MEMORY, 2003, 11 (2), 137±149

The identity function of autobiographical memory:

Time is on our side

Anne E. Wilson
Psychology Department, Wilfrid Laurier University, Waterloo, ON, N2L 3C5 Canada.

Michael Ross
University of Waterloo, Canada

Autobiographical memory plays an important role in the construction of personal identity. We review
evidence of the bi-directional link between memory and identity. Individuals' current self-views, beliefs,
and goals influence their recollections and appraisals of former selves. In turn, people's current self-views
are influenced by what they remember about their personal past, as well as how they recall earlier selves
and episodes. People's reconstructed evaluations of memories, their perceived distance from past
experiences, and the point of view of their recollections have implications for how the past affects the
present. We focus on how people's constructions of themselves through time serve the function of creating
a coherentÐand largely favourableÐview of their present selves and circumstances.

``We are what we eat'' is a currently popular autobiographical memory, but the nature and
mantra. More interested in cognition than nutri- strength of the association depends on qualities of
tion, psychologists are likely to assert, ``We are both the self-identity and the memories. More-
what we remember'' (Albert, 1977; Conway & over the relation is reciprocal: People's recollec-
Pleydell-Pearce, 2000; James, 1890/1950). Noting tions influence their self-views and vice versa
the dependence of self-identity on auto- (Figure 1). We describe motives and cognitive
biographical memory, William James (1890/1950) processes that connect self-identity to auto-
remarked that were an individual to awake one biographical memory. We begin by considering
morning with all personal memories erased, he or the influence of current self-views and beliefs on
she would essentially be a different person. Along people's reconstructions of the past. We then
the same lines, Schacter (1996) described a head- describe how people's motives and cognitive
injury patient who lost his autobiographical processes affect their reactions to their pasts, as
memories and, as a result, his associated sense of well as the impact of their recalled pasts on cur-
self. Logically, autobiographical memory plays an rent self-views.
indirect role in even the social sources of self-
knowledge (e.g., reflected appraisals, social com-
parisons; Sedikides & Skowronski, 1995), because CURRENT SELF-VIEWS INFLUENCE
much of this knowledge may be stored in other AUTOBIOGRAPHICAL MEMORY
people's memories of interactions with self. In the
current paper we examine the links between The current selfÐwith its associated character-
autobiographical memory and self-identity. It may istics, goals, and beliefsÐinfluences how indivi-
be a truism to say that self-identity depends on duals recall their pasts (Bartlett, 1932; Fischhoff &

Requests for reprints should be sent to Anne E. Wilson, Psychology Department, Wilfrid Laurier University, Waterloo, ON, N2L
3C5 Canada.

# 2003 Psychology Press Ltd

[Link] DOI:10.1080/09658210244000324
138 WILSON AND ROSS

may not be perfect now but I was much worse

before taking the course.''
At the conclusion of the study skills course,
Conway and Ross asked participants to recall
their pre-programme evaluations of their study
skills. They were told to recall their ratings as
accurately as possible and were reminded that the
researcher had their prior evaluations. Course
participants remembered their ratings as being
worse than they had reported initially. In contrast,
Figure 1. Bi-directional relation between autobiographical
memory and current identity. waiting-list control participants who did not take
the course showed no systematic recall bias over
the same time period. The biased recollections of
individuals who took the course would support a
Beyth, 1975; Greenwald, 1980; Mead, 1929/64; belief that the programme improved their skills.
Ross, 1989; Ross & Buehler, 1994; Singer & Sal- Other research has demonstrated that when
ovey, 1993). Ross (1989) focused on how people people possess a theory of decline (e.g., memory
reconstruct their earlier attributes and feelings. in older adults), they may retrospectively revise
He reasoned that such recollections often involve the past upwards, recalling themselves as better
a two-step process. Because present attributes and than they likely had been (e.g., McFarland, Ross,
feelings are frequently more accessible than past & Giltrow, 1992).
ones, individuals start with current self-appraisals, Although Ross (1989) reported research
such as ``How do I feel about `X' today.'' Next, showing exaggeration of both consistency and
people invoke implicit theories about the stability change, the former tendency seemed much more
of their own attributes and feelings to construct a common. Ross noted developmental evidence
past that is similar to or different from the present. suggesting that people's attributes are quite
Ross (1989) reviewed a large number of studies stable and concluded that perhaps people simply
that provided support for this analysis. For over-generalise an often valid theory of con-
example, several researchers obtained evidence sistency. Alternatively, consistency biases may be
that people reconstruct their past attitude on an common because people are motivated to seek
issue by first considering their present opinion and evidence from the past that implies a constant
then evaluating whether or not they had reason to self-identity through time. James (1890/1950)
suppose that their views had shifted over time. claimed that a sense of personal identity requires
People tend to assume that their attitudes are that one perceive one's self as continuous
stable and therefore infer (sometimes incorrectly) through time. Albert (1977) concurred and sug-
that their prior attitudes are similar to their gested further that people are motivated to
present views. establish a consistent self-identity through time.
According to the implicit theories approach, According to Albert, this consistency motivation
people who believe that their attributes are stable has implications for memory: People are inclined
will tend to construct a past consistent with this to recall pasts that are consonant with their cur-
belief. By the same token, people who expect rent self-views. In other words, autobiographical
change (progress or decline) on a dimension will memory may serve an identity function by
revise the past upward or downward accordingly. enhancing individuals' feelings of personal con-
Conway and Ross (1984) studied participants in a sistency through time.
study skills programme. Participants believe in the Our recent research indicates, however, that
efficacy of such courses, but like most similar skills people do not always appear to value personal
programmes Conway and Ross's was ineffective. consistency; instead they often highlight shifts in
Participants evaluated their study skills before and their identities over time. The tendency to exag-
after participating in the skills programme. gerate personal change seems to be more wide-
Conway and Ross reasoned that participants spread than Ross (1989) supposed. We next
could maintain their belief in the value of an examine the role of autobiographical memory in
ineffective treatment by retrospectively amplify- people's perceptions of a particular kind of change
ing their deficiencies prior to the programme. In in themselves, a tendency to perceive improve-
essence participants could say to themselves: ``I ment over time.
IDENTITY FUNCTION OF AM 139

THE SELF-ENHANCEMENT psychologically remote former selves as no longer

FUNCTION OF associated with their current identityÐdistant
AUTOBIOGRAPHICAL MEMORY failures lose their power to taint and glories to
flatter the present self. As a result, people can
In his autobiography, Arthur Koestler (1961) view distant selves more dispassionately, heaping
remarked that people are critical of their past scorn when it is due.
selves: ``The gauche adolescent, the foolish young Indeed, people might think of past selves as
man that one has been, appears so grotesque in akin to other individuals who vary in closeness to
retrospect and so detached from one's own iden- their current self. Recent selves may be compar-
tity that one automatically treats him with amused able to intimate others and distant selves to mere
derision. It is a callous betrayal, yet one cannot acquaintances or even strangers. People tend to be
help being a traitor to one's past'' (p. 96). Our own blind to the faults of their intimates (Murray,
research findings support Koestler's observation. Holmes, Dolderman, & Griffin, 2000), but judge
Across various samples (e.g., university students, distant acquaintances and strangers more severely
middle-aged individuals, celebrity interviews) and (e.g., Taylor & Koivumaki, 1976).
on a variety of dimensions, people reported their Of course, criticism of others is not always
past selves to be inferior to their present self valid. For example, people sometimes unfairly
(Wilson & Ross, 2000, 2001a). Karney and his derogate others to enhance their own accom-
associates found a similar pattern in people's ret- plishments (e.g., Tesser, 1988; Wills, 1981). Simi-
rospective evaluations of their marriages (Karney larly, a tendency to disparage earlier selves may
& Coombs, 2000; Karney & Frye, 2002). Although reflect concerns for self-enhancement rather than
marital satisfaction tends to decrease over the accuracy (Ross & Wilson, 2000; Wilson & Ross,
early years of a marriage, spouses underestimated 2001a). We suggest that people appraise the past
their past contentment and often recalled it as in ways that allow them to view their current self
lower than their present satisfaction. By depre- favourably. Although dissociated from people's
cating their former satisfaction levels, individuals current self, psychologically distant selves may
create the illusion of improvement even in the still serve an identity function. Conceivably, peo-
face of actual decline. ple systematically devalue their distant former
One explanation of such findings is that people selves to create the illusion that they (or their
are not actually traitors to their past selves, but relationships) have improved over time. People
merely impartial observers of former selves. Per- find an improving trajectory to be particularly
haps people see a past self as it really was, and attractive and gratifying (Carver & Sheier, 1990;
view their present self too favourably. Along these Frijda, 1988; Hsee, Abelson, & Salovey, 1991;
lines, George Orwell (1946) recommended jud- Loewenstein & Prelec, 1993; Loewenstein &
ging the truth of an autobiography by its unflat- Thaler, 1989), partly because they adapt to their
tering content: ``Autobiography is only to be current states and so even consistently favourable
trusted when it reveals something disgraceful. A circumstances become less satisfying over time
man who gives a good account of himself is (Brickman, Coates, & Janoff-Bulman, 1978).
probably lying, since any life when viewed from Indeed, Brickman and Campbell (1971) suggested
the inside is simply a series of defeats'' (p. 170). that ``perhaps the happiest adult is one who had a
Karney and Frye (2002) offered a similar inter- moderately unhappy childhood'' (p. 293), because
pretation of their data on marital satisfaction. current state can best be appreciated in contrast
They suggested that spouses' retrospective eva- with an inferior past. We suggest that it can be just
luations of earlier stages of their marriage are as effective to recall a past as inferior, whether or
more valid than their current assessments. not it actually was.
Social psychologists have provided consider- Examining prospective and retrospective tra-
able evidence that people in Western cultures are jectories of newlyweds' relationship satisfaction,
motivated to view their current self favourably Karney and Frye (2002) showed that the percep-
(Baumeister,, 1998; Higgins, 1996; Sedikides, tion of improvement is linked to other indicants of
1993). It seems reasonable to suppose that people relationship success. Spouses' retrospective
are motivated to enhance current and recent reports of increases in relationship satisfaction
former selves, because those selves and their predicted optimism about the relationship's
associated outcomes ``belong to'' people's present future, even after controlling for any actual
identity. In contrast, individuals may regard more change in satisfaction. In contrast, absolute levels
140 WILSON AND ROSS

of relationship satisfaction were unrelated to people continually boosted their current selves
expectations. By derogating earlier aspects of rather than criticising their earlier selves, their
themselves and their relationships, people can present self-regard might become so inflated as to
make their current state seem superior by com- be highly inconsistent with objective indicators
parison and foster optimism about the future. and difficult to maintain (Baumeister, 1989).
By examining conditions in which self- Moreover, there are advantages to having a rela-
enhancement goals may be particularly strong, we tively accurate appraisal of one's present attri-
obtained more direct evidence that disparaging butes. When confronted with a puddle it is useful
the past benefits the present self (Wilson & Ross, to know how far one can jump. By derogating the
2000; studies 4 & 5). We manipulated people's past, individuals are able to create an illusion of
objectives while they described themselves, improvement without greatly misrepresenting
encouraging participants to adopt the goal of their present strengths and weaknesses.
either evaluating themselves favourably or accu-
rately. Participants with a self-enhancement goal
were more likely to include an inferior past self in APPRAISING SUBJECTIVELY
their self-description than were participants with RECENT VERSUS DISTANT FORMER
an accuracy goal. Self-enhancement motives also SELVES
tend to be exacerbated by a threat to self-regard
(e.g., Wills, 1981;; Wood & Taylor, 1991). McFar- We have proposed that people are more inclined
land and Alvaro (2000) asked individuals who had to criticise distant than close former selves. In our
experienced a personally disturbing or traumatic research we have operationalised closeness in
past event to evaluate what they were like prior to terms of both actual and subjective time. As actual
the episode. Some participants were reminded of time increases, people become more critical of
the disturbing episode before completing the earlier selves. For example, in one study middle-
evaluation and others were not reminded. Parti- aged participants (M age = 50 years) evaluated
cipants who were reminded provided inferior their present selves and retrospectively appraised
evaluations of their earlier, pre-trauma selves. In their former selves at ages 35, 19, and 16 on a host
addition, people were more critical of former of attributes (Wilson & Ross, 2001a). Theoreti-
selves after being reminded of severely rather cally, however, we have proposed that evaluations
than mildly disturbing experiences. This recon- of former selves depend more on the subjective
struction of the past may protect current identity: experience of temporal distance than on the actual
By focusing on how a personally distressing event passage of time. Subjective distance is often rela-
led to growth or positive outcomes for the self, ted to clock or calendar time: Yesterday typically
individuals may minimise the negative impact of feels closer than last month or last year. However,
the trauma. psychologists have long recognised that the sub-
Can people maintain a consistent identity jective experience of time is affected by a variety
(Albert, 1977; James, 1890/1950; Swann & Read, of factors and is sometimes independent of actual
1981; Swann, Stein-Seroussi, & Giesler, 1992) time (e.g., Block, 1989; Brown, Rips, & Shevell,
while still perceiving improvement? Some 1985; James, 1890/1950; Ross & Wilson, 2002). In
researchers have argued that change over time our theory of temporal self-appraisal (Ross &
represents a threat to self-consistency, even Wilson, 2000, 2002; Wilson & Ross, 2001a), we
though it is emotionally gratifying (Keyes & Ryff, reasoned that when people feel close to a past self,
2000). On the contrary, we propose that people its successes and failures psychologically belong to
can forge a personal narrative that explains and the present, regardless of their actual temporal
justifies change. Such narratives allow individuals distance. To test this idea, we manipulated
to view themselves as the same person, despite ``apparent time'' while holding actual time con-
change and improvement (Gergen & Gergen, stant. In one study (Wilson & Ross, 2001a),
1988). university students evaluated their current self
Although individuals can satisfy a desire for and a self of 2 months ago. In the psychologically
self-enhancement by derogating their past, they close condition, participants were asked to ``think
could perhaps achieve the same end by con- of a point in time in the recent past, the beginning
tinually inflating their assessments of present self. of this term. What were you like then?'' In the
There may be psychological advantages, however, psychologically distant condition, participants
to manipulating the past rather than the present. If were instructed to ``Think all the way back to the
IDENTITY FUNCTION OF AM 141

beginning of this term. What were you like way job interview. She could perhaps restore her self-
back then?'' Even though participants were con- regard by erasing her interview from memory, but
sidering the identical time period, this subjective such forgetting may not be possible. The human
distance manipulation affected their recall of for- brain is not a computer disk from which material
mer selves. Those who were induced to regard the can be erased by the tap of a key. Assume instead
time period as recent recalled their former selves that our job applicant is able to distance the
as being just as impressive as they were in the interview, to feel far away from it. By distancing
present, whereas those who were encouraged to the interview, the interviewer can render it less
see the same period as distant were significantly relevant to her current self. The poor performance
more critical of their former than of their current belongs to an earlier and conceivably less able self.
self. Additionally, we reasoned that if criticism Although regarding a negative episode as distant
and praise are motivated by self-enhancement is not the same as forgetting it, the psychological
concerns, then the effects of subjective distance consequences may be comparable. Distancing
should be strongest when participants evaluate helps individuals to put their undesirable beha-
personally significant attributes. Presumably, viour behind them. The behaviour belongs to an
important dimensions have the greatest impact on ``old me''.
overall self-regard (e.g., Crocker & Wolfe, 2001). A double-edged sword, feelings of subjective
As predicted, participants were particularly likely distance have implications for the impact on the
to praise psychologically recent and criticise dis- current self of past achievements as well as fail-
tant former selves when appraising the dimension ures. As a prior success fades into the distance, its
they had nominated as most important to them. value to the current self diminishes. The achieve-
The effect of temporal distance on the appraisal of ment belongs to an earlier self. In temporal self-
past and present selves disappeared when parti- appraisal theory, we hypothesised that individuals
cipants evaluated themselves on their least can mitigate the effects of time by continuing to
important attribute. feel close to an episode. If an outcome feels
recent, the current self can continue to claim
credit for it.
FEELINGS OF SUBJECTIVE Although the hypothesised asymmetry in feel-
DISTANCE: PUSHING THE PAST ings of distance could include divergent estimates
AWAY AND PULLING IT FORWARD of calendar time, this need not be the case. In our
research, we control for actual time and examine
In addition to influencing what people remember feelings of subjective distance for past episodes
about their former selves, self-enhancement goals that could have negative or positive implications
can affect people's subjective judgements of when for the current self. In one study (Ross & Wilson,
episodes occurred. To this point we have discussed 2002), we randomly assigned participants to
subjective distance as an independent variable: remember the course in the previous semester in
Variations in subjective distance alter appraisals of which they received either their best or worst
a past self or episode. In addition, subjective dis- grade. After reporting their grade, participants
tance can operate as a dependent variable. indicated how distant they felt from the target
Although our reconstruction of dates and times can course on a scale with end-points labelled, ``feels
function to organise our autobiographical memory like yesterday'' and ``feels far away''. The results,
into a chronological sequence (e.g., Thompson, shown in Figure 2, evidence the predicted asym-
Skowronski, Larsen, & Betz, 1996), our subjective metry: Participants felt farther from a course in
experience of time does not always correspond to which they obtained a relatively low grade, even
clock or calendar time (e.g., Ross & Wilson, 2002). though the actual passage of time did not differ in
We hypothesised (Ross & Wilson, 2000, 2002) that the two conditions. In subsequent research, we
differences in the evaluative implications of past found that this asymmetry reveals both a tendency
episodes affect people's feelings of the subjective to pull favourable outcomes forward in subjective
distance of those events. To protect their current time and push inauspicious outcomes backward
self-regard, people are motivated to feel farther (Ross & Wilson, 2002), although the latter effect
from past failings than from achievements, even may be somewhat stronger. We also found that
when calendar time does not differ. the asymmetry was obtained for personal out-
For example, suppose that a woman suffers a comes, but not for outcomes of acquaintances.
blow to her self-esteem by performing poorly on a This self±other difference points to the functional
142 WILSON AND ROSS

Figure 2. Subjective distance of good and bad past grades, controlling for calendar time. Higher numbers indicate greater subjective
distance, controlling for actual time. Calendar time since last academic term did not differ by grade condition.

significance of feelings of subjective distance. The recent past (e.g., Age 16 to Today). They were
asymmetry reflects a motivation to protect one's instructed to locate and mark a ``target'' event
own self-regard; there exists no corresponding (e.g., a good or bad past outcome in high school)
inclination to alter the distance of others' experi- on the time line. As Figure 3 illustrates, people
ences. Arguably, shifting the subjective distance would be induced to place a target event (in this
of personal events can satisfy self-enhancement case, their last semester of high school) much
goals without necessarily distorting the date or closer to ``today'' when the time line spanned
other pertinent facts about the episode. many years than when it only included the past
few years. Moreover, this manipulation altered
people's reports of subjective distance from the
AUTOBIOGRAPHICAL MEMORIES target event: They felt psychologically closer to
INFLUENCE CURRENT SELF-VIEWS the events that were spatially closer to the present.
Next, we assessed the impact of feeling close to or
We have reviewed evidence that people's mem- far from past episodes on participants' current
ories of their personal pasts (both what and when) evaluations. Respondents who were induced to
are malleable and may be influenced by current feel close to former failures evaluated their
self-identity and self-motives. Next, we examine current self less favourably than those who were
the other side of the bi-directional relation persuaded to feel distant from the same failures.
between autobiographical memory and self- In contrast, participants encouraged to feel close
identity. Do people's memory revisions and dis- to earlier successes appraised their current self
tancing manoeuvres actually alter the effect of more favourably than those who were persuaded
remembered outcomes on current self-regard? to see the same successes as more remote.
We tested this question directly in several studies In these experiments, the actual temporal dis-
by experimentally varying people's feelings of tance of past events did not differ by condition and
distance from past outcomes that differed in we controlled for actual distance in the statistical
valence (Wilson, 2000; Wilson & Ross, 2001b). We analyses. Although one might expect an equiva-
varied subjective distance by changing repre- lent pattern of results if the events differed in
sentations of the spatial distance between two actual as well as subjective time (e.g., Suh, Diener,
points on a time line. University students were & Fujita, 1996), our studies demonstrate that
presented with a time line that spanned many subjective distance alone can moderate the impact
years (e.g., Birth to Today) or only the fairly of remembered outcomes on current self-regard.
IDENTITY FUNCTION OF AM 143

Figure 3. Time line manipulation of spatial distance. The Birth to Today time line makes high school seem subjectively more recent
than the Age 16 to Today time line.

The remembered past affects people's current feelings of closeness should make happy mem-
views of themselves, but how they remember ories even more pleasurable and distressing events
matters as much as what they remember. The more disturbing.
same event has a different impact, depending on In temporal self-appraisal theory, we focus on
whether it feels near or far. the self-esteem maintenance function of auto-
In the course of everyday life, individuals are biographical memory, highlighting how people
unlikely to encounter ``time lines'' that alter their tend spontaneously to regard negative outcomes
experience of temporal distance. We argue, how- as more distant than positive episodes. People
ever, that there are many real-life experiences that differ in the motivation to self-enhance, however,
affect feelings of subjective distance from past and variations in motivation should predict feel-
episodes. For example, transitions such as chan- ings of subjective distance. Individuals with high
ging jobs, cities, or romantic partners may cause self-esteem are more inclined to engage in cogni-
the pre-transition self to seem especially remote. tive strategies that serve to maintain or enhance
In one study, Wilson and Ross (1998) asked stu- self-regard than are individuals with lower self-
dents who had moved away from their family esteem (Ross & Wilson, 2002). Not surprisingly,
home to attend university to indicate how distant then, people with high self-esteem are more likely
they felt from their 17-year-old (pre-transition) than their low self-esteem counterparts to distance
self. Half of the students were reminded of their unflattering events and to feel close to praise-
move (by answering questions about it) whereas worthy episodes (Ross & Wilson, 2002). There
the remaining participants were not reminded. may also be occasions, either due to individual
Students who were reminded of the move repor- differences or to the nature of an event, when
ted feeling significantly more distant from their individuals cannot help but feel psychologically
high-school self than did the participants who close to threatening past experiences. Holman and
were not reminded. Milestones such as birthdays, Silver (1998) described how some people appear
religious conversions, marriages, and even physi- to become ``stuck in the past'', unable to put
cal or material changes (e.g., getting a new haircut earlier traumas behind them. Similarly, Pillemer,
or car) might represent transitions to some indi- Desrochers, and Ebanks (1998) reported that
viduals, and serve to distance earlier selves. narrators sometimes spontaneously switch to the
Similarly, severely disturbing personal events may present tense when describing emotionally intense
act as transitions that increase feelings of remo- past incidents, perhaps ``reliving'' these earlier
teness from prior selves. Feelings of temporal events whether they want to or not.
distance could thus contribute to the derogation of
pre-trauma selves evidenced in the McFarland
and Alvaro (2000) study. THE PSYCHOLOGICAL
Personal experiences may cause the past to feel IMPORTANCE OF POINT OF VIEW
close as well as distant. For example, revisiting a
childhood haunt or attending a school reunion Nigro and Neisser (1983) reported that individuals
may pull ancient history back into the psycholo- visually recall memories from either a first-person
gical present in much the same way as tasting the or a third-person perspective. When adopting a
madeleine did for Proust (1934). Such enhanced first-person perspective, people perceive
144 WILSON AND ROSS

memories through their own eyes. When assuming disturbing events had health benefits, leading for
a third-person perspective, individuals view their example to fewer visits to physicians during sub-
memories from the vantage point of an observer. sequent months. However, the disclosure±health
The fundamental attribute of a third-person relationship has not been consistently obtained by
memory is that individuals can see themselves in other researchers, who have used methodologies
the recollection. similar to Pennebaker's (e.g., Greenberg & Stone,
Like subjective distance, point of view is a 1992; Murray, Lamnin, & Carver, 1989). Fergus-
variable that relates to how people remember, son (1993) examined whether the benefits of
rather than what they remember. Moreover, like disclosure would be enhanced if participants
subjective distance, point of view is associated wrote in the third rather than the first person. In
with both actual temporal distance and the self- the third-person condition of her study, partici-
concept. Nigro and Neisser found that third-per- pants described their own distressing experiences,
son memories tended to be older and less vivid but they employed a pseudonym and the pronouns
than first-person memories. They also reported he or she. They thus wrote as if the distressing
that memory perspective is malleable: participants events had occurred to someone else. In the first-
were more likely to recall an episode from a first- person condition, participants wrote about their
person perspective when asked to focus on the upsetting experiences using the first-person pro-
emotions associated with an event rather than its noun. In the control conditions, participants wrote
objective circumstances. In an intriguing set of about trivial issues (e.g., their activities during the
studies, Libby and Eibach (2002, Libby, Eibach, & previous day or a description of their living
Gilovich, 2002) recently related the visual per- accommodation) in either the third or first person.
spective of autobiographical memories to the self- Fergusson (1993) hypothesised that writing in
concept. Individuals were more likely to invoke a the third person about negative, personally
first-person perspective when recalling actions meaningful events would be especially beneficial
consistent with their current self-concept. For because it circumvents people's tendencies to
example, participants who were induced to feel avoid dealing with traumatic events and, at the
religious (by means of a biased questionnaire) same time, serves a psychological distancing
were highly likely to recall a religious memory function that permits individuals to reframe,
from a first-person perspective. Participants who ``work through'', and ultimately leave painful
were encouraged to feel irreligious were sig- experiences behind them. The research of Libby
nificantly more likely to report that they viewed a and Eibach on visual perspective suggests further
religious memory from a third-person perspective that third-person narratives may be effective, in
(Libby & Eibach, 2002). In another study (Libby part, because they enable individuals to view
et al., 2002), participants were randomly assigned negative experiences as occurring to a different
to recall the same episode from either a first- self. As a result, the implication of those experi-
person or a third-person perspective. Participants ences for the current self is lessened. There are
who invoked a third-person perspective reported thus a variety of reasons for supposing that third-
that they had changed more since the time of the person writing could yield greater psychological
episode. A third-person perspective seems to and health benefits than first-person writing.
operate as a distancing mechanism, leading indi- Fergusson assessed the effects of writing in a
viduals to perceive that a past self is a different questionnaire completed 4 weeks following the
person from the current self. final writing session. Third-person writers repor-
Point of view can refer more generally to the ted lower levels of distress associated with the
degree to which individuals personalise a memory. events about which they wrote than did first-per-
People can think of a remembered event as son writers. Third-person writers also reported a
occurring to themselves or they can remember the better understanding of the episodes. A sub-
same episode as if it happened to someone else. sequent examination of records obtained from the
Fergusson (1993) varied respondents' point of university health centre revealed additional ben-
view in a study in which university students wrote efits of third-person writing. Third-person writers
about disturbing personal experiences. Previous made significantly fewer illness visits to the health
work by Pennebaker and his associates (e.g., centre than did first-person writers during the 50
Pennebaker & Beall, 1986; Pennebaker, Colder, & days following the writing sessions, and marginally
Sharp, 1990; Pennebaker, Kiecolt-Glaser, & fewer visits than did control participants.
Glaser, 1988a) has indicated that writing about Although the data are somewhat equivocal about
IDENTITY FUNCTION OF AM 145

the value of writing about personally disturbing stances that influence whether direct or contrast
events, they do suggest that third-person writing is effects occur. They proposed that direct effects
more beneficial than first-person writing. arise when respondents: (1) recall how an event
Clinicians working with children have noted happened, (2) recall an event in vivid detail, (3)
the advantage of encouraging their patients to ruminate upon an event, or (4) retrieve a recent
distance themselves psychologically from distres- past episode. Conversely, contrast effects occur
sing experiences (Bettelheim, 1979; Fergusson, when people: (1) recount why an event occurred,
1993; Kalter, Schaeffer, Lesowitz, Alpern, & (2) recall an episode sketchily or briefly, (3) fail to
Pickar, 1988; Mishne, 1993). Lawrence (1990) ruminate upon an episode (or are distracted from
proposed the use of ``third person analysis'' with it), or (4) retrieve an event from the distant past.
adults. She suggested that a patient speaking in Although they focus on different manipulations
the third person is able to adopt ``a more dis- and measures, Clark et al. (1994) describe effects
passionate, detached, retrospective view of him/ that are similar to those obtained for manipula-
herself'' (p. 97). Lawrence claimed that, as a tions of subjective distance and point of view. It
result, third-person analysis yields less guilt and may be that the same basic phenomenon underlies
fewer defensive justifications. The distancing of these various processes: Past episodes directly
events in third-person accounts involves reducing influence current self-appraisals when the epi-
the psychological threat of negative experiences, sodes are ascribed to the present self or seen as
not forgetting or denying their occurrence. Dis- representative of one's current life (Strack et al.,
tancing should thus be distinguished from the 1985). Past outcomes either have no influence or a
dissociation or repression of traumatic episodes, contrasting effect when dissociated from, or seen
which may have very different psychological as unrepresentative of, the present self. One can
implications (e.g., Schacter, 1996; Terr, 1994). accomplish this link or separation between the
The Ross and Wilson (2002) and Libby and past and present by varying perceptions of dis-
Eibach (2002) findings suggest novel ways of tance or perspective, and by altering the qualities
assessing the effectiveness of therapeutic proce- of the memory (e.g., its vividness).
dures. Successful therapy may lead individuals to These various memory processes are likely to
view past disturbing episodes as subjectively dis- be interrelated. Because people typically associate
tant and to adopt a third-person visual perspective a third-person perspective with older memories
when remembering the events. Such effects would (Nigro & Neisser, 1983), individuals who are
provide evidence that individuals have success- induced to adopt a third-person perspective might
fully put disturbing incidents behind them. feel farther from an episode than would indivi-
duals who remember the same event from a first-
person perspective. Similarly, recall of recent
MEMORIES ALTER CURRENT events is typically more vivid and less abstract
AFFECT AND LIFE SATISFACTION than recall of distant events (Semin & Smith,
1999). As a result of this common association,
Memories function to regulate people's emotion people may recall vivid, detailed memories in the
and their satisfaction with different aspects of first person and judge them to be recent (Brown et
their lives. As in research on distancing and point al., 1985). Rumination and rehearsal may serve to
of view, it is not what individuals remember but maintain memory vividness, and thereby induce
how they remember that determines the direction both feelings of nearness and a first-person per-
of influence. Memories often have a direct effect: spective. Many people may be more inclined to
People's moods and reports of life satisfaction revisit and rehearse positive than negative events,
improve when they recall pleasant personal both privately and in conversation with others.
experiences and worsen when they remember This differential rehearsal may cause flattering
distressing personal episodes (e.g., Martin, 1990; memories to be better remembered over time and
Salovey, 1992). However, sometimes a contrast contribute to the discrepancy in the subjective
effect occurs: A pleasant memory depresses and distance of positive and negative events. People
an unpleasant memory boosts people's current who show a heightened tendency to ruminate
mood and life satisfaction (e.g., Nolen-Hoeksema, about unflattering eventsÐfor example, indivi-
1987; Strack, Strack, & Gschneidinger, 1985; duals who are dysphoric (Nolen-Hoeksema &
Tversky & Griffin, 1991; Wilson, 2000). Clark, Morrow, 1991)Ðshould be less likely to distance
Collins, and Henry (1994) summarised circum- negative episodes. In one study of the effects of
146 WILSON AND ROSS

rehearsal, participants who were encouraged to the transgressor responsible for the misconduct
ruminate about a negative event felt subjectively and be less convinced of his or her metamorphosis
closer to it than did those who were distracted (Baumeister et al., 1990; Cameron et al., 2002).
from the event (McLellan, Wilson & Ross, 2002). More generally, claims of personal improvement
In addition, ruminators experienced more that involve distancing past negative performances
unpleasant affect about the event. should often appear unfounded to an observer who
Individuals think about why an event occurs is less motivated to distance the performer's
when they seek to understand an episode (Clark et objectionable actions. Assertions of subjective
al., 1994). People generally devote more conscious distance are especially likely to be challenged to
problem-solving resources as well as unconscious the extent that the claims are at considerable
defensive processes to making sense out of dis- variance with calendar time and imply major or
tressing as opposed to favourable events (Gilbert, improbable personal improvement.
Pinel, Wilson, Blumberg, & Wheatley, 1998; In many cases, rememberers may be quite
Taylor, 1991). In fact if making sense of events aware that their audience does not share their
helps individuals to put episodes behind them perceptions of change. For example, people may
(Silver, Boon & Stones, 1983; Taylor, 1983; firmly believe that elements from their sordid
Weber, Harvey & Stanley, 1987), then people may pasts no longer reflect on the person they cur-
prefer not to make sense of positive events. T. rently are, but may still keep their pasts secret
Wilson (2002) recently reported that people because they are not confident that others will
sometimes favour remaining uncertain or con- agree with their view of a changed self. Movie
fused about positive events. The uncertainty stars have often suppressed their embarrassing (or
keeps the event ``open'', and promotes longer- pornographic) early screen appearances, and
lasting positive affect. President George W. Bush attempted to hide his
arrest for driving under the influence even though
he no longer drinks alcohol and regards himself as
REMEMBERING IS OFTEN A a changed man. Conversely, individuals who have
SOCIAL ACT experienced a religious conversion seem eager to
describe how they have exchanged misguided and
We have argued that one function of auto- evil ways for a good and loving life (Ross &
biographical memory is to maintain a favourable Konrath, in press). Proud and persuaded of their
view of self. However, because autobiographical transformation, these individuals often don't hes-
memory can serve multiple functions, there exists itate to share it with others. They presumably
the potential for different memory goals to conflict. suppose that they can convince others of the
We have focused on the intrapersonal benefits of validity and wonder of their conversion.
recalling the self as continually improving, as well Although we have emphasised the parallels
as regarding failings as remote and glories as close. between distancing and point of view, the impli-
When recalled in conversation, these same mem- cations of the two memory processes for social
ories may also serve interpersonal functions (e.g., remembering are quite different. Point of view is a
Pasupathi, 2001). Just as other types of self- private experience. People typically don't tell
enhancement can have social costs as well as others: ``Hmm I am seeing this memory from a
benefits (e.g., Paulhus, 1998), so too may self- first(third)-person point of view.'' Even if they did
serving remembering. Individuals may encounter communicate their visual perspective, they are
some tension between their goal of achieving a unlikely to arouse the ire of an audience. No one
preferred view of a former self and maintaining else can meaningfully disagree with a remember-
closeness and harmony in relationships with oth- er's point of view. A first-person or third-person
ers. Consider a past conflict between intimate point of view cannot be right or wrong and has no
partners. The transgressor may quickly attribute direct pejorative implications.
the misconduct to an ``old me'' and claim that he or
she has improved since the episode (Baumeister,
Stillwell, & Wotman, 1990; Cameron, Ross & THE UTILITY AND VERITY OF
Holmes, 2002). If the transgressor communicates PERSONAL MEMORIES
this insight to the victim, a new conflict could
ensue: Less motivated to distance the transgres- Neisser (1988) proposed that any act of remem-
sor's behaviour, the victim may continue to hold bering lies on a continuum between ``utility (using
IDENTITY FUNCTION OF AM 147

the past to accomplish some present end) and Cameron, J.J., Ross, M., & Holmes. J.G. (2002). Loving
verity (using memory to recapture what really the one you hurt: Positive effects of recounting a
transgression against an intimate partner. Journal of
happened in the past)'' (p.357). While not denying
Experimental Social Psychology, 38, 307±314.
the importance of ``verity'', we have focused on the Carver C.S., & Scheier, M.F. (1990). Origins and
personal utilityÐor functionÐof memories in the functions of positive and negative affect: A control-
current article. Identity construction is not a pas- process view. Psychological Review, 97, 19±35.
sive process. Individuals actively seek information Clark, L.F., Collins, J.E., & Henry, S.M. (1994). Biasing
effects of retrospective reports on current self-
that helps to confirm their desired self-views.
assessments. In N. Schwarz & S. Sudman (Eds.),
Personal memory plays an important role in Autobiographical memory and the validity of retro-
identity construction because it provides pertinent spective reports (pp. 291±304). New York: Springer-
and plentiful information. Also because the past is Verlag.
ephemeral, there is often little concrete evidence to Conway, M., & Ross, M. (1984). Getting what you want
by revising what you had. Journal of Personality and
contradict individuals' versions of their personal
Social Psychology, 47, 738±748.
histories (although accounts may be disputed when Conway, M.A., & Pleydell-Pearce, C.W. (2000). The
they are publicly shared). People can revise their construction of autobiographical memories in the
appraisals of past selves and events, and shift the self-memory system. Psychological Review, 107,
subjective distance, point of view, or way of 261±288.
Crocker, J., & Wolfe, C.T. (2001). Contingencies of self-
recollecting the episodes. These revisions make it
worth. Psychological Review, 108, 593±623.
possible to use the richness of autobiographical Fergusson, P. (1993). Writing about traumatic events
memories, partly for their verity, but often for their using the third person pronoun: Psychological and
utility in creating a preferred representation of self. health effects. Unpublished doctoral dissertation,
University of Waterloo, Ontario, Canada.
Fischhoff, B., & Beyth, R. (1975). ``I knew it would
REFERENCES happen'': Remembered probabilities of once-future
things. Organizational and Human Performance, 13,
Albert, S. (1977). Temporal comparison theory. Psy- 1±16.
chological Review, 84, 485±503. Frijda, N.H. (1988). The laws of emotion. American
Bartlett, F.C. (1932). Remembering: A study in experi- Psychologist, 43, 349±358.
mental and social psychology. London: Cambridge Gergen, K.J., & Gergen, M.M. (1988). Narrative and
University Press. the self as relationship. In L. Berkowitz (Ed.),
Baumeister, R.F. (1989). The optimal margin of illusion. Advances in experimental social psychology, Vol. 21:
Journal of Social and Clinical Psychology, 8, Social psychological studies of the self: Perspectives
176±189. and programs (pp. 17±56). San Diego, CA:
Baumeister, R.F. (1998). The self. In D.T. Gilbert, S.T. Academic Press, Inc.
Fiske, and G. Lindzey (Eds.), Handbook of social Gilbert, D.T., Pinel, E.C., Wilson, T.D., Blumberg, S.J.,
psychology (4th edn., pp. 680±740). New York: & Wheatley, T.P. (1998). Immune neglect: A source
McGraw-Hill. of durability bias in affective forcasting. Journal of
Baumeister, R., Stillwell, A., & Wotman, S.R. (1990). Personality & Social Psychology, 75, 617±638.
Victim and perpetrator accounts of interpersonal Greenberg, M.A., & Stone, A.A. (1992). Emotional
conflict: Autobiographical narratives about anger. disclosure about traumas and its relation to health:
Journal of Personality and Social Psychology, 59, Effects of previous disclosure and trauma severity.
994±1005. Journal of Personality & Social Psychology, 63,
Bettelheim, B. (1979). Surviving, and other essays. New 75±84.
York: Knopf. Greenwald, A.G. (1980). The totalitarian ego: Fabri-
Block, R.A. (1989). Experiencing and remembering cation and revision of personal history. American
time: Affordances, context, and cognition. In I. Levin Psychologist, 35, 603±618.
& D. Zakay (Eds.), Advances in psychology: Vol. 59. Higgins, E.T. (1996). Self digest: Self-knowledge serving
Time and human cognition: A life-span perspective self-regulatory functions. Journal of Personality and
(pp. 333±363). Amsterdam: North-Holland. Social Psychology, 71, 1062±1083.
Brickman, P., & Campbell, D.T. (1971). Hedonic rela- Holman, E.A., & Silver, R.C. (1998). Getting ``stuck'' in
tivism and planning the good society. In M.H. the past: Temporal orientation and coping with
Appley (Ed.), Adaptation-level theory (pp. 278±302). trauma. Journal of Personality and Social Psychol-
New York: Academic Press. ogy, 75, 1146±1163.
Brickman, P., Coates, D., & Janoff-Bulman, R. (1978). Hsee, C.K., Abelson, R.P., & Salovey, P. (1991). The
Lottery winners and accident victims: Is happiness relative weighting of position and velocity in satis-
relative? Journal of Personality and Social Psychol- faction. Psychological Science, 2, 263±266.
ogy, 36, 917±927. James, W. (1950). Principles of psychology. New York:
Brown, N.R., Rips, L.J., & Shevell, S.K. (1985). The Dover. [Originally published in 1890.]
subjective dates of natural events in very long-term Kalter, N., Schaefer, M., Lesowitz, M., Alpern, D., &
memory. Cognitive Psychology, 17, 139±177. Pickar, J. (1988). School-based support groups for
148 WILSON AND ROSS

children of divorce: A model of brief intervention. In views of each other. Journal of Experimental Social
B.H. Gottlieb (Ed.), Marshaling social support: Psychology, 36, 600±620.
Formats, processes, and effects (pp. 165±185). Neisser, U. (1988). What is ordinary memory the
Thousand Oaks, CA: Sage Publications, Inc. memory of? In U. Neisser & E. Winograd (Eds.),
Karney, B.R., & Coombs, R.H. (2000). Memory bias in Remembering reconsidered: Ecological and tradi-
long-term close relationships: Consistency or tional approaches to the study of memory. Emory
improvement? Personality and Social Psychology symposia in cognition, 2 (pp. 356±373). New York:
Bulletin, 26, 959±970. Cambridge University Press.
Karney, B.R., & Frye, N.E. (2002). ``But we've been Nigro, G., & Neisser, U. (1983). Point of view in
getting better lately'': Comparing prospective and personal memories. Cognitive Psychology, 15,
retrospective views of relationship development. 467±482.
Journal of Personality and Social Psychology, 82, Nolen-Hoeksema, S. (1987). Sex differences in unipolar
222±238. depression: Evidence and theory. Psychological
Keyes, C.L.M., & Ryff, C.D. (2000). Subjective change Bulletin, 101, 259±282.
and mental health: A self-concept theory. Social Nolen-Hoeksema, S., & Morrow, J. (1991). A
Psychology Quarterly, 63, 264±279. prospective study of depression and posttraumatic
Koestler, A. (1961). Arrow in the blue, an auto- stress symptoms after a natural disaster: The 1989
biography. New York: Macmillan. Loma Prieta earthquake. Journal of Personality and
Lawrence, L. (1990). On the theory and application of Social Psychology, 61, 115±121.
third person analysis in the practice of psychother- Orwell, G. (1946). Benefit of clergy: Some notes on
apy. The Journal of Mind and Behavior, 11, 97±104. Salvador Dali. In G. Orwell (Ed.), Dickens, Dali &
Libby, L.K., & Eibach, R.P. (2002). Looking back in others: Studies in popular culture. Cornwall, NY: The
time: Self-concept change affects visual perspective Cornwall Press.
in autobiographical memory. Journal of Personality Pasupathi, M. (2001). The social construction of the
and Social Psychology, 82, 167±179. personal past and its implications for adult devel-
Libby, L.K., Eibach, R.P., & Gilovich, T. (2002, Feb- opment. Psychological Bulletin, 127, 651±672.
ruary). It depends on how you look at it: Memory Paulhus, D.L. (1998). Interpersonal and intrapsychic
perspective affects perceived and actual change in the adaptiveness of trait self-enhancement: A mixed
self. Poster presented at the 3rd Annual Meeting of blessing? Journal of Personality & Social Psychol-
the Society for Personality and Social Psychology, ogy, 74, 1197±1208.
Savannah, GA. Pennebaker, J.W., & Beall, S.K. (1986). Confronting a
Loewenstein, G., & Prelec, D. (1993). Preferences over traumatic event: Toward an understanding of inhi-
outcome sequences. Psychological Review, 100, bition and disease. Journal of Abnormal Psychology,
91±108. 95, 274±281.
Loewenstein, G., & Thaler, R.H. (1989). Intertemporal Pennebaker, J.W., Colder, M., & Sharp, L.K. (1990).
choice. Journal of Economic Perspectives, 3, Accelerating the coping process. Journal of Person-
181±193. ality & Social Psychology, 58, 528±537.
Martin, M. (1990). On the induction of mood. Clinical Pennebaker, J.W., Kiecolt-Glaser, J.K., & Glaser, R.
Psychology Review, 10, 669±697. (1988). Disclosure of traumas and immune function:
McFarland, C., & Alvaro, C. (2000). The impact of Health implications for psychotherapy. Journal of
motivation on temporal comparisons: Coping with Consulting & Clinical Psychology, 56, 239±245.
traumatic events by perceiving personal growth. Pillemer, D.B., Desrochers, A.B., & Ebanks, C.M.
Journal of Personality and Social Psychology, 79, (1998). Remembering the past in the present: Verb
327±343. tense shifts in autobiographical memory narratives.
McFarland, C., Ross, M., & Giltrow, M. (1992). Biased In C.P. Thompson, D.J. Herrmann, D. Bruce, J.D.
recollections in older adults: The role of implicit Read, D.G. Payne, & M.P. Toglia (Eds.), Auto-
theories of aging. Journal of Personality and Social biographical memory: Theoretical and applied
Psychology, 62, 837±850. perspectives. (pp. 145±162). Mahwah, NJ: Lawrence
McLellan, L., Wilson, A.E., & Ross, M. (2002). To dwell Erlbaum Associates Inc.
or distract: The effects of rumination versus distraction Proust, M. (1934). Remembrance of things past. New
on subjective distance from negative events. Unpub- York: Random House.
lished thesis, Wilfrid Laurier University, Canada. Ross, M. (1989). The relation of implicit theories to the
Mead, G.H. (1929/1964). Selected writings. Indianapolis: construction of personal histories. Psychological
Bobbs-Merrill. Review, 96, 341±357.
Mishne, J.M. (1993). The evolution and application of Ross M., & Buehler, R. (1994). Creative remembering.
clinical theory: Perspectives from four psychologies. In U. Neisser & R. Fivush (Eds.), The remembering
New York: The Free Press. self (pp. 205±235). New York: Cambridge University
Murray, E.J., Lamnin, A.D., & Carver, C.S. (1989) Press.
Emotional expression in written essays and psy- Ross, M., & Konrath, S. H. (in press). Synergies. Psy-
chotherapy. Journal of Social & Clinical Psychology, chological Inquiry.
8, 414±429. Ross, M., & Wilson, A.E. (2000). Constructing and
Murray, S.L., Holmes, J.G., Dolderman, D., & Griffin, appraising past selves. In D.L. Schacter & E. Scarry
D.W. (2000). What the motivated mind sees: Com- (Eds.), Memory, brain, and belief (pp. 231±258).
paring friends' perspectives to married partners' Cambridge, MA: Harvard University Press.
IDENTITY FUNCTION OF AM 149

Ross, M., & Wilson, A.E. (2002). It feels like yesterday: Tesser, A. (1988). Toward a self-evaluation main-
Self-esteem, valence of personal past experiences, tenance model of social behavior. In L. Berkowitz
and judgments of subjective distance. Journal of (Ed.), Advances in experimental social psychology,
Personality and Social Psychology, 82, 792±803. Vol. 21 (pp. 181±227). New York: Academic Press.
Salovey, P. (1992). Mood-induced self-focused atten- Thompson, C.P., Skowronski, J.J., Larsen, S.F., & Betz,
tion. Journal of Personality & Social Psychology, 62, A. (1996). Autobiographical memory: Remembering
699±707. what and remembering when. Mahwah, NJ:
Schacter, D.L. (1996). Searching for memory: The brain, Lawrence Erlbaum Associates Inc.
the mind, and the past. New York: Basicbooks, Inc. Tversky, A., & Griffin, D. (1991). On the dynamics of
Sedikides, C. (1993). Assessment, enhancement, and hedonic experience: Endowment and contrast in
verification determinants of the self-evaluation judgments of well-being. In F. Strack, N. Schwarz, &
process. Journal of Personality and Social Psychol- M. Argyle (Eds.), Subjective well-being (pp.
ogy, 65, 317±338. 108±118). London: Pergamon Press.
Sedikides, C., & Skowronski, J.J. (1995). On the sources Weber, A.L., Harvey, J.H., & Stanley, M.A. (1987). The
of self-knowledge: The perceived primacy of self- nature and motivations of accounts for failed rela-
reflection. Journal of Social and Clinical Psychology, tionships. In R. Burnett & P. McGhee (Eds.),
14, 244±270. Accounting for relationships: Explanation, repre-
Semin, G.R., & Smith, E.R. (1999). Revisiting the past sentation and knowledge (pp. 114±133). New York:
and back to the future: Memory systems and the Methuen.
linguistic representation of social events. Journal of Wills, T.A. (1981). Downward comparison principles in
Personality & Social Psychology, 76, 877±892. social psychology. Psychological Bulletin, 90,
Silver, R.L., Boon, C., & Stones, M.H. (1983). Searching 245±271.
for meaning in misfortune: Making sense of incest. Wilson, A.E. (2000). How do people's perceptions of
Journal of Social Issues, 39, 81±101. their former selves affect their current self-appraisals?
Singer, J.A., & Salovey, P. (1993). The remembered self: Unpublished doctoral dissertation, University of
Emotion and memory in personality. Toronto: Waterloo, Canada.
Maxwell Macmillan International. Wilson, A.E., & Ross, M. (1998). [Life transitions and
Strack, F., Strack, N., & Gschneidinger, E. (1985). subjective distance.] Unpublished raw data.
Happiness and reminiscing: The role of time per- Wilson, A.E., & Ross, M. (2000). The frequency of
spective, affect, and mode of thinking. Journal of temporal-self and social comparisons in people's
Personality and Social Psychology, 49, 1460±1469. personal appraisals. Journal of Personality and
Suh, E., Diener, E., & Fujita F. (1996). Events and Social Psychology, 78, 928±942.
subjective well-being: Only recent events matter. Wilson, A.E., & Ross, M. (2001a). From chump to
Journal of personality and Social Psychology, 70, champ: People's appraisals of their earlier and
1091±1102. current selves. Journal of Personality and Social
Swann, W.B., Read, S.J. (1981). Acquiring self-knowl- Psychology, 80, 572±584.
edge: The search for feedback that fits. Journal of Wilson, A.E., & Ross, M. (2001b, February). How do
Personality and Social Psychology, 41, 1119±1128. perceptions of former selves affect current self-
Swann, W.B., Stein-Seroussi, A., & Giesler, R.B. (1992). appraisals? Paper presented at the Society for
Why people self-verify. Journal of Personality & Personality and Social Psychology Convention, San
Social Psychology, 62, 392±401. Antonio, TX
Taylor, S.E. (1983). Adjustment to threatening events: Wilson, T. (2002). Self knowledge and the pleasures of
A theory of cognitive adaptation. American Psy- uncertainty. Keynote address at the 3rd Annual
chologist, 38, 1161±1173. Meeting of the Society for Personality and Social
Taylor, S.E. (1991). Asymmetrical effects of positive Psychology, Savannah, GA.
and negative events: The mobilization-minimization Wood, J.V., & Taylor, K.L. (1991). Serving self-relevant
hypothesis. Psychological Bulletin, 110, 67±85. goals through social comparison. In J. Suls & T.A.
Taylor, S.E., & Koivumaki, J.H. (1976). The perception Wills, (Eds.), Social comparison: Contemporary
of self and others: Acquaintanceship, affect, and theory and research (pp. 23±49). Hillsdale, NJ:
actor-observer differences. Journal of Personality Lawrence Erlbaum Associates Inc.
and Social Psychology, 33, 403±408.
Terr, L.C. (1994). Unchained memories. New York:
Basic Books.