. L. Garfield & A E. Bergin (Eds.), Handbook of psychotherapy and
behavior change (3rd ed., pp. 157-211). New York: Wiley.
Landman, J. T., & Dawes, R M. (1982). Smith and Glass' conclusions stand up
under scrutiny. American Psychologist, 37, 504-516.
Lane, J. D., & Wegner, D. M. (1995). The cognitive consequences of secrecy. Journal
of Personality and Social Psychology, 69, 237-253.
Lanzetta, J. T., Cartwright-Smith, J., & Kleck, R E. (1976). Effects of nonverbal
dissimulation on emotional experience and autonomic arousal. Journal of Per-
sonality and Social Psychology, 33, 354-370.
Larson, D. G. (1993a). The helper's journey: Working with people facing grief, loss, and
life-threatening illness. Champaign, IL: Research Press.
Larson, D. G. (1993b). Self-concealment: Implications for stress and empathy in
oncology care. Journal of Psychosocial Oncology, 11, 1-16.
Larson, D. G., & Chastain, R L. (1990). Self-concealment: Conceptualization, mea-
surement, and health implications. Journal of Social and Clinical Psychology, 9,
439-455.
Lazarus, R S. (1985). The trivialization of distress. In J. c. Rose & L. J. Solomon
(Eds.), Primary prevention of psychopathology: Vol. 8. Prevention in Health Psychol-
ogy (pp. 279-298). Hanover, NH: University Press of New England.
Lazarus, R 5., & Folkman, S. (1984). Stress, appraisal, and coping. New York:
Springer.
Lazarus, R S., Speisman, J. c., Mordkof, AM., & Davison, L. A (1962). A labora-
tory study of psychological stress produced by a motion picture film. Psycho-
logical Monographs, 76(34, entire No. 553).
Leary, M. R, & Kowalski, R M. (1990). Impression management: A literature
review and two-component model. Psychological Bulletin, 107, 34-47.
236 REFERENCES

Lehman, D. R, & Hemphill, K. J. (1990). Recipients' perceptions of support at-

tempts and attributions for support attempts that fail. Journal of Social and
Personal Relationships, 7, 563-574.
Lehman, D. R, Ellard, J. H., & Wortman, C. B. (1986). Social support for the
bereaved: Recipients' and providers' perspectives on what is helpful. Journal
of Consulting and Clinical Psychology, 54, 438-446.
Lehman, D. R, Wortman, C. B., & Williams, A. F. (1987). Long-term effects of losing
a spouse or child in a motor vehicle crash. Journal of Personality and Social
Psychology, 52, 218-23l.
Lepore, S. J. (1992). Social conflict, social support, and psychological distress:
Evidence of cross-domain buffering effects. Journal of Personality and Social
Psychology, 63, 857-867.
Lepore, S. J. (1997a, April). Social constraints, intrusive thoughts, and negative affect in
women with cancer. Paper presented at the annual meeting of the Society of
Behavioral Medicine, San Francisco, CA.
Lepore, S J. (1997b). Expressive writing moderates the relation between intrusive
thoughts and depressive symptoms. Journal of Personality and Social Psychology,
73,1030-1037.
Lepore, S. J., & Helgeson, V. S. (1998). Social constraints, intrusive thoughts, and
mental health after prostate cancer. Journal of Social and Clinical Psychology, 17,
89-106.
Lepore, S. J., Silver, R c., Wortman, C. B., & Wayment, H. A. (1996). Social con-
straints, intrusive thoughts, and depressive symptoms among bereaved
mothers. Journal of Personality and Social Psychology, 70,271-282.
Lepore, S. J., Ragan, J. D., & Jones, S. (2000). Talking facilitates cognitive-emotional
processes of adaptation to an acute stressor. Journal of Personality and Social
Psychology, 78, 499-508.
Lerner, M. J. (1980). The belief in a just world: A fundamental delusion. New York:
Plenum Press.
Levy, R L. (1983). Social support and compliance: A selective review and critique of
treatment integrity and outcome measurement. Social Science and Medicine, 17,
1329-1338.
Lieberman, M., Yalom, I., & Miles, M. (1973). Encounter groups: First facts. New York:
Basic Books.
Lifton, R J. (1986). The Nazi doctors: Medical killing and the psychology ofgenocide. New
York: Basic Books.
Lindberg, F. H., & Distad, L. J. (1985). Survival responses to incest: Adolescents in
crisis. Child Abuse and Neglect, 9, 521-526.
Liotti, G. (1987). Structural cognitive therapy. In W. Dryden & W. L. Golden (Eds.),
Cognitive-behavioral approaches to psychotherapy (pp. 92-128). Cambridge, UK:
Hemisphere.
Lodge, D. (1995). Therapy: A novel. New York: Viking.
Loehlin, J. C. (1992). Genes and environment in personality development. Thousand
Oaks, CA: Sage.
Lubell, D., & Soong, W. (1982). Group therapy with sexually abused adolescents.
Canadian Journal of Psychiatry, 27, 311-315.
REFERENCES 237

Luck, H. E., & Timaeus, E. (1969). Scales for the measurement of manifest anxi-
ety (MAS) and social desirability (SDS-E and SOS-CM). Diagnostica, 15,
134-141.
Luminet, 0., Bouts, P., Delie, E, Manstead, A. S. R, & Rime, B. (2000). Social sharing
of emotion following exposure to a negatively valenced situation. Cognition
and Emotion, 14,661-688.
Lutgendorf, S. K., Antoni, M. H., Kumar, M., & Schneiderman, N. (1994). Changes
in cognitive coping strategies predict EBV-antibody titre change following a
stressor disclosure induction. Journal of Psychosomatic Research, 38, 63-78.
Lyubomirsky, S., & Nolen-Hoeksema, S. (1993). Self-perpetuating properties of
dysphoric rumination. Journal of Personality and Social Psychology, 65, 339-349.
Lyubomirsky, S., & Nolen-Hoeksema, S. (1995). Effects of self-focused rumination
on negative thinking and interpersonal problem solving. Journal of Personality
and Social Psychology, 69, 176-190.
Macdonald, J., & Morley, I. (2001). Shame and non-disclosure: A study of the
emotional isolation of people referred for psychotherapy. British Journal of
Medical Psychology, 74, 1-21.
Maddison, D. C, & Walker, W. L. (1967). Factors affecting the outcome of conjugal
bereavement. British Journal of Psychiatry, 113, 1057-1067.
Mahony, P. J. (1996). Freud's Dora: A psychoanalytic, historical, and textual study. New
Haven, CT: Yale University Press.
Major, B., & Gramzow, R H. (1999). Abortion as stigma: Cognitive and emotional
implications of concealment. Journal of Personality and Social Psychology, 77,
735-745.
Major, B., Cozzarelli, C, Sciacchittano, A. M., Cooper, M. L., Testa, M., & Mueller, P.
M. (1990). Perceived social support, self-efficacy, and adjustment to abortion.
Journal of Personality and Social Psychology, 59, 452-463.
Major, B., Zubek, J. M., Cooper, M. L., Cozzarelli, C, & Richards, C (1997). Mixed
messages: Implications of social conflict and social support within close rela-
tionships for adjustment to a stressful life event. Journal of Personality and Social
Psychology, 72, 1349-1363.
Manne, S. L., Taylor, K. L., Dougherty, J., & Kemeny, N. (1997). Supportive and
negative responses in the partner relationship: Their association with psycho-
logical adjustment among individuals with cancer. Journal of Behavioral Medi-
cine, 20, 101-125.
Maracek, J., & Mettee, D. R (1972). Avoidance of continued success as a function of
self-esteem, level of esteem certainty, and responsibility for success. Journal of
Personality and Social Psychology, 22, 90-107.
Margolis, G. J. (1966). Secrecy and identity. International Journal of Psycho-Analysis,
47,517-522.
Margolis, G. J. (1974). The psychology of keeping secrets. International Review of
Psycho-Analysis, 1, 291-296.
Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954-969.
Martin, J. (1984). The cognitive mediational paradigm for research on counseling.
Journal of Counseling Psychology, 31, 558-571.
Martin, J., Martin, w., Meyer, M., & SIemon, A. (1986). Empirical investigation of
238 REFERENCES

the cognitive mediational paradigm for research on counseling. Journal of

Counseling Psychology, 33, 115-123.
Martin, J., Martin, w., & SIemon, A G. (1987). Cognitive mediation in person-
centered and rational-emotive therapy. Journal of Counseling Psychology, 34,
251-260.
Martin, L. L., Tesser, A, & McIntosh, W. D. (1993). Wanting but not having: The
effects of unattained goals on thoughts and feelings. In D. M. Wegner & J. w.
Pennebaker (Eds.), Handbook of mental control (pp. 552-572). Englewood Cliffs,
NJ: Prentice-Hall.
McArdle, J. (1974). Impression management by alcoholics. Quarterly Journal of the
Study of Alcoholism, 35, 911-916.
McCanne, T. R, & Anderson, J. A (1987). Emotional responding following experi-
mental manipulation of facial electromyographic activity. Journal of Personality
and Social Psychology, 52, 759-768.
McCarthy, B. W. (1995). Learning from unsuccessful sex therapy patients. Journal of
Sex and Marital Therapy, 21, 31-38.
McCrae, R R, & Costa, P. T., Jr. (1995). Trait explanations in personality psychology.
European Journal of Personality, 9, 231-252.
McDaniel, S. H., Stiles, W. B., & McGaughey, K. J. (1981). Correlations of male
college students' verbal response mode use in psychotherapy with measures
of psychological disturbance and psychotherapy outcome. Journal of Consult-
ing and Clinical Psychology, 49, 571-582.
McFarlin, D. B., & Blascovich, J. (1981). Effects of self-esteem and performance
feedback on future affective preferences and cognitive expectations. Journal of
Personality and Social Psychology, 22, 90-107.
McGaughey, K. J., & Stiles, W. B. (1983). Courtroom interrogation of rape victims:
Verbal response mode use by attorneys and witnesses during direct examina-
tion vs. cross-examination. Journal of Applied Social Psychology, 13,78-87.
McKillop, K. J. (1991). Multiple self-presentations and the resiliency of the self.
Dissertation Abstracts International, 52(5-B), 2822.
McKillop, K. J., & Schlenker, B. R (1988). Audience effects on the internalization of
depressed vs. nondepressed self-presentations. Paper presented at the Annual
Meeting of the American Psychological Association, Atlanta, Georgia.
McKillop, K. J., Berzonsky, M. D., & Schlenker, B. R (1992). The impact of self-
presentations on self-beliefs: Effects of social identity and self-presentational
context. Journal of Personality, 60, 789-808.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Meichenbaum, D. (1977). Cognitive-behavior modification: An integrative approach.
New York: Plenum Press.
Meiselman, K. C. (1978). Incest. San Francisco: Jossey-Bass.
Mendola, M. (1980). The Mendola Report: A new look at gay couples. New York: Crown.
Mendola, M., & Kleck, R E. (1993). Effects of talking about a stressful event on
arousal: Does what we talk about make a difference? Journal of Personality and
Social Psychology, 64, 283-292.
Merton, R K. (1948). The self fulfilling prophecy. Antioch Review, 8,193-210.
Miell, D. E., & Duck, S. W. (1986). Strategies in developing friendship. In V. J.
REFERENCES 239

Derlega & B. A Winstead (Eds.), Friendship and social interaction (pp.129-144).

New York: Springer-Verlag.
Milgram, S. (1967). The small world problem. Psychology Today, 1, 60-67.
Miller, R, Goldman, E., & Bor, R (1994). Application of a family systems approach
to working with people affected by HIV disease-Two case studies. Journal of
Family Therapy, 16, 295-312.
Minoff, L. A (1992). A case of rape: Real and imagined. Psychoanalytic Review, 79,
537-553.
Mischel, w., & Peake, P. K. (1982). Beyond deja vu in the search for cross-situational
consistency. Psychological Review, 89, 730-755.
Mitchell, R G. (1993). Secrecy and fieldwork. Thousand Oaks, CA: Sage.
Mook, D. G. (1983). In defense of external invalidity. American Psychologist, 38,
379-387.
Moos, R H., & Solomon, G. F. (1965). Psychologic comparisons between women
with rheumatoid arthritis and their non-arthritic sisters: IT. Content analysis of
interviews. Psychosomatic Medicine, 27, 150-164.
Moreno, J. (1958). Psychodrama (Vol. IT). New York: Beacon House.
Moultrop, D. {l992). Husbands, wives, and lovers. New York: Guilford.
Mueller, D. P. (1980). Social networks: A promising direction for research on the
relationship of the social environment to psychiatric disorder. Social Science
Medicine, 14, 147-161.
Murphy, B. (1989). Lesbian couples and their parents: The effects of perceived
parental attitudes on the couples. Journal of Counseling and Development, 68,
46-5l.
Murray, E. J., & Segal, D. L. (1994). Emotional processing in vocal and written
expression of feelings about traumatic experiences. Journal of Traumatic Stress,
7,391-405.
Murray, E. J., Lamnin, A D., & Carver, C. S. (1989). Emotional expression in written
essays and psychotherapy. Journal of Social and Clinical Psychology, 8,414-429.
Myers, L. B., & Brewin, C. R (1994). Recall of early experience and the repressive
coping style. Journal of Abnormal Psychology, 103, 288-292.
Nash, E. H., Hoehn-Saric, R, Battle, C. c., Stone, A R, Imber, S. D., & Frank, J. D.
(1965). Systematic preparation of patients for short-term psychotherapy. IT.
Relations to characteristics of patient, therapist, and the psychotherapeutic
process. Journal of Nervous and Mental Disease, 140, 374-383.
Nichols, M. (1974). Outcome of brief cathartic psychotherapy. Journal of Consulting
and Clinical Psychology, 42, 403-410.
Nichols, M., & Efran, J. (1985). Catharsis in psychotherapy: A new perspective.
Psychotherapy, 22, 46-58.
Nisbett, R E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports
and mental processes. Psychological Review, 84, 231-259.
Nolen-Hoeksema, S., & Morrow, J. (1993). Effects of rumination and distraction on
naturally occurring depressed mood. Cognition and Emotion, 7, 561-570.
Nolen-Hoeksema, S., McBride, A, & Larson, J. (1997). Rumination and psychologi-
cal distress among bereaved partners. Journal of Personality and Social Psychol-
ogy, 72, 855-862.
240 REFERENCES

Norton, R, Feldman, c., & Tafoya, D. (1974). Risk parameters across types of
secrets. Journal of Counseling Psychology, 21, 450-454.
Nowak, A., Szamrej, J., & Latane, B. (1990). From private attitude to public opinion:
A dynamic theory of social impact. Psychological Review, 97, 362-376.
O'Brien, R A., & Pilar, B. R (1997). Application of solution-focused interventions to
nurse home visitation for pregnant women and parents of young children.
Journal of Community Psychology, 25, 47-57.
Okun, B. F. (1997). Effective helping: Interviewing and counseling techniques (5th ed.).
Pacific Grove, CA: Brooks/Cole.
Olson, J. M., Barefoot, J. c., & Strickland, L. H. (1976). What the shadow knows:
Person perception in a surveillance situation. Journal of Personality and Social
Psychology, 34, 583-589.
Orlinsky, D. E., & Howard, K. I. (1966). Therapy Session Report, Form P. Chicago:
Institute for Juvenile Research.
Palazzoli, M. S., & Prata, G. (1982, October). Snares in family therapy. Journal of
Marital and Family Therapy, 443-450.
Paul, N. L., & Bloom, J. D. (1970). Multiple-family therapy: Secrets and scapegoat-
ing in family crisis. International Journal of Group Psychotherapy, 20(1), 37-47.
Pennebaker, J. W. (1985). Traumatic experience and psychosomatic disease: Explor-
ing the roles of behavioral inhibition, obsession, and confiding. Canadian
Psychology, 26, 82-95.
Pennebaker, J. W. (1989). Confession, inhibition, and disease. Advances in Experi-
mental Social Psychology, 22, 211-244.
Pennebaker, J. W. (1990). Opening up: The healing powers of confiding in others. New
York: Morrow.
Pennebaker, J. W. (1993). Mechanisms of social constraint. In D. Wegner & J. w.
Pennebaker (Eds.), Handbook of mental control (pp. 200-212). Englewood Cliffs,
NJ: Prentice-Hall.
Pennebaker, J. W. (1997a). Writing about emotional experiences as a therapeutic
process. Psychological Science, 8, 162-166.
Pennebaker, J. W. (1997b). Opening up: The healing power of expressing emotion. New
York: Guilford.
Pennebaker, J. w., & Beall, S. K. (1986). Confronting a traumatic event: Toward an
understanding of inhibition and disease. Journal of Abnormal Psychology, 95,
274-281.
Pennebaker, J. w., & Chew, C. H. (1985). Behavioral inhibition and electrodermal
activity during deception. Journal of Personality and Social Psychology, 49, 1427-
1433.
Pennebaker, J. w., & Hoover, C. W. (1985). Inhibition and cognition: Toward an
understanding of trauma and disease. In R J. Davidson, G. E. Schwartz, & D.
Shapiro (Eds.), Consciousness and self-regulation (Vol. 4, pp.107-136). New York:
Plenum Press.
Pennebaker, J. w., & O'Heeron, R C. (1984). Confiding in others and illness rates
among spouses of suicide and accidental-death victims. Journal of Abnormal
Psychology, 93, 473-476.
Pennebaker, J. W., & Harber, K. D. (1993). A social stage model of collective coping:
REFERENCES 241

The Lorna Prieta earthquake and the Persian Gulf War. Journal of Social Issues,
49,125-145.
Pennebaker, J. w., & Susman, J. R. (1988). Disclosure of traumas and psychosomatic
processes. Social Science and Medicine, 26, 327-332.
Pennebaker,1. w., Hughes, C. E, & O'Heeron, R. C. (1987). The psychopathology of
confession: Linking inhibitory and psychosomatic processes. Journal of Person-
ality and Social Psychology, 52, 781-793.
Pennebaker, J. W., Kiecolt-Glaser, J. K, & Glaser, R. (1988). Disclosure of traumas
and immune function: Health implications for psychotherapy. Journal of Con-
sulting and Clinical Psychology, 56, 239-245.
Pennebaker, J. w., Barger, S. D., & Tiebout, J. (1989). Disclosure of traumas and
health among holocaust survivors. Psychosomatic Medicine, 51, 577-589.
Pennebaker, J. w., Colder, M., & Sharp, L. K (1990). Accelerating the coping
process. Journal of Personality and Social Psychology, 58, 528-537.
Pennebaker, J. w., Mayne, T. 1., & Francis, M. E. (1997). Linguistic predictors of
adaptive bereavement. Journal of Personality and Social Psychology, 72, 863-871.
Peskin, J. (1992). Ruse and representations: On children's ability to conceal informa-
tion. Developmental Psychology, 28, 84-89.
Peters-Golden, H. (1982). Breast cancer: Varied perceptions of social support in the
illness experience. Social Science and Medicine, 16, 483-491.
Petrie, K J., Booth, R. J., Pennebaker, J. w., Davison, K P., & Thomas, M. G. (1995).
Disclosure of trauma and immune response to a hepatitis B vaccination pro-
gram. Journal of Consulting and Clinical Psychology, 63, 787-792.
Petronio, S. (2000). Balancing the secrets of private disclosures. Mahwah, NJ: Erlbaum.
Petronio, S., & Bantz, C. (1991). Controlling the ramifications of disclosure: "Don't
tell anybody but ... " Journal of Language and Social Psychology, 10, 263-269.
Pincus, L., & Dare, C. (1978). Secrets in the family. New York: Pantheon Books.
Pittman, E (1989). Private lies. New York: Norton.
Polivy, J. (1998). The effects of behavioral inhibition: Integrating internal cues,
cognition, behavior, and affect. Psychological Inquiry, 9, 181-204.
Ponse, B. (1978). Identities in the lesbian world. Westport, CT: Greenwood.
Premo, B. E., & Stiles, W. B. (1983). Familiarity in verbal interactions of married
couples versus strangers. Journal of Social and Clinical Psychology, 1, 209-230.
Quintana, S. M., & Meara, N. M. (1990). Internalization of therapeutic relationships
in short-term psychotherapy. Journal of Counseling Psychology, 37, 123-130.
Rachman, S. (1980). Emotional processing. Behaviour Research and Therapy, 18, 51-60.
Rando, T. (1993). Treatment of complicated mourning. Champaign, IL: Research Press.
Regan, A. M., & Hill, C. E. (1992). Investigation of what clients and counselors do
not say in brief therapy. Journal of Counseling Psychology, 39, 240-246.
Regan, J. W. (1968). Guilt, inequity, and altruistic behavior. Unpublished doctoral
dissertation, as cited in Freedman, J. L., Carlsmith, J. M., & Sears, D. O. (1970).
Social psychology. Englewood Cliffs, NJ: Prentice Hall.
Reichert, E. (1994). Expressive group therapy with adult survivors of sexual abuse.
Family Therapy, 21, 99-105.
Reik, T. (1945). The compulsion to confess: On the psychoanalysis of crime and punish-
ment. New York: Grove Press.
242 REFERENCES

Rennie, D. L. (1985, June). Client deference in the psychotherapy relationship. Paper

presented at the annual meeting of the Society for Psychotherapy Research,
Evanston, 11.
Rennie, D. L. (1992). Qualitative analysis of client's experience of psychotherapy:
The unfolding of reflexivity. In S. G. Toukmanian & D. L. Rennie (Eds.),
Psychotherapy process research: Paradigmatic and narrative approaches. Sage focus
editions (Vol. 143, pp. 211-233). Newbury Park, CA: Sage.
Rhodes, R. H., Hill, C. E., Thompson, B. J., & Elliott, R. (1994). Client retrospective
recall of resolved and unresolved misunderstanding events. Special section:
Qualitative research in counseling process and outcome. Journal of Counseling
Psychology, 41, 473-483.
Rhodewalt, F., & Agustsdottir, S. (1986). Effects of self-presentation on the phenom-
enal self. Journal of Personality and Social Psychology, 50, 47-55.
Richardson, L. (1988). Secrecy and status: The social construction of forbidden
relationships. American Sociological Review, 53, 209-219.
Rime, B. (1995a). Mental rumination, social sharing, and the recovery from emo-
tional exposure. In J. w. Pennebaker (Ed.), Emotion, disclosure, and health (pp.
271-291). Washington, DC: American Psychological Association.
Rime, B. (1995b). The social sharing of emotion as a source for the social knowledge
of emotion. In J. A. Russell, D. Fernandez, & M. Jose (Eds.), Everyday concep-
tions of emotion: An introduction to the psychology, anthropology and linguistics of
emotion. NATO ASI series D: Behavioural and social sciences (Vol. 81, pp. 475-489).
Norwell, MA: Kluwer.
Rime, B., & Schiaratura, L. (1991). Gesture and speech. In R. S. Feldman & B. Rime
(Eds.), Fundamentals of nonverbal behavior. Studies in emotion and social interaction
(pp. 239-281). New York: Cambridge University Press.
Rime, B., Mesquita, B., Philippot, P., & Boca, S. (1991a). Beyond the emotional event:
Six studies on the social sharing of emotion. Cognition and Emotion, 5, 435-465.
Rime, B., Noel, P., & Philippot, P. (1991b). Episode emotionnel, reminiscences
mentales et reminiscences sociales. (Mental rumination, social sharing, and
everyday life emotions.) Cahiers Internationaux de Psychologie Sociale, 7(11),
93-104.
Rime, B., Philippot, P., Boca, S., & Mesquita, B. (1992). Long-lasting cognitive and
social consequences of emotion: Social sharing and rumination. In W. Stroebe
& M. Hewstone (Eds.), European review of social psychology (Vol. 3, pp. 225-258).
New York: Wiley.
Rime, B., Philippot, P., Finkenauer, c., Legast, S., Moorkens, P., & Tornqvist, J.
(1995). Mental rumination and social sharing in current life emotion. Unpublished
manuscript. Louvain-Ia-Neuver, Belgium: University of Louvain.
Rippa, B. (1994). Groups in Israel during the Gulf War. Group Analysis, 27, 87-94.
Roback, H. B., & Shelton, M. (1995). Effects of confidentiality limitations on the
psychotherapeutic process. Journal of Psychotherapy Practice and Research, 4,
185-193.
Robertson, J. M., & Fitzgerald, L. F. (1992). Overcoming the masculine mystique:
Preferences for alternative forms of assistance. Journal of Counseling Psychology,
39, 240-246.
REFERENCES 243

Rodriguez, N., & Ryave, A. L. (1992). The structural organization and micropolitics
of everyday secret telling interactions. Qualitative Sociology, 15, 297-318.
Roger, D., & Najarian, B. (1998). The relationship between emotional rumination
and cortisol secretion under stress. Personality and Individual Differences, 24,
531-538.
Rogers, C. R. (1951). Client-centered therapy: Its current practice, implications, and
theory. Boston: Houghton Mifflin.
Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic person-
ality change. Journal of Counseling Psychology, 21, 95-103.
Rook, K. S. (1984). The negative side of social interaction: Impact on psychological
well-being. Journal of Personality and Social Psychology, 46, 1097-1108.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Rosenfeld, D. (1980). The handling of resistances in adult patients. International
Journal of Psychoanalysis, 61, 71-83.
Ross, L., Greene, D., & House, P. (1977). The false consensus effect: An egocentric
bias in social perception and attribution processes. Journal of Experimental
Social Psychology, 13, 279-301.
Roth, S. (1985). Psychotherapy with lesbian couples: Individual issues, female
socialization, and the social context. Journal of Marital and Family Therapy, 11,
273-286.
Russell, D. E. (1986). The secret trauma: Incest in the lives ofgirls and women. New York:
Basic Books.
Russell, P. D., & Snyder, W. U. (1963). Counselor anxiety in relation to amount of
clinical experience and quality of affect demonstrated by clients. Journal of
Consulting Psychology, 27, 358-363.
Saffer, J. B., Sansone, P., & Gentry, J. (1979). The awesome burden upon the child
who must keep a family secret. Child Psychiatry and Human Development, 10,
35-40.
Sagarin, B. J., Rhoads, K. V. L., & Cialdini, R. B. (1998). Deceiver's distrust: Denigra-
tion as a consequences of undiscovered deception. Personality and Social Psy-
chology Bulletin, 24, 1167-1176.
Salkovskis, P. M., & Campbell, P. (1994). Thought suppression induces intrusion in
naturally occurring negative intrusive thoughts. Behaviour Research and Ther-
apy, 32, 1-8.
Scheff, T. J. (1979). Catharsis in healing, ritual, and drama. Berkeley: University of
California Press.
Schlenker, B. R. (1980). Impression management: The self-concept, social identity, and
interpersonal relations. Monterey, CA: Brooks/Cole.
Schlenker, B. R. (1985). Identity and self-identification. In B. R. Schlenker (Ed.), The
self and social life (pp. 65-99). New York: McGraw-Hill.
Schlenker, B. R. (1986). Self-identification: Toward an integration of the private and
public self. In R. Baumeister (Ed.), Public self and private self (pp. 21-62). New
York: Springer-Verlag.
Schlenker, B. R. (1987). Threats to identity: Self-identification and social stress. In
C. R. Snyder & c. E. Ford (Eds.), Coping with negative life events: Clinical and
social psychological perspectives (pp. 273-321). New York: Plenum Press.
244 REFERENCES

Schlenker, B. R, & Trudeau, J. V. (1990). Impact of self-presentations on private self-

beliefs: Effects of prior self-beliefs and misattribution. Journal of Personality and
Social Psychology, 58, 22-32.
Schlenker, B. R, & Weigold, M. F. (1989). Goals and the self-identification process:
Constructing desired identities. In L. A. Pervin (Ed.), Goal concepts in person-
ality and social psychology (pp. 243-290). Hillsdale, NJ: Erlbaum.
Schlenker, B. R, & Weigold, M. F. (1990). Self-consciousness and self-presentation:
Being autonomous versus appearing autonomous. Journal of Personality and
Social Psychology, 59, 820-828.
Schlenker, B. R, & Weigold, M. F. (1992). Interpersonal processes involving impres-
sion regulation and management. Annual Review of Psychology, 43, 133-168.
Schlenker, B. R, Dlugolecki, D. W., & Doherty, K. (1994). The impact of self-
presentations on self-appraisals and behavior: The power of public commit-
ment. Personality and Social Psychology Bulletin, 20, 20-33.
Schlenker, B. R, Britt, T. w., & Pennington, J. (1996). Impression regulation and
management: Highlights of a theory of self-identification. In R M. Sorrentino
& E. T. Higgins (Eds.), Handbook of motivation and cognition: The interpersonal
context (Vol. 3, pp. 118-147). New York: Guilford.
Schmidt,1. A., & Fox, N. A. (1995). Individual differences in young adults shyness
and SOCiability: Personality and health correlates. Personality and Individual
Differences, 19, 455-462.
Schoicket, S. (1980). Secrets. American Journal of Psychoanalysis, 40, 179-182.
Schwartz, G. E. (1990). Psychobiology of repression and health: A systems approach.
In 1. L. Singer (Ed.), Repression and dissociation: Implications for personality theory,
psychopathology, and health (pp. 405-434). Chicago: University of Chicago Press.
Schwartz, G. S., Friedlander, M. L., & Tedeschi, J. T. (1986). Effects of clients'
attributional explanations and reasons for seeking help on counselors' impres-
sions. Journal of Counseling Psychology, 33, 90-93.
Schwartz, R S. (1984). Confidentiality and secret-keeping on an inpatient unit.
Psychiatry, 47, 279-284.
Segal, D. L., & Murray, E. J. (1994). Emotional processing in cognitive therapy and
vocal expression of feeling. Journal of Social and Clinical Psychology, 13,189-206.
Servaes, P., Vingerhoets, A. J. J. M., Vreugdenhil, G., Keuning, J. J., & Broekhuijsen,
A. M. (1999). Inhibition of emotional expression in breast cancer patients.
Behavioral Medicine, 25, 23-27.
ShapirO, D. A., & Firth, J. A. (1987). Prescriptive vs. exploratory psychotherapy:
Outcomes of the Sheffield Psychotherapy Project. British Journal of Psychiatry,
151, 790-799.
Shapiro, D., Jamner, L. D., & Goldstein, I. B. (1993). Ambulatory stress psycho-
physiology: The study of "compensatory and defensive counterforces" and
conflict in a natural setting. Psychosomatic Medicine, 55, 309-323.
Shlien,1. (1984). Secrets and the psychology of secrecy. In R Levant & J. Shlien
(Eds.), Client centered therapy and the person centered approach (pp. 390-399).
New York: Praeger.
Shneidman, E. S., & Farberow, N. L. (1961). Statistical comparisons between at-
REFERENCES 245

tempted and committed suicides. In N. L. Farberow & E. S. Shneidman (Eds.),

The cry for help (pp. 19-47). New York: McGraw-Hill.
Shneidman, E. 5., & Farberow, N. L. (1970). Attempted and committed suicides. In
E. S. Shneidman, N. L. Farberow, & R E. Litman (Eds.), The psychology of suicide
(pp. 199-225). New York: Science House.
Shoham-Salomon, V., & Rosenthal, R (1987). Paradoxical interventions: A meta-
analysis. Journal of Consulting and Clinical Psychology, 55, 22-28.
Shortt, J. w., & Pennebaker, J. W. (1992). Talking versus hearing about Holocaust
experiences. Basic and Applied Social Psychology, 13, 165-179.
Shrauger, J. 5., & Schoeneman, 1. J. (1979). Symbolic interactionist view of self-
concept: Through the looking glass darkly. Psychological Bulletin, 86, 549-573.
Siegman, A. W. (1993). Cardiovascular consequences of expressing, experiencing,
and repressing anger. Journal of Behavioral Medicine, 16, 539-569.
Siegman, A. W. (1994a). From type A to hostility to anger: Reflections on the history
of coronary-prone behavior. In A. W. Siegman & T. W. Smith (Eds.), Anger,
hostility, and the heart (pp. 1-21). Hillsdale, NJ: Lawrence Erlbaum.
Siegman, A. W. (1994b). Cardiovascular consequences of expressing and repressing
anger. In A. W. Siegman & 1. W. Smith (Eds.), Anger, hostility, and the heart (pp.
173-197). Hillsdale, NJ: Lawrence Erlbaum.
Siegman, A. w., & Snow, S. C. (1997). The outward expression of anger, the inward
experience of anger and CVR: The role of vocal expression. Journal of Behavioral
Medicine, 20, 29-45.
Siegman, A. w., Anderson, R w., & Boyle, S. (1991, March). Repression, impression
management, trait anxiety, and cardiovascular reactivity in men and women. Paper
presented at the annual meeting of the Society for Behavioral Medicine, Wash-
ington, IX.
Siegman, A. W., Anderson, R, Herbst, J., Boyle,S., & WIlkinson, J. (1992). Dimen-
sions of anger-hostility and cardiovascular reactivity in provoked and an-
gered men. Journal of Behavioral Medicine, 15, 257-272.
Sifneos, P. (1979). Short-term dynamic psychotherapy. New York: Plenum Press.
Silver, D. (1983). Psychotherapy and the characterologically difficult patient. Cana-
dian Journal of Psychiatry, 28, 513-521.
Silver, R L., & Wortman, C. B. (1980). Coping with undesirable life events. In J.
Garber & M. E. P. Seligman (Eds.), Human helplessness: Theory and applications
(pp. 279-375). New York: Academic Press.
Silver, R L., Boon, c., & Stones, M. H. (1983). Searching for meaning in misfortune:
Making sense of incest. Journal of Social Issues, 39, 81-102.
Silver, R c., Wortman, C. B., & Crofton, C. V. (1990). The role of coping in support
provision: The self-presentation dilemma of victims of life crises. In B. R
Sarason, I. G. Sarason, & G. R Pierce (Eds.), Social support: An interactional view
(pp. 397-426). New York: Wiley.
Silverstein, J. L. (1993). Secrets versus privacy in group psychotherapy. Group, 17,
107-114.
Silvia, P. J., & Duval, 1. S. (2001). Predicting the interpersonal targets of self-serving
attributions. Journal of Experimental Social Psychology, 37, 333-340.
246 REFERENCES

Sloan, W W, & Stiles, W B. (1994). Client self-disclosure and psychotherapy outcome.

Unpublished manuscript.
Smart, L., & Wegner, D. M. (1999). Covering up what can't be seen: Concealable
stigma and mental control. Journal of Personality and Social Psychology, 77,
474-486.
Smith, M. L., Glass, G. v., & Miller, T. I. (1980). The benefits of psychotherapy. Bal-
timore: The Johns Hopkins University Press.
Smyth, J. M., (1998). Written emotional expression: Effect sizes, outcome types, and
moderating variables. Journal of Consulting and Clinical Psychology, 66,174-184.
Smyth, J. M., Stone, A A, Hurewitz, A, & Kaell, A (1999). Effects of writing about
stressful experiences on symptom reduction in patients with asthma or rheu-
matoid arthritis: A randomized trial. Journal of the American Medical Association,
281,1304-1309.
Sophie, J. (1988). Internalized homophobia and lesbian identity. In E. Coleman
(Ed.), Integrated identity for gay men and lesbians (pp. 53-65). New York: Har-
rington Park.
Spera, S. P., Buhrfeind, E. D., & Pennebaker, J. W (1994). Expressive writing and
coping with job loss. Academy of Management Journal, 37, 722-733.
Spiegel, D. (1992). Effects of psychosocial support on patients with metastatic
breast cancer. Journal of Psychosocial Oncology, 10, 113-120.
Spiegel, D., Bloom, J. H., Kraemer, H. c., & Gottheil, E. (1989). Effects of psychoso-
cial treatment of patients with metastatic breast cancer. Lancet, 2,888-891.
Spielberger, C. D., Crane, R. S., Kearns, W D., Pellegrin, K. L., Rickman, R. L., &
Johnson, E. H. (1991). Anger and anxiety in essential hypertension. In C. D.
Spielberger, I. G. Sarason, Z. Kulcsar, & G. L. Van Heck (Eds.), Stress and
emotion: Anxiety, anger, and curiosity (Vol. 14, pp. 265-283). Washington, DC:
Hemisphere.
Starr, K. W (1998, September 9). Narrative. In The Starr report (online). Available:
http://www.alllinks.com/icreport/6narrit.htm#N_42......
Steinbeck, J. (1961). The winter of our discontent. New York: Viking Press.
Stice, E. (1992). The similarities between cognitive dissonance and guilt: Confession
as a relief of dissonance. Current Psychology: Research and Reviews, 11, 69-77.
Stiles, W. B. (1984). Client disclosure and psychotherapy session evaluations. British
Journal of Clinical Psychology, 23, 311-314.
Stiles, W B. (1987). "1 have to talk to somebody." A fever model of disclosure. In V. J.
Derlega & J. H. Berg (Eds.), Self-disclosure: Theory, research, and therapy (pp. 257-
282). New York: Plenum Press.
Stiles, W. B. (1995). Disclosure as a speech act: Is it psychotherapeutic to disclose? In
J. W Pennebaker (Ed.), Emotion, disclosure, and health (pp. 71-91). Washington,
DC: American Psychological Association.
Stiles, W B., & Shapiro, D. A (1994). Disabuse of the drug metaphor: Psycho-
therapy of process-outcome correlations. Journal of Consulting and Clinical
Psychology, 62, 942-948.
Stiles, W B., & Sultan, F. E. (1979). Verbal response mode use by clients in psycho-
therapy. Journal of Consulting and Clinical Psychology, 47, 611-613.
REFERENCES 247

Stiles, W. B., McDaniel, S. H., & McGaughey, K. (1979). Verbal response mode
correlates of experiencing. Journal of Consulting and Clinical Psychology, 47,
795-797.
Stiles, W. B., Putnam, S. M., & Jacob, M. C. (1982). Verbal exchange structure of
initial medical interviews. Health Psychology, 1, 315-336.
Stokes, J. P. (1983). Predicting satisfaction with social support from social network
structure. American Journal of Community Psychology, 11, 141-152.
Stoler, N. (1963). Client likability: A variable in the study of psychotherapy. Journal
of Consulting Psychology, 27, 175-178.
Stone, A., Kennedy-Moore, E., & Neale, J. (1995). Association between daily coping
and end-of-the-day mood. Health Psychology, 14, 341-349.
Strack, S., & Coyne, J. C. (1983). Social confirmation of dysphoria: Shared and
private reactions to depression. Journal of Personality and Social Psychology, 44,
798-806.
Strassberg, D. S., & Anchor, K. N. (1977). Ratings of client self-disclosure and
improvement as a function of sex of client and therapist. Journal of Clinical
Psychology, 33, 239-241.
Strassberg, D. S., Roback, H. B., Anchor, K. N., & Abramowitz, S. I. (1975). Self-
disclosure in group therapy with schizophreniCs. Archives of General Psychiatry,
32,1259-1261.
Strassberg, D. S., Anchor, K. N., Gabel, H., & Cohen, B. (1978). Client self-disclosure
in short-term psychotherapy. Psychotherapy: Theory, Research, and Practice, 15,
153-157.
Strong, S. R (1968). Counseling: An interpersonal influence process. Journal of
Counseling Psychology, 15, 215-224.
Strong, S. R (1987). Interpersonal influence theory as a common language for
psychotherapy. Journal of Integrative and Eclectic Psychotherapy, 6,173-184.
Strong, S. R (1991). Theory-driven science and naive empiricism in counseling
psychology. Journal of Counseling Psychology, 38, 204-210.
Strong, S. R:. (1995). Interpersonal influence theory: The situational and individual
determinants of interpersonal behavior. In D. J. Lubinski & R Dawis (Eds.),
Assessing individual differences in human behavior: New concepts, methods, and
findings (pp. 263-295). Palo Alto, CA: Davies-Black Publishing.
Strong, S. R, & Claiborn, C. D. (1982). Change through interaction: Social psychological
processes of counseling and psychotherapy. New York: Wiley-Interscience.
Stroop, J. R (1935). Studies of interference in serial verbal reactions. Journal of
Experimental Psychology, 18, 643-662.
Strupp, H. H., & Hadley, S. W. (1979). Specific vs. nonspecific factors in psycho-
therapy: A controlled study of outcome. Archives of General Psychiatry, 36, 1125-
1136.
Sturkie, K. (1983). Structured group treatment for sexually abused children. Health
and Social Work, 8, 299-308.
Suarez, E. c., & Williams, R B. (1990). The relationships between dimensions of
hostility and cardiovascular reactivity as a function of task characteristics.
Psychosomatic Medicine, 52, 558-570.
248 REFERENCES

Suedfeld, P., & Pennebaker, J. W. (1997). Health outcomes and cognitive aspects of
recalled negative life events. Psychosomatic Medicine, 59, 172-177.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: W. W. Norton.
Suls, J., & Fletcher, B. (1985). The relative efficacy of avoidant and nonavoidant
coping strategies: A meta-analysis. Health Psychology, 4, 249-288.
Suomi, S. J. (1991). Uptight and laid-back monkeys: Individual differences in the
response to social challenges. In S. Branch, W. Hall, & E. Dooling (Eds.),
Plasticity of development (pp. 27-55). Cambridge, MA: MIT Press.
Swann, W. B. (1987). Identity negotiation: Where two roads meet. Journal of Person-
ality and Social Psychology, 53, 1038-1051.
Swann, W. B. (1996). Self-traps: The elusive quest for higher self-esteem. New York:
W. H. Freeman.
Swann, W. B., & Ely, R J. (1984). A battle of wills: Self-verification versus behavioral
confirmation. Journal of Personality and Social Psychology, 46, 1287-1302.
Swann, W. B., & Hill, c. A. (1982). When our identities are mistaken: Reaffirming
self-conceptions through social interaction. Journal of Personality and Social
Psychology, 43, 59-66.
Swann, W. B., & Predmore, S. C. (1985). Intimates as agents of social support:
Sources of consolation or despair? Journal of Personality and Social Psychology,
49,1609-1617.
Swann, W. B., & Read, S. J. (1981). Acquiring self-knowledge: The search for
feedback that fits. Journal of Personality and Social Psychology, 41, 1119-1128.
Swanson, L., & Biaggio, M. K (1985). Therapeutic perspectives on father-daughter
incest. American Journal of Psychiatry, 142, 667-674.
Sweetser, E. E. (1987). The definition of "lie": An examination of the folk models
underlying a semantic prototype. In D. Holland & N. Quinn (Eds.), Cultural
models in language and thought (pp. 43-66). New York: Cambridge University
Press.
Swink, K K, & Leveille, A. E. (1986). From victim to survivor: A new look at the
issues and recovery process for adult incest survivors. Women and Therapy, 5,
119-14l.
Szajnberg, N. (1988). The developmental continuum from secrecy to privacy in a
psychodynamic milieu. Residential Treatment for Children and Youth, 6, 9-28.
Tait, R, & Silver, R C. (1989). Coming to terms with major negative life events. In J.
S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 351-381). New York:
Guilford.
Tannenbaum, P. H., & Gaer, E. P. (1965). Mood changes as a function of stress of
protagonist and degree of identification in a film viewing situation. Journal of
Personality and Social Psychology, 2, 612-616.
Tausk, V (1933). On the origin of the "influencing machine" in schizophrenia.
Psychoanalysis Quarterly, 2, 519-556.
Taylor, S. E. (1990). Health psychology: The science and the field. American Psycholo-
gist, 45, 40-50.
Taylor, S. E., & Armor, D. A. (1996). Positive illusions and coping with adversity.
Journal of Personality, 64, 873-898.
REFERENCES 249

Taylor, S. E., & Brown, J. D. (1988). illusion and well-being: A social psychological
perspective on mental health. Psychological Bulletin, 103, 193-210.
Taylor, S. E., & Brown, J. D. (1994). Positive illusions and well-being revisited:
Separating fact from fiction. Psychological Bulletin, 116, 21-27.
Taylor, S. E., & Gollwitzer, P. M. (1995). Effects of mindset on positive illusions.
Journal of Personality and Social Psychology, 69, 213-226.
Temoshok, L. (1983). Emotion, adaptation, and disease: A multidimensional theory.
In L. Temoshok, C. v. Dyke. & L. S. Zegans (Eds.), Emotions in health and illness.
New York: Grune & Stratton.
Temoshok, L. (1987). Personality, coping style, emotion, and cancer: Towards an
integrative model. Cancer Surveys, 6, 545-567.
Tesser, A (1988). Toward a self-evaluation maintenance model of social behavior. In
L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 181-
227). San Diego, CA: Academic Press.
Tesser, A, Leone, c., & Clary, E. (1978). Affect control: Process constraints versus
catharsis. Cognitive Therapy and Research, 2, 265-274.
Thompson, B. J., & Hill, C. E. (1991). Therapist perceptions of client reactions.
Journal of Counseling and Development, 69, 261-265.
Thoits, P. (1982). Conceptual, methodological and theoretical problems in studying
social support as a buffer against life stress. Journal of Health and Social Behavior,
23,145-159.
Tice, D. M. (1992). Self-concept change and self-presentation: The looking glass self
is also a magnifying glass. Journal of Personality and Social Psychology, 63, 435-451.
Tice, D. M., & Ciarocco, N. J. (1998). Inhibition and self-control. Psychological
Inquiry, 9,228-231.
Tichenor, v., & Hill, C. E. (1989). A comparison of six measures of working alliance.
Psychotherapy, 26, 195-199.
Traue, H. C. (1995). Inhibition and muscle tension in myogenic pain. In J. w.
Pennebaker (Ed.), Emotion, disclosure, and health (pp. 155-175). Washington,
DC: American Psychological Association.
Traue, H. c., & Kraus, W. (1988). Ausdruckshemmung als Risikofaktor: Eine ver-
haltensmedizinische Analyses. Praxis der angewandten Verhaltensmedizin und
Rehabilitation, 2, 85-95.
Traue, H. c., & Michael, A (1993). Behavioral and emotional inhibition in head
pain. In H. C. Traue & J. w. Pennebaker (Eds.), Emotion, inhibition, and health
(pp. 226-246). Kirkland, WA: Hogrefe & Huber.
Trinder, H., & Salkovskis, P. N. (1994). Personally relevant intrusions outside the
laboratory: Long-term suppression increases intrusion. Behaviour Research and
Therapy, 32, 833-842.
Truax, C. B. (1966). Therapist empathy, warmth, and genuineness and patient
personality change in group psychotherapy: A comparison between inter-
action unit measures, time sample measures, and patient perception measures.
Journal of Clinical Psychology, 22, 225-229.
Truax, C. B., & Carkhuff, R. R. (1967). Toward effective counseling and psychotherapy:
Training and practice. Chicago: Aldine.
250 REFERENCES

Truax, C. B., & Mitchell, K. M. (1971). Research on certain therapist interpersonal

skills in relation to process and outcome. In A E. Bergin & S. 1. Garfield (Eds.),
Handbook of psychotherapy and behavior change (pp. 299-344). New York: Wiley.
Tsai, M., & Wagner, N. N. (1978). Therapy groups for women sexually molested as
children. Archives of Sexual Behavior, 7,417-427.
Turner, R J. (1983). Direct, indirect, and moderating effects of social support on
psychological distress and associated conditions. In H. B. Kaplan (Ed.), Psycho-
logical stress: Irends in theory and research (pp. 105-1SS). New York: Academic Press.
Tversky, A, & Kahneman, D. (1973). Availability: A heuristic for judging frequency
and probability. Cognitive Psychology, 5, 207-232.
Vallacher, R R, & Wegner, D. M. (1987). What do people think they're doing?
Action identification and human behavior. Psychological Review, 94, 3-15.
Vangelisti, A 1. (1994). Family secrets: Forms, functions, and correlates. Journal of
Social and Personal Relationships, 11, 113-135.
Vangelisti, A 1., & Caughlin, J. P. (1997). Revealing family secrets: The influence of
topic, function, and relationships. Journal of Social and Personal Relationships,14,
679-705.
Van Rood, Y. R, Bogaards, M., Goulmy, E., & Van Houwelingen, H. C. (1993). The
effects of stress and relaxation on the in vitro immune response in man: A
meta-analytic study. Journal of Behavioral Medicine, 16, 163-181.
Vinokur, A D., & Van Ryn, M. (1993). Social support and undermining in close
relationships: Their independent effects on the mental health of unemployed
persons. Journal of Personality and Social Psychology, 65, 350-359.
Wallerstein, R S. (1986). Forty-two lives in treatment: A study of psychoanalysis and
psychotherapy (pp. 265-277). New York: Guilford.
Watson, D., Clark, 1. A, & Tellegen, A (1988). Development and validation of brief
measures of positive and negative affect: The PANAS scales. Journal of Person-
ality and Social Psychology, 54, 1063-1070.
Wegner, D. M. (1989). White bears and other unwanted thoughts. New York: Penguin
Books.
Wegner, D. M. (1992). You can't always think what you want: Problems in the
suppression of unwanted thoughts. In M. Zanna (Ed.), Advances in experimental
social psychology (Vol. 25, pp. 193-225). San Diego, CA: Academic Press.
Wegner, D. M. (1994). Ironic processes of mental control. Psychological Review, 101,
34-52.
Wegner, D. M., & Erber, R (1992). The hyperaccessibility of suppressed thoughts.
Journal of Personality and Social Psychology, 63, 903-912.
Wegner, D. M., & Gold, D. B. (1995). Fanning old flames: Emotional and cognitive
effects of suppressing thoughts of a past relationship. Journal of Personality and
Social Psychology, 68, 782-792.
Wegner, D. M., & Wenzlaff, R M. (1996). Mental control. In E. T. Higgins & A W.
Kruglanski (Eds.), Social psychology: Handbook of basic principles. New York:
Guilford.
Wegner, D. M., Schneider, D. J., Carter, S. R, ill, & White, T. (1987). Paradoxical
effects of thought suppression. Journal of Personality and Social Psychology, 53,
5-13.
REFERENCES 251

Wegner, D. M., Shortt, J.w., Blake, A W., & Page, M. S. (1990). The suppression of
exciting thoughts. Journal of Personality and Social Psychology, 58, 409-418.
Wegner, D. M., Lane, J. D., & Dimitri, S. (1994). The allure of secret relationships.
Journal of Personality and Social Psychology, 66, 287-300.
Weigel, R G., Dinges, N., Dyer, R, & Straumfjord, A A (1972). Perceived self-
disclosure, mental health, and who is liked in group treatment. Journal of
Counseling Psychology, 19, 47-52.
Weinberger, D. A (1990). The construct validity of the repressive coping style. In J.
L. Singer (Ed.), Repression and dissociation: Implications for personality theory,
psychopathology, and health (pp. 337-386). Chicago: Chicago University Press.
Weinberger, D. A, Schwartz, G. E., & Davidson, R J. (1979). Low-anxious, high-
anxious, and repressive coping styles: Psychometric patterns and behavioral
and physiological responses to stress. Journal of Abnormal Psychology, 88,
369-380.
Weinberger, J. (1995). Common factors aren't so common: The common factors
dilemma. Clinical Psychology: Science and Practice, 2, 45-69.
Weinstein, H. M., & Richman, A (1984). The group treatment of bulimia. Journal of
American College Health, 32, 208-215.
Weiss, J. (1995). Bernfeld's "The facts of observation in psychoanalysis: A response
from psychoanalytic research." Psychoanalytic Quarterly, 64, 699-716.
Welsh, A (1994). Freud's wishful dream book (pp. 29-50). Princeton, NJ: Princeton
University Press.
Wenzlaff, R M., Wegner, D. M., & Klein, S. B. (1991). The role of thought suppres-
sion in the association of thought and mood. Journal of Personality and Social
Psychology, 60,500-508.
Wergeland, H. (1980). Elective mutism. Annual Progress in Child Psychiatry and Child
Development, 65, 373-385.
Westfall, A (1989). Extramarital sex: The treatment of the couple. In G. R Weeks
(Ed.), Treating couples (pp. 163-190). New York: Brunner/Mazel.
Wills, T. A (1978). Perceptions of clients by professional helpers. Psychological
Bulletin, 85, 968-1000.
Wilner, N., & Horowitz, M. J. (1975). Intrusive and repetitive thoughts after a
depressing film: A pilot study. Psychological Reports, 37, 135-138.
Wilson, T. D., & Kraft, D. (1993). Why do I love thee? Effects of repeated introspec-
tions about a dating relationship on attitudes toward the relationship. Person-
ality and Social Psychology Bulletin, 19, 409-418.
Wilson, T. D., Lisle, D. J., & Schooler, J. (1988). Some undesirable effects of self-
reflection. Unpublished manuscript, Charlottesville: University of Virginia.
Wilson, T. D., Dunn, D. S., Kraft, D., & Lisle, D. J. (1989). Introspection, attitude
change, and attitude-behavior consistency: The disruptive effects of explain-
ing why we feel the way we do. In L. Berkowitz (Ed.), Advances in experimental
social psychology (Vol. 19, pp. 123-205). Orlando, FL: Academic Press.
Wmdle, M. (1994). Temperamental inhibition and activation: Hormonal and psy-
chosocial correlates and associated psychiatric disorders. Personality and Indi-
vidual Differences, 17, 61-70.
Wmnicott, D. W. (198O). Playing and reality. New York: Penguin Books.
252 REFERENCES

Winter, D. A (1985). Group psychotherapy with depressives: A personal construct

theory perspective. International Journal of Mental Health, 13, 26-38.
Wolff, C. T., Friedman, S. B., Hofer, M. A, & Mason, J. W. (1964). Relationship
between psychological defenses and mean urinary 17-hydroxycorticosteroid
excretion rates: I. A predictive study of parents of fatally ill children. Psycho-
somatic Medicine, 26, 576-591.
Woods, K., & McNamara, J. R. (1980). Confidentiality: Its effect on interviewee
behavior. Professional Psychology, 11,714-721.
Wortman, C. B. (1984). Social support and the cancer patient: Conceptual and
methodological issues. Cancer, 53, 2339-2360.
Wortman, C. B., & Conway, T. (1985). The role of social support in adaptation and
recovery from phYSical illness. In S. Cohen (Ed.), Social support and health (pp.
281-308). New York: Academic.
Wortman, C. B., & Lehman, D. R (1985). Reactions to victims of life crises: Support
attempts that fail. In I. G. Sarason & B. R Sarason (Eds.), Social support: Theory,
research, and applications (pp. 463-489). Dordrecht, The Netherlands: Martinus
Nijhoff.
Wortman, C. B., & Silver, R C. (1987). Coping with irrevocable loss. In G. R Vanden
Bos & B. K. Bryant (Eds.), Cataclysms, crises, and catastrophes: Psychology in
action (pp. 189-235). Washington, DC: American Psychological Association.
Wortman, C. B., & Silver, R C. (1989). The myths of coping with loss. Journal of
Consulting and Clinical Psychology, 57, 349-357.
Wortman, C. B., Adesman, P., Herman, E., & Greenberg, R (1976). Self-disclosure:
An attributional perspective. Journal of Personality and Social Psychology, 33,
184-191.
Wright, F. (1990). Discussion: Of "common dilemmas in combined individual and
group treatment" by Anne Alsonso and J. Scott Rutan. Group, 14, 13-15.
Wright, T. L., Ingraham, L. J., Chemtob, H. J., & Perez-Arce, P. (1985). Satisfaction
and things not said: Clinical tools for group therapists. Small Group Behavior,
16,565-572.
Yalom,I. D. (1985). The theory and practice ofgroup psychotherapy (4th ed.). New York:
Basic Books.
Yep, G. A (2000). Disclosure of HIV infection in interpersonal relationships: A
communication boundary management approach. In S. Petronio (Ed.), Balanc-
ing the secrets of private disclosures. LEA's communication series (pp. 83-96).
Mahwah, NJ: Erlbaum.
Yovetich, N. A, & Drigotas, S. M. (1999). Secret transmission: A relative intimacy
hypothesis. Personality and Social Psychology Bulletin, 25, 1135-1146.
Zeigarnik, B. (1927). Uber das behalten von erledigten und underledigten han-
dlungen. Psychologische Forschung, 9, 1-85.
Zuckerman, M., Klorman, R, Larrance, D. T., & Spiegel, N. H. (1981). Facial,
autonomic, and subjective components of emotion: The facial feedback hy-
pothesis versus the externalizer-internalizer distinction. Journal of Personality
and Social Psychology, 41, 929-944.
Zung, W. w., & Gianturco, J. A (1971). Personality dimension and the Self-Rating
Depression Scale. Journal of Clinical Psychology, 27, 247-248.
INDEX

Abortion, concealment or disclosure of, 50, Arousal (con t.)

54,179,192-193,194,202 negative emotional, effect of revelation
Acquaintances, shared, 200 of secrets on, 174
Advice, unhelpful, 170 repression-related increase in, 27-28
Aggression, avoidance of, 24-25 as response to others' distress, 168-169
AIDS (acquired immunodeficiency Asthma patients, trauma writing
syndrome), as family secret, 112 experiment with, 77-78
Alcoholism Audience effects, in self-presentation, 138-
revelation of, in group therapy, 114-116 140, 166, 167
as trauma response, 82 Autonomic reactivity: see also Arousal
Allen, Joan, 187 repression-related increase in, 25, 27-28
Amygdala, overresponsive, 58, 59 Availability bias phenomenon, 201
Anger
as cardiovascular hyperactivity risk Back pain
factor, 29 repression as risk factor for, 25
as coronary heart disease risk factor, 29, 31 self-concealment as risk factor for, 32, 37
expression of, 29, 87, 90-01 Behavioral control, negative correlation
suppression of, as low-back pain risk with self-concealment, 34, 35
factor, 25 Belonging, as basic human need, 16, 169-
Anxiety 170
genetic factors in, 59 Bereaved persons
effect on immunologic functioning, 30-31 depression in, 83
positive correlation with secrecy, 23-24 mothers, of sudden infant death
positive correlation with self- syndrome victims, 180
concealment, 32, 37 negative advice given to, 170
Anxiety disorders, exposure therapy for, spouses, of suicide and automobile crash
213 victims, 210-211
Arabs, in psychotherapy training groups, Blaming, of victims, 169
114 Borderline personality disorder, 107
Arousal Boundaries
catharsis-related decrease in, 86 personal, effect of secrecy on, 172-173
endogenous opioids-mediated, 31 for revelation of secrets, 206-208

253
254 INDEX

Breast cancer, repression as risk factor for, Clinton, Hillary, 161-162

25-27 Closure
Breast cancer patients as component of coping, 91
emotional expression by, 179-180 new insights as, 95-96, 99
participation in support groups, 77 Cognitive accessibility, of secrets, 48-49
significant others' support for, 179-180 Cognitive-behavioral therapy, clients'
survival time of, effect of self- humiliating disclosures during, 153
concealment/revelation of secrets Cognitive load, high, 201-202
on, 36, 69 Cognitive therapy, for traumatic events, 87
Breuer, Josef, 85 College students
Brief Symptom Inventory, 32 adjustment to stressors, 180-181
Brief therapy, 82, 143 attitudes toward admission of cheating,
Bulimia 212
revelation of, in group therapy, 114-116 emotional inhibition in, 43, 44-45
as trauma response, 82 expression of traumatic experiences by
Bush, George, 3-4, 20 spoken expression, 73-75
written expression, 72-73, 76-78, 120
Cancer, secrecy as risk factor for, 23-24 frequency of lying by, 9
among homosexuals, 35, 36, 56 hidden reactions of, 122
Cancer patients low self-esteem of, 54
emotional expression by, 179 pain suppression by, 43
fear experienced by, 169 rating of desired qualities of confidants
repressive coping style of, 25-27 by, 184-185
Cardiovascular disease revelation of secrets by, 17, 69-70
excessive self-disclosure as risk factor as source of catharsis versus new
for, 25 insights, 89-90, 91-98
repression as risk factor for, 28-29, 31 secret keeping by, 16
Case study method, of psychoanalysis, types of secrets kept, 10-12
106-107 self-concealment by, 32-33
Catharsis positive correlation with social
definition of, 85 anxiety, 33-34
revelation of secrets-related, 81, 85-88 sexual secrets of, 10, 11
as negative affect cause, 87, 88, 90-91, stigmatizing characteristics of, 179
94-95, 96, 99 thought suppression by, 48-50
negative effects of, 184 traumatic events exposure by, 211
versus new insights, 88-99 Colon cancer patients, emotional
positive effects of, 98 expression by, 180
self-presentational view of, 153 Communality, of secrets, 213
Censorship, Freud's concept of, 103 Compulsive behavior, repression-related,
Cheating, students' revelation of, 212 104-105
Child abuse, psychotherapy clients' Confidants
admission of, 204 appropriate/helpful
Childhood abuse, forgetting of, during discretion of, 182, 184, 185
adult life, 52-53 nonjudgmental nature of, 182-183, 184
Children in particular relationships, 196-199
exposure to violence, 180 as source of new insights, 183-184, 185
inhibited and noninhibited feedback from, 178-181, 186
temperaments of, 58-59 judgmental nature of, 162
self-concealment by, 172 optimal number of, 185
Clinton, Bill, 8-9, 20, 205 qualities of, 178-185
INDEX 255

Confidants (cant.) Diagnostic Interview for Children and

rejection by, 162 Adolescents-Parent Version, 62
revelation of secrets by, 162, 164-165, 173 Disapproval, as reason for secrecy, 16
inadvertent, 198 Disclosure: see also Revelation
revelation of secrets to of psychological distress, to therapist,
negative effects of, 168-171, 173 60-61
positive effects of, 174-178 Discretion, of confidants, 182, 184, 185
Consumer Reports, 117 Dissociative identity disorder, 106
"Contender, The" (movie), 187-188 Dream analysis, 104
Conventional topics, as family secrets, 14 Drug abuse, 3-4
Coping by Bill Clinton and George Bush, 3-4, 20
defensive strategies for, 6
repressive strategies for: see Repression Earthquake survivors, 36
Coronary heart disease, repression as risk Eating disorders, concealment of, 50-52
factor for, 28, 31 Emotional constriction, correlation with
Cortisol secretion self-concealment, 34, 35
behavioral inhibition-related increase in, Emotional experiences: see Traumatic
63 experiences
repression-related increase in, 29-30, 31 Emotional expression
Countertransference, 107 as catharsis, 85-88
Covert processes, in psychotherapy, 121- conflict over, positive correlation with
126 self-concealment, 34
Co-workers, revelation of personal distortions of, 171-172
information to, 167-168 inhibition of: see also Repression
Cross-sex relationships: see also Romantic effect on experience of negative
relationships emotions, 45-47
types of secrets kept in, 13-14 effect on experience of positive
Crying, effect on immunologic function, 45 emotions,44-45,46-47
genetic factors in, 57-59
Dahmer, Jeffrey, 23-24, 39, 41, 149 negative correlation with self-
Dahmer, Lionel, 23 concealment, 34
"Dead Man Walking" (movie), 67-69 as source of new insights, 88
Death as stress response, 179
of family members, 97, 111 effect on cognitive and emotional
of infants, mothers' intrusive thoughts adjustment, 180-181
about, 180 Emotion Control Questionnaire, 29
Defensive coping mechanisms, repression Empathy, of therapists, 182-183, 206
as, 6 Encounter groups, 19-20, 119
Depressed persons, others' negative Epstein-Barr virus, as immunologic
reactions to, 168, 169, 171 function/ stress response indicator,
Depression 30-31,70-71, 75
bereavement-related, 83 Exposure therapy, 82, 119-120
genetic factors in, 59 for anxiety disorder patients, 213
negative life events-related, 97 for rape victims, 119-120, 153
relationship with repression, 25 self-presentational view of, 153-154
secret keeping-related, 23-24 Expression, as motivation for revealing
self-concealment-related, 32, 37, 38, 56 secrets, 17, 18
shyness-related, 63 Expressive therapies, 88
Diagnostic and Statistical Manual of Mental Extramarital! extrarelationship affairs
Disorders, 205 revelation of, 1-2, 4-5, 174-175
256 INDEX

Extramarital/ extrarelationship affairs Guilt

(cont.) experienced by psychotherapy clients,
revelation of (cont.) 204
in family therapy, 109, 110 extramarital affairs-related, 1-2
secrecy of, 14, 109-110 effect of revelation of secrets on, 174-175,
as source of self-identity, 113 213,214

Failure-related secrets, 10, 15 Headaches

prevalence of, 21 repression as risk factor for, 25
of psychotherapy clients, 13 self-concealment as risk factor for, 32, 37
False consensus effect, 175-176 Health, effect of secrets or revelation of
Family members, death of, 97, 111 secrets on, 69-79
Family secrets, 14-15 cancer risk, 23-24, 35, 36, 56
revelation of cardiovascular disease risk, 25, 28-29, 31
adverse effects of, 126 inhibition model of, 42-47, 56-57
during family therapy, 107-1U, 126 predisposing/ genetic factors in, 57-64
Family therapy, clients' secrecy and preoccupation model of, 47-54, 57
openness in, 107-112, 116 of previously-disclosed versus
Fear undisclosed events, 75-76
experienced by cancer patients, 169 of real versus imagined events, 76-77,
of public speaking, 86-87 79
of the unfamiliar, 58 versus revelation of trivial events, 69-72
Fearfulness, positive correlation with through written versus vocal
shyness, 63 expressions, 74-75
"Four Weddings and a Funeral" (movie), in undergraduate versus clinical
215-216 samples, 77-78
Free association, 104 Heart rate, effect of repression on, 28
Freud, Signum, 5, 6,85,101,103-105,106, Hepatitis B antibodies, 70
202 Hidden reactions, in psychotherapy, 121-
Friends 124
lies told by, 7-8 Holocaust survivors, 85-86, 169
revelation of secrets by, 17 Homosexuals
as social support, 155-156 concealment of sexual orientation by
as cancer risk factor, 35, 36, 56
Genetic factors effect on HIV infection progression,
in anxiety, 59 35-36,168
in depression, 59 revelation of sexual orientation by
in inhibition of emotional expression, effect on health, 168
57-59 to parents, 111
in pain, 59 stigmatization of, 178-179
in secrecy-related illness, 57-59, 63-64 Hostility
Genital herpes infection, revelation to effect on cardiovascular fitness, 28
sexual partner, 215 as low back pain risk factor, 25
Graham, David, 1-2, 4-5, 21 in therapist-client relationship, 124
Group therapy Human immunodeficiency virus (HIV)
members' hidden reactions in, 122 infection, progression of, effect of
members' openness and secrecy in, 114- sexual orientation concealment on,
117 35-36, 168
negative effects of self-disclosure in, 102 Human immunodeficiency virus (HIV)
schizophrenics' self-disclosure in, 118, 125 infection testing, 4
INDEX '157

Humiliation, effect of revelation of secrets Insights, new, derived from revelation of

on, 213 secrets, 81-85, 174
Humor, effect on immunologic function, 45 versus catharsis, 88-99
Hygiene rituals, 4 as closure, 95-96, 99
Hypertension, positive correlation with confidants as source of, 183-184, 185
secrecy, 36 therapists as source of, 90
Internalization, of self-beliefs, 136-141, 144,
Identity 147,156-157
effect of extramarital affairs on, 113 public commitment theory of, 166-167
effect of false judgments on, 187-188 reasons for, 166-168
public commitment to, 166, 167 Interpersonal influence theory, of
relationship with secrecy, 20, 21-22 psychotherapy, 155
of revealers of secrets, disclosure of, 164 Interpersonal relationships: see also
undermining effects of, 165-168, 173 Romantic relationships
Identity development, role of secrecy in, 10 model of revelation of secrets in, 187-217
during psychoanalysis, 105-106 communality of secrets component,
Identity management, 111 213
Immunologic functioning controllability of secrets component,
effect of anxiety on, 30-31 211-212
effect of crying on, 45 determination of confidant's
effect of humorous stimuli on, 45 appropriateness component, 196-
effect of repression on, 30-31 199, 206-207, 214, 216
effect of revelation of secrets on, 69, 70- emotional effects component, 213-214
71 limitations to, 209-214
effect of revelation of traumatic events negotiation of, 208
on, 75 partial! modified revelations
Impulsivity, positive correlation with self- component 203-209
concealment, 38, 39 personal privacy boundaries
Incest, as family secret, 108-109 component 206-208
Incest survivors, 82, 84 probability of being discovered
repression in, 108-109 component, 199-202, 214, 216-217
revelation of incest by reciprocity component, 198-199
in family therapy, 109 responsibility for secrets component,
in group therapy, 114-116 212
search for meaning by, 96, 211 revelation of themes vs. issues
Infants, inhibited and uninhibited component 203-206, 209-210
temperaments of, 58 revelation of troubling secrets
Infectious diseases, positive correlaiton component, 202-203
with secrecy, 23-24 secrecy vs. privacy component, 191-
Influenza, positive correlation with secrecy, 196, 206-208
36 spousal privilege consideration
Inhibited temperament types, 57-59, 62- component, 199
63,64 termination of relationship
Inhibition, 5 component, 209
positive correlation with shyness, 63 testing of, 214-216
Inhibition model, of secrecy-related illness, thought suppression component, 209-
42-47,56-57 210
Insight-oriented psychotherapy, timing component, 197-198
comparison with cathartic types of secrets revealed component,
psychotherapy, 86 209-210,214
258 INDEX

Interpersonal relationships (cont.) Mental illness-related secrets

model of revelation of secrets in (cont.) prevalence of, 21
writing about secrets component, 209 of psychotherapy clients, 13
need for, as reason for secrecy, 16 Metzinger, Alice, 41
taboo topics within, 13-14, 15-16 Military veterans, behavioral inhibition
Introversion, positive correlation with self- among, 63
concealment, 38 Millon Behavioral Health Inventory, 27
Izqueido, Elisa, 129 Minnesota Multiphasic Personality
Inventory (MMPI), 63, 118, U5
Jackson, Michael, 161 Mood changes, as response to others'
Jews, in psychotherapy training groups, distress, 168-169
114 Mothers, of sudden infant death victims,
Job loss, reemployment after, 71-72 180
Jones, Adrianne, 1-2 Murderers, confessions by, 1-2, 21, 163
Jung, Carl, 103-104
Negative affect
Kemper, Edmund Emil, 163 catharsis-related, 87, 88, 90-91, 94-95,
Kennedy, John F., 24 96,99
effect on cardiovascular fitness, 28
Lesbians, revelation of sexual orientation Neuroticism, positive correlation with
by, 110, 111 shyness, 63
Lewinsky, Monica, 8-9, 161-162, 205 New identity therapy, 85
Life satisfaction, effect of secrecy on, 35 Nonjudgmental nature, of confidants, 182-
Likability, effect of revelation of secrets on, 183,184
177-178
Limbic system, overarousal of, 58, 59 Obsessions, obsessive preoccupation-
Loma Prieta earthquake survivors, 36 related,47
Loneliness Obsessive-compulsiveness, 34
positive correlation with self- Offenses, as secrets, 11
concealment, 34, 38, 39 Onassis, Jacqueline Bouvier Kennedy, 24
positive correlation with shyness, 63 Opioids, endogenous, hyperactivity of, 31
Los Angeles, California, 168
Lying Pain
distinguished from secrecy, 7-9 genetic factors in, 59
false consensus effect in, 175 suppression of, 43, 46
prevalence of, 9 Parents, children's revelation of their
sexual orientation to, 110-111
Manning, Andrew, 187 Phobias, exposure therapy for, 82
Marijuana use, by Bill Clinton, 20 Play therapy, 111-1U
Marital satisfaction, effect of secrecy on, 173 Posttraumatic stress disorder, exposure
Marlowe-Crowne Social Desirability Scale, therapy for, 82, 119-UO
6,30-31 Power, Katherine, 41, 64
Masochism, as trauma response, 82 Powerless groups, lack of privacy of, 172
Masturbation, 104-105 Preface, to revelation of secrets, 17
McKearney, Jennifer, 2 Preference for Solitude Scale, 33-34
Memories Preoccupation model, of health
enhanced recall of, 53 consequences of secrecy, 47-54, 57
secret, 35 Primal therapy, 85
positive correlation with illness, 35, 37 Privacy
recall of, 61 distinguished from secrecy, 4-7, 210
INDEX 259

Privacy (cont.) Psychotherapy (cont.)

in psychotherapy, 158-159 clients' self-presentation in (cont.)
relationship with social status, 172 boundary conditions for, 148-151
Probability estimation, of discovery of components of, 130-131, 144
secrets, 199-202 counterarguments to, 151-157
Prolonged exposure therapy, for rape definition of, 132
victims, 120 description of, 143-148
Prostate cancer patients, emotional feedback component, 135, 136-137, 138,
expression by, 180 144, 145, 146, 148, 150-151, 156, 166
Psychiatric inpatients, self-presentation by, internalization component, 136-141,
133-134 144, 147, 156-157, 166-168
Psychoanalysis research about, 132-141
clients' openness during, 103-107, 116 self-beliefs component, 136-138, 140
fundamental rule of, 101, 154-155 self-change component, 134-136
Psychodrama therapy, 85, 105 self-identification theory as basis for,
Psychosomatic illness 131
inhibition model of, 42 theoretical background of, 131, 141,
repression as risk factor for, 25 142-143
shyness as risk factor for, 63 common factors in, 142-143
Psychotherapy interpersonal influence theory of, 155
clients' disclosure and secrecy in, 32, 33, privacy in, 158-159
60,101-127 utilization of, effect of self-concealment
countertransference and, 107 on,32-33,39
covert processes of, 121-126 variants of, 142
effect on disclosure/improved Public commitment, to an identity, 166, 167
symptomatology relationship, 117- Public speaking, fear of, 86-87
118,125-126
disclosure of psychological distress, Rape victims, 119-120, 153
60-61 Reciprocity, in revelation of secrets, 198-199
discussion of themes versus details, Reevaluation therapy, 85
204-206, 217 Relationship development, as motivation
empirical findings on, 117-127 for revealing secrets, 17, 18
in family therapy, 107-114, 116 Religiosity, as trauma response, 82
frequency of secrecy, 124-125 Repression
in group therapy, 114-117 as compulsive behavior cause, 104-105
hidden reactions in, 121-124 definition of, 24
intrusive thoughts and, 209 distinguished from secrecy, 5-7
negative effects of, 126-127, 209-210 as Freudian term, 103
versus private revelations, 72-74 physical consequences of, 24-31
in psychoanalysis, 103-107, 116 cancer, 25-27, 31
reasons for secrecy, 15-16 cortisol secretion, 29-30
relationship with symptom heart disease, 28-29
improvement, 118-120 immunologic functioning, 30-31, 38
effect on secret keeping/improved physiological arousal, 27-28
symptomatology relationship, 124- relationship with self-concealment, 5-7
126 Reputation
types of secrets kept, 12-13 measurement of, 200
clients' self-presentation in, 129-159 protection of, 188-189
audience effects component, 138-140, effect of social participation level on,
166,167 200-201
260 INDEX

Resolution, as motivation for revelation of Rheumatoid arthritis patients, trauma

secrets, 19 writing experiment with, 77-78
Restraint, negative correlation with self- Rich, Marc, 8
concealment, 34 Rituals, secret, 4
Revelation, of secrets, 55, 161-186 Robert Lee, Willie, 68-69
apprehension about, 59, 61, 64 Rogerian therapy, 150, 155
health effects of, 69-79 Romantic relationships
cancer risk, 23-24, 35, 36, 56 emotional expression in, 171
cardiovascular disease risk, 25, 28-29, false consensus effect in, 175-176
31 maintenance of personal boundaries in,
inhibition model of, 42-47, 56-57 172-173
preoccupation model of, 47-54, 57 partners' assumed honesty in, 175-176
motivation for, 1-2, 17-21 revelation of secrets within, 19-20
negative consequences of, 164-173, 185- secret keeping in, 172-173
186,188 beneficial effect of, 113
confidant feels burdened, 168-171, 173 types of secrets kept, 13-14
damage to social reputation, 185-186 taboo topics within, 13-14, 15-16
negative self-concept, 185-186 Rule violations, as family secrets, 14
rejection by listeners, 185-186 Rumination
revealer's identity is undermined, about negative life events, 97
165-168,173 positive correlation with self-
secrecy becomes more "rea!," 171-172, concealment, 34-35, 37, 39
173 as response to others' distress, 168-169
secret is repeated to others, 164-165, 173
violation of personal boundaries, 172- Scapegoating, within families, 108, 111, 116
173 Schizophrenics
positive consequences of, 67-79, 174-178, self-disclosure by, during group therapy,
186 118,125
avoidance of suspicions of reciprocal self-presentation by, 133-134
secrecy, 175-176 Secrecy
increased likability, 177-178, 186 contextual, 173
makes secrets seem less harmful, 176- definitions of, 3-9
177 developmental role of, 10
meaningful resolution of secrets, 174- distinguished from lying, 7-9
175, 186 distinguished from privacy, 4-7
new insights, 186 distinguished from repression, 5-7
relief from burden of secrecy, 186 individual differences in, 23-39
preface to, 17 effect on life satisfaction, 35
rules of, 16-17 nature of, 1-22
sequence of, 16-17 physical and psychological consequences
of sexual orientation of,202-203
effect on health, 168 cancer risk, 23-24, 25, 36, 56
to parents, 110-111 cardiovascular disease risk, 25, 28-29,
social context of, 161-162 31
as source of new insights genetic factors in, 57-59, 63-64, 217
as closure, 95-96, 99 inhibition model of, 42-47, 56-57
confidants as source of, 183-184, 185 predisposition to, 57-64
therapists as source of, 90 preoccupation model of, 47-54, 57
traumatic events as source of, 81-85 prevalence of, 9-10
stigmatization associated with, 161-162 privacy versus, 210
INDEX 261

Secrecy (cont.) Self-esteem (cont.)

reasons for, 15-16 low (cont.)
relationship with identity, 20, 21-22 positive correlation with self-
as stress cause, 41-42 perception, 54
Secret frame, 16-17 positive correlation with shyness, 63
Secret keepers, profile of, 38-39 self-enhancement-related increase in, 136
Secrets: see also Failure-related secrets; Self-identification theory, 131
Sexual secrets; Violence-related Self-perception, 166-167
secrets self-scanning, 166-167
cognitive accessibility of, 48-49 as source of low self-worth, 174
controllability of, 211-212 effect of stigmatization on, 178-179
hyperaccessibility of, 48 Self-perception model, of secrecy, 54-55,
probability of discovery of, 199-202 57
types of, 10-15 Self-presentation, 129-159
"Secrets and Lies" (movie), 67, 161 audience effects component, 138-140,
Self-beliefs 166,167
internalization of, 136-141, 144, 147, 156-157 boundary conditions for, 148-151
public commitment theory of, 166-167 components of, 130-131, 144
reasons for, 166-168 counterarguments to, 151-157
self-perception theory of, 166-167 definition of, 132
effect of self presentation on, 136-138 description of, 143-148
audience effects in, 138-140 feedback component, 135, 136-137, 138,
Self-clarification, as motivation for 144, 145, 146, 148, 150-151, 156, 166
revealing secrets, 17 internalization component, 136-141, 144,
Self-concealment, 3, 5-7 147,156-157,166-168
effect on attitudes toward counseling, research about, 132-141
32-33 self-beliefs component, 136-138, 140
correlates of, 31-38 self-change component, 134-136
repression, 5-7 self-concept change and, 134-136
social support, 32-33, 56 self-identification theory as basis for, 131
physical and psychological consequences theoretical background of, 131, 141, 142-
of,209-210 143
genetic factors in, 188, 217 Sexual assault victims: see also Rape
self-report measures of, 24 victims
tendency to, 59-60 exposure therapy for, 82
Self-Concealment Scale, 6, 32-35, 59-60 Sexual orientation
Self-concept concealment of, 35-36
"one true self," 151 revelation of
self-presentation-related change in, 134- effect on health, 168
136,157 to parents, 110-111
Self-control, negative correlation with self- Sexual partners, number of, 215-216
concealment, 34 Sexual secrets: see also Extramarital!
Self-deception, negative correlation with extrarelationship affairs
self-concealment, 34, 39 of college students, 10, 11
Self-disclosure, 3 of families, revelation during family
health benefits of, 42-43 therapy, 109
Self-esteem prevalence of, 21
low of psychotherapy clients, 13
positive correlation with self- revelation to relationship partner, 215-
concealment, 34, 38, 54 216
262 INDEX

Shame Sorrows, as secrets, 11, 15

correlation with secrecy, 54 Spousal privilege, 199
experienced by psychotherapy clients, Spouses, of suicide victims, 210-211
204 Starr, Kenneth, 8
effect of revelation of secrets on, 174 Starr Report, 205
stigmatizing traumatic events-related, Steinbeck, John, 162, 183
174 Stigmatization
Sheffield Psychotherapy Project, 118 associated with revelation of secrets,
Short-term dynamic psychotherapy, 85 161-162
Shyness of homosexuals, 178-179
physical and psychological consequences Stress
of,62-63 secrecy-related, 41-42
positive correlation with self- self-concealment-related, 56
concealment, 23-24, 38, 39 Stress inoculation training, for rape
Significant others, as social support, 179- victims, 120
180 Stress response
Sins, as secrets, 11 cortisol secretion during, 29-30, 31
Sister Prejean, 68, 69 emotional expression as, 179
Skin conductance levels effect on cognitive and emotional
catharsis-related decrease in, 85-86 adjustment, 180-181
emotional inhibition-related increase in, Epstein-Barr virus as indicator of, 30-31,
42,45-46 70-71,75
effect of expression of pain on, 43 effect of repression on, 6-7, 27-28
secret revelation-related decrease in, 169 effect of social support on, 169
self-concealment-related increase in, 42 Sudden infant death, 180
"Small-world" problem, of social Suicide, 97
networks, 200-201 as family secret, 113
Smokers, emotional expression by, 179 by group therapy members, 119
Social anxiety as trauma response, 82
positive correlation with self- Suicide victims, spouses of, 210-211
concealment, 33-34, 38, 61-62 Support groups, for breast cancer patients,
positive correlation with shyness, 63 77
Social networks Supportive counseling, for rape victims,
alienation of, 169 120
interconnectedness of Suppression, 19
estimation of, 189 effect on awareness of information, 19
"small-world" problem of, 200-201 as Freudian term, 103
need for, 169-170 of intrusive thoughts, 47-54, 57
unsupportive or critical, 162 of pain, 43, 46
Social status, relationship with privacy, 172 Symptom prescription, 142-143
Social support
negative correlation with self- Taboo topics
concealment, 32-33 in close relationships, 13-14, 15-16
significant others as, 179-180 as family secrets, 14
effect on stress response, 169 of psychotherapy clients, 15-16
therapists as, 155-156 Talking, about traumatic events, 72-75
Social validation, as motivation for Taylor Manifest Anxiety Scale, 6, 26, 30-31
revealing secrets, 17-18 Temperament types, inhibited and
Solitude, need for, 33-34, 39 noninhibited,57-59
Solution-focused therapy, 146 Terminal illness, as family secret, 111
INDEX 263

Themes, of secrets, vs. details, 203-206, Traumatic experiences (cont.)

207 as source of new insights, 81-85
Therapeutic alliance, supportive, 142-143 writing about, 69-79, 120
Therapist-client relationship, hostility in, 124 as catharsis, 88, 99
Therapists versus catharsis, 91-96
empathy of, 182-183, 206 delayed sense of relief after, 213-214
primary role of, 182-183 health effects of, 88
as social support, 155-156 positive effects of versus talking about
as source of new insights, 90 traumatic events, 72-73, 74-75
Therapy Session Report, 89-90 as source of new insights, 82-85, 88,
Thoughts, intrusive 99
in bereaved mothers, 180 Type A behavior, 28
suppression of, 47-54, 57
writing about, 84 Unfamiliar, fear of, 58
Time-limited therapy, clients' hidden
reactions in, 122-123 Varnado, Mike, 68
T-lymphocyte CD4 and CD8 subsets, effect Victims
of emotional expression on, 71 blaming of, 169
Traumatic experiences others' negative responses to, 168-171
controllability of, 211-212 Violence-related secrets, 11
personal interpretation of, 211
revelation of, 18-19, 21, 69-79, 210-211 Widows/widowers, 36-37, 210-211
as catharsis, 85-86 Workplace, revelation of secrets in, 167-
of previously-disclosed versus 168, 194, 195
undisclosed events, 75-76 Worries, as secrets, 11
to psychotherapists, 72-74 Writing, about traumatic experiences, 69-
of real versus imagined events, 76-77, 79, 120
79 as catharsis, 88, 99
versus revelation of trivial events, 69- delayed sense of relief after, 213-214
72 health effects of, 88
as secrets, 11, 15 versus talking about traumatic events,
through written versus vocal 72-73, 74-75
expressions, 74-75
in undergraduate versus clinical Zamora, Diane, 1-2, 4-5, 21,188
samples, 77-78 Zeignamik effect, 19, 95-96