Toward A Neuroscience of Attachment

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/230669685

Toward a Neuroscience of Attachment

Chapter · August 2008
CITATIONS READS
166 10,863
1 author:
James Coan
University of Virginia
135 PUBLICATIONS 8,751 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Neural Mechanisms involved in Emotion, Depression, and/or Anxiety View project
Takeover Transitions in Automated Vehicles(Safe-SCAD) View project
All content following this page was uploaded by James Coan on 22 May 2014.
The user has requested enhancement of the downloaded file.

T O WA R D A
NEUROSCIENCE OF
AT TA C H M E N T
Ja me s A . C o a n
FROM THE HANDBOOK OF ATTACHMENT:
THEORY, RESEARCH, AND CLINICAL IMPLICATIONS

JUDE CASSIDY AND PHILIP R. SHAVER, EDS.
2nd Edition
Pages 241 - 265
The Guilford Press, NY
U n i v e r s i t y o f Vi r g i n i a , D e p a r t m e n t o f P s y c h o l o g y, 1 0 2 G i l m e r H a l l P O B o x 4 0 0 4 0 0 , C h a r l o t t e s v i l l e , VA 2 2 9 0 4
• t e l e p h o n e : 4 3 4 - 2 4 3 - 2 3 2 2 • e m a i l : j c o a n @ v i r g i n i a . e d u • w w w. s o c i a l b a s e l i n e . c o m
Toward a Neuroscience of Attachment
James A Coan
Neurobiological studies of attachment are facts (and a cursory glance at the table of con-
either abundant or scarce, depending on one’s tents for this volume) underscore the complex-
research tradition and scientific understanding ity of attachment as a domain of inquiry, and
of the term “attachment.” On the one hand, suggest that, at present, any neuroscience of
the past two decades have seen a great deal of attachment is likely to strike some as limited in
nonhuman animal work detailing the various both empirical foundation and theoretical
neural manifestations of social bonding, famili- scope.
arity, affiliation, caregiving, and other behaviors
that can (and often do) fall under the general Nevertheless, it is important to make a begin-
rubric of “attachment.” On the other hand, ning somewhere, and a neuroscience of at-
neuroscientific investigations of normative attachment has much to gain from the integra-
tachment in humans have been limited and tion of multiple research perspectives. Follow-
slow to develop, and similar investigations of ing Bowlby (1969/1982) and Ainsworth (1989),
the neural circuits supporting, or even associ- attachment bonds are considered in the present
ated with, individual differences in attachment chapter to be those characterized by a high fre-
(e.g., secure, anxious, avoidant, in the social quency of close proximity to the putative “at-
psychology tradition; autonomous, preoccu- tachment figure,” especially during times of
pied, and dismissing, in the clinical and devel- emotional stress. Moreover, attachment rela-
opmental tradition; see Crowell & Fraley, Chap- tionships are considered in this chapter to serve
ter 26, this volume) are exceedingly rare. These regulatory functions, often in relation to basic
C o a n! To w a r d a N e u r o s c i e n c e o f A t t a c h m e n t
1
physiological needs, but also with respect to Theorists have long argued that social bonding
many forms of emotional responding. These ser ves security-provision and distress-
regulatory functions are social insofar as they alleviation regulatory functions with respect to
result from interaction with conspecifics (other negative affect and arousal (Bowlby, 1973;
members of the same species). Some of the Mikulincer, Shaver, & Pereg, 2003). Prominent
regulatory functions of attachment relation- evolutionary theorists dating to Darwin have
ships are obvious and fundamental. For exam- even argued that because mammalian emo-
ple, human infants literally cannot survive tional responding evolved in a social context,
without the assistance of an adult caregiver. In emotional behavior is virtually inextricable
later childhood, however, and in adult attach- from social behavior (Brewer & Caporeal, 1990;
ment relationships, emotion becomes the pri- Buss & Kenrick, 1998; Darwin, 1872/1998).
mary target of social regulation (Mikulincer & These diverse perspectives and literatures sug-
Shaver, Chapter 23, this volume). A major gest that any robust conception of attachment
source of interest here is that the likely mecha- will include multiple, distributed subsystems
nism underlying the well-known link between including (but probably not limited to) those
social contact and health is the social regulation devoted to emotion, motivation, emotion regu-
of emotion, particularly the social regulation of lation, and social affiliation.
threat responding. The social regulation of
threat responding is itself a major feature of The promise of the emerging field of what we
attachment (Carter & DeVries, 1999; Edens, can here consider to be “attachment neurosci-
Larkin, & Abel, 1992; Hofer, 1995). ence” is at once to provide critical information
about how the brain supports attachment be-
A large literature now suggests that a range of haviors and to forge links between research
interactive social behaviors target physiological traditions as diverse as the basic neurosciences,
systems, temperamental dispositions, and overt behavioral ecology, and various subdomains of
behaviors associated with the stress response psychology such as developmental, social, and
(Berscheid, 2003; Diamond, 2001; Sapolsky, clinical, as well as affective science. In this
1998; Uchino, Cacioppo, & Kiecolt-Glaser, chapter, the neural systems supporting emo-
1996). For example, supportive social behaviors tion, motivation, emotion regulation, and social
are known to attenuate stress-related activity in behavior are first reviewed. Following this, the
the autonomic nervous system (ANS) and the social regulation of emotion and individual dif-
hypothalamic-pituitary-adrenal (HPA) axis ference in attachment behavior will be consid-
(Boccia, Reite, & Laudenslager, 1989; Flinn & ered from the perspective of behavioral neuro-
England, 1997; Lewis & Ramsay, 1999; Weiss, science. Based on these reviews, the social base-
1990; Wiedenmayer, Magarinos, McEwen, & line model of social affect regulation will be
Barr, 2003). Maternal grooming behaviors af- proposed. The social baseline model uses a
fect glucocorticoid receptor gene expression neuroscientific framework to integrate models
underlying hippocampal and HPA-axis stress of attachment with a neuroscientific principle,
reactivity in rat pups (Weaver et al., 2004). In economy of action, in the management of
the context of a novel, mildly stressful new en- metabolic resources devoted to emotional and
vironment, rats in the company of a familiar social behavior. Finally, recommendations are
companion engage in more exploration and made for the development of a robust future
play-soliciting behavior compared to rats in the neuroscience of attachment.
company of an unfamiliar companion (Terra-
nova, Cirulli, & Laviola, 1999).
2
Attachment as a Neural Construct tionally, by the presence of separation distress
or physiological soothing as a function of close
Although attachment bonds are widely believed proximity, or both. By contrast, social, clinical,
to result from a universal, innate “attachment and developmental psychologists often focus
behavioral system,” attempts to locate a single, their efforts on “behavioral systems,” seeking
dedicated attachment circuit are likely to be, to to understand how humans behave in and, im-
paraphrase Wittgenstein, a bit like trying to find portantly, what they have to say about, rela-
the real artichoke by peeling away all its leaves. tional contexts.
Almost any interpretation of the attachment
This is not to say that research on attachment
behavioral system reveals it to be a higher order
in humans has not utilized physiological meas-
construct comprised of constituent behaviors
urement. On the contrary, psychologists have
about which a great deal is known, even at the
used measures of autonomic physiology, elec-
neural level (Fox & Hane, Chapter 10, this vol-
troencephalography (EEG), glucocorticoid lev-
ume; Polan & Hofer, Chapter 7, this volume).
els, and, more recently, functional magnetic
For example, many studies have addressed the
resonance imaging (fMRI). These measures
neurobiology of social behaviors such as rec-
have provided valuable insights into human
ognition and familiarity, proximity seeking,
social behavior, but they are rarely capable of
separation distress, soothing behaviors, and ma-
identifying causal brain- behavior relationships
ternal caregiving. Thus, one of the goals of this
(Norris, Coan, & Johnstone, 2007), and their
chapter is to introduce the neuroscientific study
frequent dependence on self-report measures
of attachment from the perspective of what is
(including coded interviews) may result in neu-
currently known about its social and emotional
robiological correlates that are quite distinct
constituents.
from those of behaviorally defined animal
A corollary goal is to move toward bridging models (cf. Williamson, 2006).
two broad, rigorous, productive, and unfortu-
Yet another difficulty presents itself in bridging
nately disparate literatures. One is a thriving
these literatures. Even if the definitions of at-
animal literature dedicated to what is variously
tachment were perfectly matched and each neu-
termed “social bonding,” “pair bonding,” and
ral measure applied to humans and non- human
“attachment bonding.” The other contains a
animals were identical, the neural processes as-
vast body of research on human attachment
sociated with attachment behaviors in non-
behavior, including studies of individual differ-
human animals may not generalize perfectly to
ences in internal working models of attachment
those in humans. Work on the social communi-
(reviewed in Mikulincer & Shaver, 2007, and in
cation value of pheromones provides an excel-
J. Feeney, Chapter 21, this volume). Tradition-
lent example of this point. Pheromones are
ally, these two worlds have had little to say to
chemical substances that convey information
each other—a reflection of their starkly differ-
between members of the same species (Insel &
ent research strategies as much as their differ-
Fernald, 2004). It is certain that nearly all ani-
ent subject populations. Animal models, partly
mals, including humans, show at least some
by virtue of what is ethically permissible with
evidence of two distinct olfactory systems. The
the population, often emphasize the study of
primary olfactory system is dedicated to the
social processes in terms of specific causal neu-
detection of odors that convey information
ral structures, circuits, neurotransmitters, neu-
about food or the presence or predators, and
ropeptides, pheromones, or hormones. At-
this system is most commonly associated with
tachment relationships are defined observa-
the sense of smell. By contrast, the accessory
3
olfactory system is, in many species, dedicated mation about the neurobiology of attachment,
to the detection of specific pheromonal infor- without which any understanding of human
mation. This accessory olfactory system con- attachment would, at the neural level, be se-
sists of the vomeronasal organ (VNO) and the verely impoverished. Moreover, advanced neu-
accessory olfactory bulb (AOB). Pheromones roimaging techniques such as high density
make contact with the VNO, exciting EEG, positron emission tomography (PET),
pheromone-specific sensory neurons projecting transcranial magnetic stimulation (TMS), and
to the AOB. functional magnetic resonance imaging (fMRI)
promise access to human neural processes at a
In a wide variety of species, this system is ca- level of detail undreamed of until the very end
pable of providing rapid and powerful infor- of the 20th century. Hence the potential for
mation about sex, reproductive capacity, mate building bridges between the animal and hu-
location, territorial boundaries, and even social man attachment literatures is higher than it has
status (Insel & Fernald, 2004). Nevertheless, ever been. FMRI studies in particular are, by
the strongest of these findings derive exclu- virtue of their rapid proliferation and relative
sively from studies of animal populations, and lack of invasiveness, beginning to supply pieces
after a great deal of initial excitement about the of the human social bonding puzzle that will
possibility of a human pheromone system, en- compliment anatomical and molecular work in
thusiasm has waned significantly amid evidence animals. Such advances promise the formation
that, although there does appear to be a human of a more comprehensive neuroscience of at-
VNO, (a) there is no obvious pheromone- tachment.
specific sensory neuron associated with it; (b)
vomeronasal receptor genes present in the hu- The Neural Constituents of
man genome appear to be pseudogenes (genes Attachment
that have lost their protein-coding ability); and
(c) the AOB does not appear to exist at all in Neural systems supporting attachment are
the brains of adult humans (Meredith, 2001). likely to include, at a minimum, those underly-
In other words, the VNO—the primary and ing incentive motivation, certain forms of emo-
best-understood mechanism of socially critical tional responding, emotion regulation, and dis-
pheromonal communication in animals—ap- crete social behaviors such as the establishment
pears to be vestigial in humans. of familiarity and preference, proximity seek-
ing, separation distress, and social affect regula-
Interestingly, evidence does suggest that chemi- tion. This chapter is not intended to provide an
cal communication between humans can occur exhaustive treatment of all possible constituent
(e.g., Jacob & McClintock, 2000). However, un- systems underlying attachment. In truth, be-
like so many social species, the extent to which cause so many neural structures are involved
such effects are pheromonal, and whether they one way or another in attachment behavior, it is
have anything whatever to do with the VNO, is possible to think of the entire human brain as a
uncertain at best. It is more likely that odors neural attachment system. Auditory, olfactory,
can moderate social information in humans, and visual sensory systems are heavily impli-
and that they do so through a distinct mecha- cated for obvious reasons. Memory processes
nism that is as yet poorly defined and under- involving, for example, long-term memory
stood (Meredith, 2001). consolidation and retrieval in the hippocampus,
underlie familiarity, recognition, and the main-
Despite all of these cautions, it is clear that re-
tenance of shared histories. A wide variety of
search on animals has yielded invaluable infor-
4
regulatory needs affected by attachment rela- information, or with more evolutionarily
tionships are likely to be related to activity in “primitive” brain structures, working “up” to
the hypothalamus. Conflict monitoring de- more integrative and evolutionarily modern
mands will be made on the anterior cingulate areas such as the cortex. The process of receiv-
cortex (ACC). Each of these systems and more ing sensory inputs from the environment and
contribute to attachment in a variety of ways. converting those inputs into neural pulses that
In this chapter, however, a smaller number of are relayed to cortical structures as consciously
putatively basic elements will be reviewed. perceived information about one’s surround-
ings would be an example of this. Top-down
Behavioral versus neural systems. I should first dis- processes are essentially the opposite. In this
tinguish between what ethologists have long case, integrative and evolutionarily “new” struc-
referred to as “behavioral systems” and what tures pass neural information “down” to more
neuroscientists refer to as “neural systems.” In sensory-oriented and evolutionarily old struc-
ethology, a behavioral system is a set of behav- tures, often to suit some regulatory purpose.
iors associated with a common causal antece- One example of a top-down process might be
dent and resulting, once activated, in a common the brain’s tendency to impute information
consequent, which in turn deactivates the sys- from memory and experience into stimuli in
tem. Drawing on an ethological approach, the periphery of the visual field, thereby im-
Bowlby (1969/1982) described several behav- posing “best guesses” on visual information
ioral systems associated with attachment. When that is ambiguous.
discussing behavioral systems such as these,
there is a great temptation to view the behav- Emotional and Motivational
ioral system as having a one-to-one relationship Elements
with some underlying neural system. But such
tidy correspondences are rare. The term “neu- Incentive motivation, reward, and the dopamine system.
ral system” describes coordinated neural inputs Incentive motivation involves the acquisition of
and signaling targets among a population of rewarding stimuli. The intensity of incentive
neurons that form a circuit. Neural systems can motivation varies as a function of the state of
be tightly organized in close physical proximity the individual and the magnitude of the reward.
or distributed throughout the brain. Highly For example, if a typical Westerner is mildly
similar or even identical behaviors may, across hungry and is offered a kind of food that is
individuals, result from different combinations normally undesirable to him or her—uni (raw
of activity in dissimilar neural systems. Moreo- sea urchin), for example—there will be little
ver, similar neural activations can result in quite incentive motivation to eat the food. If the in-
distinct behaviors. Thus, the search for specific dividual is extremely hungry, however, the in-
neural circuits associated with time-honored, centive motivation to eat the uni will be high.
observationally defined behavioral systems is Similarly, if the same individual is again only
fraught with theoretical and empirical difficulty. mildly hungry, but is given a food item that is
highly desirable—say a piece of chocolate
Bottom-up versus top-down processing. Although the cake—the incentive motivation to eat the cake
terms “bottom-up” and “top-down” processing will be high.
are frequently used in the cognitive neurosci-
ences (and throughout the remainder of this Incentive motivation plays a key role in a num-
chapter), their meanings may not be immedi- ber of attachment-related processes (e.g., prox-
ately obvious. Bottom- up processes are imity seeking) and is tightly linked to the do-
thought to begin, more or less, with sensory pamine projection system of the ventral teg-
5
mental area (VTA). Dopamine is produced in to desirable unconditioned stimuli and dopa-
the VTA and substantia nigra and projected to minergic activity in the VTA increase the pre-
as many as 30 distinct networks (Le Moal & dictability of those unconditioned stimuli, and,
Simon, 1991). It has long been held that dopa- hence, the opportunities for obtaining them
minergic activity represents a neural substrate (Depue & Collins, 1999).
for the facilitation of goal- directed behavior
(Berridge, 2007; Depue & Collins, 1999). The amygdala and hippocampus in affect and memory.
Strongly implicated in this function is the nu- The amygdala is now one of the most widely
cleus accumbens, which is a major terminal area recognized brain structures associated with
of dopaminergic projections from the VTA emotion (Phelps & LeDoux, 2005). Far from a
(Tzschentke & Schmidt, 2000). Dopaminergic unitary structure, the amygdala contains many
activity within the VTA and nucleus accumbens sub-nuclei, accounting for its involvement in a
has been repeatedly associated with reinforcing vast array of emotional responses. A large body
stimuli and the experience of pleasure. For ex- of research now supports the notion that the
ample, rats capable of directly stimulating these amygdala is sensitive to both conditioned and
circuits with a lever press will repeatedly do so, unconditioned signs of threat. Moreover, at
even in lieu of access to food, water, and sex. least two pathways to amygdala activation asso-
This preference for lever pressing over food ciated with visual stimuli exist, both of which
and water will continue even to the point of can mediate fear learning. One is a very rapid
death (Bozarth & Wise, 1996). and direct route through the thalamus (the
thalamo-amygdala pathway) that processes ob-
Dopaminergic cells in the VTA are also highly vious or highly specific sensory information
responsive to conditioning (Depue & Collins, (e.g., the shape of a snake, Le Doux, 2000;
1999), especially to cues that predict the receipt Öhman, 2005). Another pathway processes
of reward (Schultz, Dayan, & Montague, 1997). slower and more complex information in the
Importantly, the VTA is also responsive to visual cortex before activating the amygdala.
stimuli that are unconditioned. Unconditioned When paired with unconditioned aversive stim-
stimuli are those that naturally, automatically, uli (e.g., a loud noise, pain), otherwise meaning-
and unconditionally trigger a response in an less stimuli quickly come to be associated with
organism. Positive unconditioned stimuli act as the presence of a threat, and this conditioning
reinforcers, and include certain flavors, water, appears to be dependent to a large degree on
sleep, touch, and the presence of a variety of amygdala functioning in humans as well as
social cues. Negative unconditioned stimuli act animals. Importantly, although it at first appears
as punishers or negative reinforcers, and in- as if threat responding in the amygdala is an
clude pain, social deprivation, and putrefying entirely bottom-up phenomenon, there is evi-
odors (Rolls, 2007a). With repeated exposure to dence that amygdala activity is modulated by
unconditionally reinforcing stimuli, dopaminer- top-down processes related to attention (Pes-
gic neurons in the VTA become sensitive to soa, Kastnerb, & Ungerleider, 2002).
cues associated with those stimuli. In this way,
the VTA begins to activate the nucleus accum- Interestingly, the amygdala is exquisitely sensi-
bens earlier and earlier in a “chain of cues” that tive to social signals expressed on the face (Be-
increase the probability of coming into contact nuzzi et al., 2007; Rolls, 2007b). Human pa-
with the original unconditioned reinforcer (e.g., tients with impaired amygdala functioning have
an attractive potential mate). Put another way, difficulty processing emotional facial expres-
conditioned associations between cues related sions, especially those communicating social
emotions (Adolphs, Baron-Cohen, & Tranel,
6
2002; Adolphs & Tranel, 2003; Adolphs, Tra- crine system, most famously in the case of cor-
nel, & Damasio, 1998). Fearful faces in particu- tisol release via the hypothalamic-pituitary-
lar reliably activate amygdala in normal human adrenal (HPA) axis (Kemeny, 2003). The hypo-
subjects (Thomas et al., 2001; Whalen, in thalamus receives inputs from a wide variety of
press), even when the presentation of faces is structures implicated in social behavior, emo-
so rapid that subjects have no conscious mem- tion, stress, and attachment, including the
ory of them (Whalen et al., 1998), or when the amygdala, prefrontal cortex, and hippocampus
faces are reduced to “essential elements,” such (McEwan, 2007). The periventricular nucleus of
as when no cue but the raised upper eyelid is the hypothalamus is capable of synthesizing
shown (Whalen et al., 2004). corticotrophin-releasing hormone (CRH;
Gainer & Wray, 1994). In threat responding,
Bearing all of this in mind, it is noteworthy that CRH released by the hypothalamus stimulates
the amygdala also plays a major role in the con- the release of adrenocorticotropic hormone
solidation of both positive and negative long- (ACTH) in the pituitary gland. ACTH causes
term memories. Amygdala activity during increased production of cortisol and catecho-
memory encoding is associated with the recall lamines (e.g., epinephrine and norepinephrine)
of emotionally salient information even weeks in the adrenal cortex. This cortisol is circulated
after testing (Hamann, Ely, Grafton, & Kilts, throughout the body, including the brain. Criti-
1999). Beta-adrenergic blockade of amygdala cally, circulating cortisol in the brain is capable
function appears to impair these effects (Cahill, of activating glucocorticoid receptors in the
Prins, Weber, & McGaugh, 1994). These find- hippocampus that feed back to inhibit the
ings suggest that the amygdala “tags” sensory HPA- axis (Kemeny, 2003).
experiences as significant or salient and that
this tagging is prominently represented in long- Importantly, the hypothalamus is one of the
term memory consolidation. Importantly, the key structures implicated in the regulatory ef-
hippocampus appears to support the forma- fects of social soothing on neural threat re-
tion, storage, and consolidation of associations sponding, including interactions with attach-
between internal states and spatial or contextual ment figures (Carter, 2003; Coan, Schaefer, &
environmental stimuli (Brasted, Bussey, Murray, Davidson, 2006b). The precise mechanisms by
& Wise, 2003; Kennedy & Shapiro, 2004). which social soothing down-regulates HPA-axis
activity are currently unknown (Coan et al.,
Ultimately, both the amygdala and the hippo- 2006b), but the hypothalamus is known to co-
campus are likely to underlie the identification ordinate the activity of many behavioral and
and consolidation of significant interactions physiological systems, including those involved
between attachment figures and emotionally in maternal behavior and pair bonding. Moreo-
salient situations. The amygdala will tag emo- ver, maternal and pair bonding behaviors are
tionally salient stimuli and participate, along strongly associated with oxytocin and vaso-
with the hippocampus, in the consolidation of pressin, neuropeptides (reviewed below) that
contextual cues associated with those stimuli in the hypothalamus is capable of synthesizing in
long-term memory. Among those contextual abundance (Carter, 2003; Gainer & Wray,
cues will be the behavior of attachment figures. 1994).
Threat responding, social soothing, and the hypothala- The prefrontal cortex (PFC), emotion, and emotion
mus. The hypothalamus regulates a variety of regulation. Many regions of the PFC are impli-
metabolic and autonomic processes, as well as cated in emotion, motivation, and emotion
linking the central nervous system to the endo- regulation (Coan & Allen, 2004; Coan, Allen, &
7
McKnight, 2006a). Indeed, portions of the working memory, language, and action planning
PFC are strongly connected to the dopaminer- operations (Ochsner et al., 2002).
gic projection system (e.g., nucleus accumbens
and VTA), and the PFC shares numerous con- Thus, the PFC may be associated with attach-
nections with the amygdala, hippocampus, and ment processes in at least two ways. First, over
hypothalamus. For example, the orbitofrontal time, medial orbital circuits may encode condi-
region of the PFC assists the amygdala and tioned or “automatic” responses to attachment
hippocampus in linking the emotional value of figures related to excitatory or inhibitory re-
secondary sensory information (e.g., place cues) sponses to threat cues. Second, dorsolateral
to primary reinforcers such as food, water, and circuits may modulate cognitive operations as-
social contact (Rolls, 2007a). sociated with attachment figures in reflective,
working memory. In truth, these distinctions
One of the major functions of the PFC is the are not likely to be as discrete as the above
regulation of emotion. Prefrontal regions may formulation suggests, but the distinction be-
bias brain circuits responsible for appraising the tween medial orbital and dorsolateral circuits of
emotional content of sensory stimuli and in- the PFC offers a useful neural heuristic for
stantiating behavior directed toward approach- thinking about the regulatory influences of at-
or avoidance- related goals (e.g., via amygdala tachment figures in automatic versus explicit
or nucleus accumbens; Davidson & Irwin, terms, respectively.
1999). Different portions of the PFC underlie
different emotion-regulation strategies (see Emotional constituents in combination. Because all of
Ochsner & Gross, 2005, for a review). These the constituent systems described above are
can include “automatic” forms of emotion- linked, it is possible for them to coordinate in
regulation and effortful forms related to the important ways. For example, dopaminergic
cognitive control of attention or stimulus ap- neurons in the VTA share connections with
praisal (Ellenbogen, Schwartzman, Stewart, & many regions other than the nucleus accum-
Walker, 2006). Automatic forms of emotion bens, including the amygdala (in various nuclei
regulation include conditioning and extinction as well as the extended amygdala), the hippo-
learning, including instrumental avoidance. campus, the hypothalamus, and the PFC
These rapid and automatic regulatory functions (Depue & Collins, 1999). In this way, these
(especially extinction learning) have been asso- structures form their own distributed networks
ciated with the ventromedial and medial orbital of often reciprocal influence. To understand
PFC (Milad et al., 2005; Quirk & Beer, 2006; how such a network may function, consider the
Sierra-Mercado, Corcoran, Lebrón-Milad, & distribution of activity following an encounter
Quirk, 2006). More “effortful” forms of regu- with an unconditionally rewarding stimulus.
lation require attention, working memory, and Dopamine is released from the VTA, which
other cognitive operations (Ochsner, Bunge, stimulates dopaminergic activity in the nucleus
Gross, & Gabrieli, 2002). For example, cogni- accumbens associated with pleasure. The
tive reappraisals have been used to alter the amygdala “tags” sensory properties of the
meaning of a stimulus, and attentional practices stimulus as affectively salient or significant,
(e.g., meditation) have been used to alter atten- placing special emphasis on those properties
tional foci associated with affective stimuli. during the process of long-term memory con-
These processes have been associated with solidation via the hippocampus, which also en-
more lateral, especially dorsolateral, portions of codes contextual information as part of the
the PFC—regions also known to support consolidation process. The PFC uses this in-
formation to effect action plans and regulate
8
subsequent behavior—both automatic and ef- the establishment and maintenance of prefer-
fortful—relevant to the stimulus. As experience ences for familiar others (caregivers, peers,
with the rewarding stimulus increases in fre- one’s mate, etc.) form the first necessary condi-
quency (partly as a function of successful regu- tion of attachment bonds. Through evolution-
lation and action planning activity in the PFC), ary time, familiarity was likely a matter of sur-
the affective “tagging” of cues associated with vival, and so it remains in the case of infants
it proceeds down a “chain of cues,” increasing and their caregivers. One of the striking things
the probability that the rewarding stimulus will about humans (and many other mammals) is
be accessed (or avoided in the case of uncondi- how well designed we are for affiliation (Depue
tionally negative reinforcers). & Morrone-Strupinsky, 2005). Many stereo-
typed behaviors, including facial expressions,
For a more concrete example, consider an en- vocalization, bodily gestures, etc., are calibrated
counter with an attractive potential mate. In to signal social closeness and/or discomfort.
many species, including humans, such an en- These signals are readily recognized by most
counter is unconditionally reinforcing. The en- humans, and may in many cases be innate
counter initially elicits pleasurable feelings and (Laird & Strout, 2007; Rolls, 2007a).
an increase in incentive motivation associated
with the partner. Amygdala tags sensory fea- More than half a century ago, Bowlby (1969/
tures of the encounter as salient during the 1982) suggested that infant-mother bonds,
process of memory consolidation in coopera- characterized by both the ability to distinguish
tion with the hippocampus, and the VTA be- the caregiver from others and a strong prefer-
comes conditioned to cues associated with (and ence for the caregiver, formed very rapidly, and
predictive of) the potential mate, thereby acti- this appears to be true in many species. Most
vating incentive motivation circuits early in the researchers who study infants agree that the
“chain of cues” that will increase the likelihood development of attachment bonds is critical,
of encountering the potential mate again. With because infants often must survive long periods
repeated exposures, and perhaps a bit of luck, of early development totally dependent upon
the potential mate may even respond in kind. their caregivers, even when those caregivers are
With this, the foundation of pair-bond attach- neglectful or abusive (Simpson & Belsky, Chap-
ment has been set, and the complex process of ter 6, this volume). The formation of such
attachment bonding has begun (see Zeifman & bonds appears mainly among birds and mam-
Hazan, Chapter 20, this volume). During the mals, and is thought to have been present in
attachment bonding process, the PFC utilizes their common ancestor, the therapsids (Insel &
information about the potential mate to adjust Winslow, 1998).
its emotion-regulatory activities, opting, in
many cases, to cede some level of regulatory Among social species, the most common mani-
effort to the potential mate, as discussed below. festation of the attachment bond—indeed its
commonest exemplar—is, as Bowlby sug-
Social Elements gested, the bond between a human infant and
its mother (Insel & Fernald, 2004). Human in-
Familiarity and preference. One of the bedrock fants have the capacity to distinguish their
features of any species deemed social (as well mother from others within hours after birth
as any conception of attachment) is the ability (DeCasper & Fifer, 1980). Most researchers
to distinguish individuals who are familiar from agree, however, that in many species attach-
those who are not—an ability that in turn is ment bonding represents a more generalized
yoked to a preference for the familiar. Indeed, capacity—one that is only very frequently ap-
9
plied to the actual mother. Indeed, many birds It is largely for this reason that at least some of
become bonded within hours to the first mov- the neural circuitry associated with attachment
ing object they encounter. Interestingly, Lorenz in infants is likely to be different from that in
(1935) discovered that geese reared by him not adults. This may explain why filial bonds occur
only bonded to him (and followed him) as if he so rapidly and unconditionally by comparison
was the parent, but also that they “courted” with attachment formation in adulthood. In
him upon reaching sexual maturity, preferring fact, filial bonding may precede birth, where
him to other geese. These observations raise learning about the mother’s voice and odor may
important questions for a neuroscience of at- occur. In many species, odor is thought to play
tachment, concerning the degree to which early a significant role in the identification of the
sensory objects associated with a caregiver are primary caregiver soon after birth and thereaf-
rapidly and permanently “etched” into the de- ter, even among human infants (Insel & Fer-
veloping brain, how such a thing can occur, and nald, 2002). For example, maternal odor has
whether a critical period of bonding formation been observed to elicit orienting responses in
exists in early development. infants, as well as having soothing effects on
human infant crying (Marlier & Schaal, 2005;
Filial bonding, the locus coeruleus, and the amygdala. Schaal, Marlier, & Soussignan, 1998).
Filial affiliations are those concerning an off-
spring relating to a parent. In humans, strong Filial bonding also occurs in a context of sig-
attachment to the caregiver usually develops at nificant neural development. The human brain
six months of age, but filial bonds resembling grows exponentially during the first year of life
this process appear from birth. Filial bonds and continues to develop rapidly into the sec-
may, however, differ from adult affiliation be- ond year (Franceschini et al., 2007). Glucose
haviors in important ways due to the dependent metabolism rises gradually until about the 4th
nature of the offspring-parent relationship. year, and on average the level of brain glucose
Many offspring of social species are totally de- metabolism is more than double that of adults
pendent upon a caregiver for survival, and at- until about age 10 (Chugani, 1998). The pro-
tachments are imperative regardless of the duction of neurotrophins— proteins that aid in
quality of the care (Hofer & Sullivan, 2001). neuron survival—are dependent upon neuronal
Indeed, nonhuman primates have been ob- activity and, by extension, environmental stim-
served to exhibit strong attachments to their uli (Berardi & Maffei, 1999; Cancedda et al.,
mothers even when the mothers are abusive, 2004). Within the first two years of develop-
and this pattern extends to human children ment in humans, the brain’s production of ax-
(Moriceau & Sullivan, 2005). Rat pups have ons, dendrites, and synapses far exceeds its
been observed to form preferences even to needs. Synaptic connections are then “pruned”
stimuli paired with electric shock, a seemingly throughout childhood due to lack of use; that
paradoxical effect thought to have developed as is, synaptic connections that go unused are dis-
a means of preventing pups from aversion carded (Reichardt, 2006). In this way, the envi-
learning while being handled roughly by the ronment exerts its influence on the otherwise
mother (Hofer, 2006), an unfortunate predica- genetically determined neural development of
ment but generally not as unfavorable as being the brain. At a systems level, neural organiza-
abandoned. Ultimately, filial bonds need to be tion tends to follow functioning— repetitive
understood in the context of this high level of and patterned activation—during development
dependence, at least early in development. (Hebb, 1949; Posner & Rothbart, 2007).
10
Throughout the earliest stages of this process, cease vocalizations in late infancy, likely due to
at least two brain structures, the locus co- a lack of auditory stimuli (Schauwers et al.,
eruleus and the amygdala, interact to facilitate 2004). Interestingly, research on the social
the familiarity and reinforcement associated complexity of rearing environments in rats
with the caregiver in filial bonding. Although in suggests that environments rich in social and
adults, norepinephrine (NE) moderates mem- cognitive complexity are associated with signifi-
ory consolidation and learning (Cahill et al., cantly more synapses per neuron throughout
1994), NE from the locus coeruleus appears to the visual cortex compared to simple socially
be both necessary and sufficient for learning in paired housing and individual housing (Briones,
human and animal neonates (Sullivan, 2003). Klintsova, & Greenough, 2004). These effects
And the neonate locus coeruleus releases large remained even after later environments were
amounts of NE early in development (Naka- changed or reversed, suggesting that plastic
mura & Sakaguchi, 1990). When combined changes associated with early experiences are
with sensory information such as the look, persistent.
sound, and smell of a caregiver, that sensory
information is likely to be learned rapidly. Im- In combination, these findings suggest that fil-
portantly, this learning is occurring alongside a ial bonding occurs rapidly and unconditionally.
neonatal amygdala that is not yet fully func- Moreover, the filial bond develops in a context
tional, making it difficult or impossible for of rapid neural development, during what ap-
aversive conditioning to occur (Sullivan, 2003). pears to be a sensitive period of learning. As
In other words, the amygdala, being immature will be discussed in greater detail below, this
during early neonatal development, may not be process, especially to the extent that it involves
capable of associating aversive stimuli with developing links between the PFC and affective
alarm or avoidance behavior, leaving virtually structures like the amygdala and nucleus ac-
all stimuli to be simply encoded as “familiar,” cumbens, may result in the development of
which is, for many intents and purposes at this different reflexive “assumptions” about the na-
stage, unconditionally reinforcing. ture of the social world, including that world as
it will be encountered in the future. This may
During this developmental period, neural set the stage for different broad strategies for
pathways linking amygdala to hippocampus are engaging (or avoiding) social stimuli, perhaps
similarly underdeveloped, as are many regions especially during emotional situations. Indeed,
within the PFC (Herschkowitz, 2000). This conditions under which the filial bond forms
suggests that learning in neonates may not in- and develops may constitute a kind of rudi-
volve the PFC, or may do so only in limited mentary “pre-working model” of interdepend-
ways. In either case, these systems begin to de- ence and affect regulation—of attach-
velop rapidly in infancy, leading many to refer ment—that is either altered or reinforced dur-
to this developmental time as a “critical” or ing the course of development throughout
“sensitive” period for neural development. Sen- childhood.
sitive periods have been studied extensively in
terms of the brain’s sensory systems. For ex- Adult affiliation, nucleus accumbens, and the social
ample, Hubel and Wiesel (1970) observed that neuropeptides. Of course, attachment bonds
a temporary blockage of visual input to one eye characterized by interdependence and affect
in cats during early development caused irre- regulation extend far beyond the prototypic
versible impairment in the visual cortex. Simi- mother/infant relationship. Adult attachments
larly, children born deaf have been observed to occur in the context of romantic relation-
ships—especially monogamous ones—but
11
adult attachment is probably not restricted to studies have linked dopamine release in the
this. Indeed, relationships that meet attachment nucleus accumbens and VTA to the spontane-
criteria have by now been documented between ous establishment of partner preferences (Ara-
pairs of individuals as diverse as adult romantic gona et al., 2006).
partners (Fraley & Shaver, 2000); captive chim-
panzee cage mates (Bard, 1983; Miller, Bard, Mating behavior in the absence of partner
Juno, & Nadler, 1986); chimpanzees and their preference is also associated with dopamine in
human caretakers (Miller, Bard, Juno, & Nadler, the nucleus accumbens (Balfour, Yu, & Coolen,
1990); and even between domesticated dogs 2004; Pfaus, Kippin, & Centeno, 2001), how-
and their owners (Topal, Miklosi, Csanyi, & ever, suggesting that dopaminergic activity in
Doka, 1998). Aspects of attachment seem to the nucleus accumbens is insufficient for the
occur even between organization members and establishment of partner preferences. This
their leaders (Davidovitz, Mikulincer, Shaver, raises the question of how the establishment of
Ijzak, & Popper, 2007). partner preferences is “linked up” to the do-
paminergic incentive motivation system. Here,
Of interest here are neural circuits that support the neuropeptides oxytocin and vasopressin
the establishment and maintenance of attach- appear to play major roles (Depue & Morrone-
ment bonds in later childhood and adulthood. Strupinsky, 2005; Young & Wang, 2004). Both
How does the brain facilitate movement from have been associated with the formation of
close proximity, to familiarity, to attachment? partner preferences regardless of mating be-
To start, positive, possibly unconditioned, so- havior, and both, but especially oxytocin, are
cial affiliation behaviors (e.g., eye gaze, soothing elicited by positive social behaviors (Uvnaes-
vocalizations, non-threatening facial and bodily Moberg, 1998).
behaviors) increase proximity between conspe-
cifics, setting the stage for motivated attach- Perhaps the most celebrated example of the
ment bonding. It is clear that some social cues function of these neuropeptides derives from
are unconditionally capable of activating neural work on pair bonding within monogamous
structures supporting incentive or reward moti- prairie voles (Borman-Spurrell, Allen, Hauser,
vation, especially the nucleus accumbens and Carter, & Cole-Detke, 1995; Carter, 2003; Insel
the VTA (Allen et al., 2003). For example, pas- & Fernald, 2004; Young & Wang, 2004). When
sively viewed images of female faces have been these animals forge a pair bond, they mate,
observed to activate the VTA and nucleus ac- share nests and territory, cooperate in care of
cumbens unconditionally in heterosexual men young, and forcefully reject intruders of either
(Aharon et al., 2001). In rats, maternal females sex (Borman-Spurrell et al., 1995). Unlike
show an increase in dopamine release in the nonmonogamous animals—including other
nucleus accumbens when exposed to pups variants of vole—the nucleus accumbens of
(Hansen, Bergvall, & Nyiredi, 1993). Depletion these animals is rich in oxytocin receptors.
of dopamine in the VTA and nucleus accum- Moreover, structures like the ventral tegmen-
bens via lesions or dopamine antagonists virtu- tum and ventral palladium are rich in receptors
ally eliminates rat maternal behavior (Hansen, for vasopressin (Lim, Hammock, & Young,
Harthon, Wallin, Löfberg, & Svensson, 1991). 2004; Lim & Young, 2006).
Interestingly, maternal behaviors not directly Findings such as these provide clues as to how
associated with caregiving, such as nest build- social cues activate incentive motives associated
ing, passive nursing, and aggression, are virtu- with dopaminergic activity and in turn the for-
ally unaffected by these manipulations. Other mation of partner preferences and proximity-
12
seeking behavior. Socially sensitive oxytocin level reward-related proximity conditioning is
and vasopressin circuits in the VTA, nucleus tightly bound to the provision of security by
accumbens, and ventral palladium probably the attachment figure in other contexts. From
stimulate dopaminergic activity linked to incen- the perspective of the VTA and nucleus ac-
tive motivation. Because activation of this do- cumbens, there may be little difference, because
paminergic system is frequently associated with they become in either case sensitized to the
positive affect and reward, it may be that the presence of the attachment figure as a positive
degree of oxytocin and vasopressin activity de- outcome.
termines the degree to which a social experi-
ence is rewarding, by virtue of the dopaminer- In addition to the reinforcing nature of dopa-
gic cascade that follows it. minergic activation, consummatory pleasure
may play a role in rewarding social interaction.
Proximity seeking, the dopamine system, and After all, positive social experiences are charac-
endogenous opiates. One of the natural conse- terized in everything from semi-structured sci-
quences of familiarity, preference, and bonding entific interviews to ancient literature as involv-
is proximity seeking, a characteristic of social ing feelings of warmth, closeness, love, affec-
behavior strongly associated with attachment. tion, and pleasure. Depue and Morrone- Strup-
Proximity seeking is likely an extension of mo- insky (2005) have argued that feelings of con-
tivational circuits associated with reward and summatory pleasure promote the development
partner preference. Of course, individuals can of contextual associative memory networks
seek close proximity as a function of positive that help both to establish and to maintain so-
affect and reward or in response to cues of cial bonds and that are ultimately responsible
punishment where the goal is the provision of for many of the regulatory effects associated
safety (Depue & Morrone-Strupinsky, 2005). In with the soothing and security provided by at-
the case of positive affect, proximity is sought tachment relationships. The critical substrate
because the attachment figure has become as- for these feelings, and perhaps for the socioaf-
sociated with rewarding feelings of pleasure, fective regulatory effects that accompany them,
and close proximity increases the frequency or may be the release of opiates that often follow
intensity of these feelings. In the case of nega- activation of oxytocin receptors, also in struc-
tive affect, the attachment figure may serve as a tures like the nucleus accumbens and VTA.
safety cue, eliciting approach behaviors ori-
ented toward the acquisition of security. In this There is abundant evidence for the role of en-
way, proximity seeking can involve both dogenous opiates in a wide variety of social
reward-related approach behaviors and ap- behaviors. In humans and other animals, these
proach behaviors associated with active avoid- opiates are released during childbirth, nursing,
ance. maternal caregiving, sexual activity, and many
modes of tactile stimulation, including groom-
Behaviorally, these motivations may appear to ing and play behavior (Carter & Keverne, 2002;
be identical, but they are likely to involve both Keverne, Martensz, & Tuite, 1989. This release
shared and distinct neural circuits. Moreover, may mediate the reward associations that are
although attachment theory emphasizes the forged between infants and mothers, as well as
emotion-regulatory function of proximity- between romantic partners, and even platonic
seeking due to the need for security, it may be friends. For example, morphine, an opiate re-
counterproductive to downplay the role of ceptor agonist, increases the reinforcing effects
proximity-seeking due to reward processes. It of a host of maternal behaviors, mother-infant
may be the case, for example, that at the neural bonding, time spent by juveniles (rats) with
13
their mothers after a brief separation, groom- tachment bonds and capabilities persist among
ing, and juvenile play behavior (Agmo, Barreau, so many species.
& Lemaire, 1997; Niesink, Vanderschuren, &
van Ree, 1996; Nocjar & Panksepp, 2007; Pank- Function can be considered in a more proxi-
sepp, Nelson, & Siviy, 1994). By contrast, opi- mal, ontogenetic sense as well, and it is at this
ate receptor antagonists such as naltrexone re- level that the regulation of affect may take cen-
duce reward conditioning effects associated ter stage. Proximal functions of the attachment
with each of these forms of social contact system are, following basic survival during in-
(Graves, Wallen, & Maestripieri, 2002; Hollo- fancy (Hofer, 2006), primarily concerned with
way, Cornil, & Balthazart, 2004). In humans, the social regulation of emotional responding.
the administration of the opiate antagonist Bowlby (1969/1982), following along with
naltrexone was associated with increased volun- Ainsworth and her colleagues (e.g., Ainsworth,
tary isolation from friends, as well as decreased Blehar, Waters, & Wall, 1978), argued that a
levels of enjoyment in the company of others critical function of attachment figures was the
(Jamner & Leigh, 1999). provision of a secure base from which infants
could explore their worlds relatively free of
Importantly, tactile stimulation appears to play anxiety, and a safe haven to which the infant
a particularly powerful role in the activation of could return when distressed. It was proposed,
affiliative reward conditioning (Burgdorf & for example, that the base from which an infant
Panksepp, 2001; Melo et al., 2006). In some could explore its world was secure to the extent
animals, the affiliative conditioning associated that the caregiver was responsive to the infant’s
with maternal behavior is attenuated in the ab- distress. Many have since proposed that the
sence of tactile stimulation (Melo et al., 2006). quality of the caregiver-infant attachment
bond—especially of the caregiver’s status as a
Attachment and the secure base—holds consequences for child and
Social Regulation of Emotion adult emotional functioning, including styles of
interpersonal relating and emotion-regulation
Many evolutionary accounts of the reproduc- capabilities. A very large behavioral database
tive advantages of infant-caregiver bonds have now supports this notion with respect to both
been proposed, but similar accounts of adult childhood and adulthood (for reviews, see
attachment bonds are relatively recent (Simp- Thompson, Chapter 16, and Mikulincer &
son & Belsky, Chapter 6, this volume; Zeifman Shaver, Chapter 23, both in this volume).
& Hazan, Chapter 20, this volume). Fraley and
Shaver proposed that adult attachments repre- Throughout childhood, and certainly by adult-
sent homologies of the infant-caregiver bond hood, the regulatory effects of attachment rela-
co-opted by natural selection to facilitate pair tionships are likely to be felt in two broad ways.
bonding (Fraley & Shaver, 2000). By this ac- The first is immediate, as when the attachment
count, adult and infant-caregiver attachment figure is present and regulating emotional re-
systems entail similar goals (the survival of off- sponding “on line.” An example of this may be
spring) and operate according to similar condi- when a caregiver holds her child’s hand during
tions of activation (e.g., presence of a threat) a blood draw at the doctor’s office, thus actively
and termination (e.g., regulation of threat re- soothing the child’s anxiety as it occurs. The
sponding by the attachment figure). Evolution- second is generalized, where the attachment
ary perspectives like these address ultimate figure is present only in the form of a mental
function, in the sense of explaining why at- representation. These representations may in
theory manifest either as “internal working
14
models” based on procedural and semantic In humans, very little work to date has actually
memory, or as declarative, explicitly recalled sought to identify how neural circuits associ-
mental images. Indeed, on-line regulation expe- ated with social affiliation and emotion func-
riences likely condition mental representations tion in a context that combines social interac-
in both implicit and declarative memory. In the tion with externally generated emotional stress.
sections that follow, immediate, “on line” regu- Recently, Coan and colleagues (Coan et al.,
lation is considered in contrast to “mental rep- 2006b) collected functional brain images from
resentations” of the attachment figure, often 16 married women as they were subjected to
referred to as “internal working models,” that the threat of mild electric shock while either
may serve to preempt the level of distress an holding their husband’s hand, holding the hand
individual experiences in the face of a potential of an anonymous male experimenter, or hold-
threat. ing no hand at all. Their results suggest that
physical contact from both attachment figures
The “On-Line” Social Regulation and strangers attenuates threat responsive neu-
of Emotion ral activity in affect-related action and bodily
arousal circuits (e.g., in the ventral anterior cin-
Many researchers have observed the stress- gulate cortex), but also that down regulation of
buffering effects of social contact on behaviors structures such as the nucleus accumbens, dor-
and physiological systems related to emotional solateral prefrontal cortex, and superior collicu-
responding. This social buffering occurs at all lus was achieved only via hand holding with the
levels (e.g., group, caregiver, familiar conspe- attachment figure. Moreover, Coan et al.
cific), but familiarity and attachment are associ- (2006b) observed that some of the regulatory
ated with the strength of social regulation ef- effects of soothing physical contact varied as a
fects. Even in rats, the presence of familiar function of relationship quality, with higher
conspecifics (“buddy” rats) increases explora- quality predicting yet greater attenuation of
tion and attenuates HPA-axis activity under threat-related neural activation in the right ante-
conditions of threat (Kiyokawa, Kikusui, rior insula, superior frontal gyrus, and hypo-
Takeuchi, & Mori, 2004; Ruis et al., 1999; Ter- thalamus during spousal, but not stranger, hand
ranova et al., 1999). Familiar conspecifics at- holding. These findings suggest that social
tenuate emotional stress responding in non- proximity in general, and the presence of an
human primates during new social group for- attachment figure in particular, exerts bottom-
mation and social conflict (Gust, Gordon, Bro- up regulatory influences on the perception of
die, & McClure, 1996; Weaver & de Waal, threat in the brain. Moreover, the fact that
2003). As reviewed above, these effects are stranger hand holding conferred regulatory
widely believed to derive from social cues that benefits at all suggests that the human brain is
activate the release of oxytocin and vasopressin unconditionally soothed to some extent by so-
in the VTA, ventral palladium, and nucleus ac- cial proximity, which may lay the groundwork
cumbens (Carter & DeVries, 1999; Heinrichs et for the additional regulatory benefits associated
al., 2001; Izzo et al., 1999; Uvnaes-Moberg, with attachment figures.
1998; Windle, Shanks, Lightman, & Ingram,
1997). This in turn is thought to activate do- Internal Working Models and
paminergic and endogenous opiate activity as- Individual Differences
sociated with consummatory pleasure and
physiological soothing. Thus far, we have primarily considered basic
systems supporting “normative” manifestations
15
of attachment behavior, as well as a concrete marily mothers) can result in socially deviant
example of the emotion-regulation functions behavior and physiology later in life (Mineka &
of the attachment system occurring “on line” Suomi, 1978). Among brown capuchins mon-
in real time. However, the emotion-regulatory keys, patterns of mother/offspring behavior
effects of caregiving experiences such as those partially determine the post-conflict reconcilia-
between infants and caregivers, or even be- tion styles of offspring during later interactions
tween romantic partners, are likely to extend far with nonfamilial conspecifics (Weaver & de
beyond online moments of soothing and secu- Waal, 2003).
rity provision. Bowlby (1979) considered many
facets of early attachment experiences to hold Neural mechanisms linking early parental care
implications for interpersonal and emotional to trait-like individual differences in threat re-
functioning “from the cradle to the grave” (p. sponding over the life span have been expertly
129), and in the past several decades many re- described by Meaney and colleagues (Weaver et
searchers have adopted this idea as one way to al., 2004). This work suggests that in rats,
understand adult interpersonal functioning and grooming behavior by the mother “sets” or
emotion-regulation capabilities. “programs” the degree to which her offspring
react to threat cues throughout their lives. This
Unfortunately, a large portion of what is modulation of threat reactivity has been ob-
known about links between early social experi- served both in behavior and in HPA- axis activ-
ence, neural development, and subsequent ity. Moreover, associations between maternal
emotional behavior derive from studies of grooming and offspring threat reactivity have
abuse and neglect. For example, neglect and been linked to the expression of specific genes
abuse (both physical and verbal aggression) are that moderate HPA-axis functioning. As re-
associated with risk for heightened stress reac- viewed above, the HPA-axis has its own built-in
tivity, anxiety, depression, and social deviance regulatory mechanism in the hippocampus,
that extend well into adulthood (Teicher, Sam- whereby circulating cortisol activates hippo-
son, Polcari, & McGreenery, 2006). In one re- campal glucocorticoid receptors, which in turn
cent study, children who had experienced social down-regulate the production of
deprivation and neglect in Romanian orphan- corticotrophin-releasing hormone in the hypo-
ages were observed to have lower overall levels thalamus. Grooming induces the expression of
of vasopressin, as well as blunted oxytocin re- genes that encode for glucocorticoid receptors
sponses to physical contact with their caregiv- in the hippocampus, thus making the hippo-
ers, relative to normally family-reared children campus more sensitive to circulating cortisol
(Wismer-Fries, Ziegler, Kurian, Jacoris, & Pol- and, hence, more susceptible to down-
lak, 2005). This is consistent with findings re- regulation during stress. Cross- fostering stud-
garding social isolation as a well-known risk ies by Meaney and colleagues strongly suggest
factor for a number of neurodevelopmental that lifelong stress reactivity, and even the sub-
and psychosocial problems, ranging from anxi- sequent maternal behavior of female rat pups,
ety and depression to increased risk of suicide, is largely attributable the degree of post-natal
family problems, and even stress-related dwarf- maternal grooming and not to genetic inheri-
ism (Barber, Eccles, & Stone, 2001; Kawachi, tance (Weaver et al., 2004).
2001; Newcomb & Bentler, 1988; Skuse, Al-
banese, Stanhope, Gilmour, & Voss, 1996). In Attachment and internal working models. According
nonhuman primates, frequent or prolonged to attachment theory (Bowlby, 1969/1982,
separation of offspring from caregivers (pri- 1973; Mikulincer & Shaver, 2007; Mikulincer
&Shaver, Chapter 23, this volume), threat de-
16
tection capabilities evolved in part to activate external signs of threat, mediated through the
the attachment behavioral system, thus increas- amygdala, nucleus accumbens and hippocam-
ing the likelihood that humans, beginning in pus, as well as portions of the prefrontal cor-
infancy, would seek out and maintain proximity tex. These conditioned associations may remain
to attachment figures. Moderating the degree to stable for long periods of time, especially to the
which proximity to attachment figures is sought extent that they continue to be reinforced by
out in the context of a threat is attachment se- internal feelings of security, prevailing social
curity, which is itself the product of many contingencies, or both.
attachment-related experiences involving both
threats and attachment figures. These experi- This process likely allows individuals to adapt
ences shape “internal working models” of at- themselves to a variety of environmental condi-
tachment that guide emotion-regulation tions (e.g., security restoring or enhancing expe-
throughout life (see Bretherton & Munholland, riences with attachment figures, frequent or
Chapter 5, this volume). According to Bowlby lengthy absence of the caregiver, abuse by the
(1969/1982), internal working models are men- caregiver, excessive caregiving). Such adapta-
tal representations of the availability and prac- tions are referred to, in various research tradi-
tical utility of attachment figures when threats tions, as attachment patterns, attachment styles,
arise, and of the self in relationship with these or attachment states of mind (e.g., secure, anx-
figures. ious, avoidant, preoccupied). These adaptations
are thought to be relatively stable when the in-
Recently, Hofer (2006) described a process by dividual remains in a stable environment, and
which very early developmental experiences in can be measured by observations, self-report
interactions with a caregiver may plausibly pro- questionnaires, and structured interviews (e.g.,
ceed from the on-line regulation of fundamen- Crowell & Fraley, Chapter 26, Kerns, Chapter
tal neural systems supporting sensory-motor, 17, and Solomon & George, Chapter 18, all this
thermal, and nutrient functions to the shaping volume).
of internal working models of attachment se-
curity. In this model, access to primary rein- Behavioral research on the effects of different
forcers (e.g., food, water, warmth, touch) is de- adult attachment styles suggests the presence of
pendent in early development on (a) caregiver two relatively independent axes regarding at-
support and (b) affective brain circuitry used to tachment insecurity—anxiety and avoid-
solicit caregiver support via expressed affect. ance—along which individuals can vary (J.
Over the course of development, what begins Feeney, Chapter 21, this volume; Mikulincer &
as the regulation of physiological needs via af- Shaver, 2007). Moreover, different combina-
fect becomes the regulation of affect per se tions of scores along these dimensions can re-
(Hofer, 2006). Throughout this process, the sult in particular styles of relating interperson-
regulatory behavior of the attachment figure ally. For example, individuals low in attachment
(e.g., the provision of security, the alleviation of anxiety and low in attachment avoidance would
distress) is likely to set expectations about the be considered generally secure in their attach-
availability of attachment figures during times ments to others. Individuals high in both
of stress—the “internal working models” re- avoidance and anxiety are thought to avoid at-
flecting attachment security. tachment relationships out of fear, while those
high in avoidance but low in anxiety are
Thus, internal working models likely reflect thought to be “dismissing” of attachments,
conditioned associations between proximity to compulsively self-reliant, and unlikely to seek
attachment figures and both internal needs and proximity to attachment figures under stress
17
(Bartholomew & Horowitz, 1991; Brennan, with activity in the same region when holding
Clark, & Shaver, 1998). Finally, individuals low the hand of a stranger. The vACC is implicated
on avoidance but high on anxiety are thought in the modulation of affect-related arousal.
to be preoccupied with attachment relation- Avoidance scores corresponded with increased
ships. activation during spouse hand holding, and de-
creased activation during stranger handholding,
Few studies to date have investigated individual in the right ventromedial PFC, a region com-
differences in attachment styles using measures monly associated with the regulation of nega-
of neural activity, and some of the work that tive affect.
has been done serves only as an approximation.
Indeed, attachment styles may, at a neural level, In another recent fMRI study, 20 women were
manifest as little more than individual differ- asked to think about—and then to stop think-
ences in response capabilities among neural ing about—various relationship scenarios (Gil-
circuits supporting emotion, emotion- lath, Bunge, Shaver, Wendelken, & Mikulincer,
regulation, and social behavior. Interestingly, 2005). Attachment anxiety was positively asso-
Dawson and colleagues (Dawson et al., 2001) ciated with activity in the dorsal anterior cingu-
observed that insecurely attached infants of late cortex, and anxiety scores were positively
depressed mothers were more likely to show correlated with brain activity in the temporal
PFC asymmetries lateralized to the right. By pole, but negatively with brain activity in the
this metric, asymmetries in EEG activity in the orbitofrontal cortex, during thoughts about
alpha (8-13Hz) range (Coan & Allen, 2003; negative relationship scenarios. This suggests
Coan & Allen, 2004) correspond with emotion that attachment-anxious individuals are not en-
regulation capabilities (Coan et al., 2006a), with gaging neural systems that would help to regu-
relatively greater left PFC activity indexing an late their emotional responses during negative
increased probability of approach behavior relationship thoughts.
(e.g., anger, joy), and relatively greater right
PFC activity indexing an increased probability More recently, Bucheim and colleagues
of withdrawal behavior (e.g., sadness, fear). (Buchheim et al., 2006) collected functional im-
Thus, according to Dawson, insecurely attached ages of the brain while adults told “attachment
infants of depressed mothers have a trait-like stories” in response to images from the Adult
propensity to engage in withdrawal behavior Attachment Projective (Lorberbaum et al.,
(Dawson et al., 2001). 1999) intended to activate the attachment be-
havioral system. Attachment stories from the
A very small number of studies have begun to AAP were used to classify individuals as either
associate adult attachment styles with brain “organized” or “disorganized.” Individuals
function using functional neuroimaging tech- classified as disorganized were more likely to
nology. Recently, Coan and colleagues (2005) show amygdala and hippocampal activation
reported a variety of interaction effects be- when shown pictures portraying traumatic as
tween self-reported attachment scores and opposed to neutral attachment situations.
hand holding condition (spouse, stranger,
alone) on threat-related neural activity through- Although findings from each of the studies
out the brain. For example, under threat of described above should be considered prelimi-
mild electric shock, secure attachment scores nary, they do contribute to our understanding
were negatively associated with activity in the of how attachment styles and internal working
ventral anterior cingulate cortex (vACC) during models may moderate neural processes associ-
spouse hand holding, and positively correlated ated with the regulation of emotion. They are
18
the initial steps in what is likely to be an in- proposed that one of the ways in which the
creasing effort to use brain imaging techniques brain manages energy expenditure is via altera-
to study the neural correlates and underpin- tions in sensory perception that aid in decision-
nings of processes studied previously only making about the deployment of an organism’s
through verbal and behavioral reactions to resources. For example, Proffitt (2006) has ob-
laboratory procedures. served that donning a heavy backpack causes
hills to appear steeper and objects to appear
The Social Baseline Model farther away, thus discouraging individuals from
using their resources to climb those hills or ap-
Social influences on the regulation of affect are proach those objects. In this way, the brain can
sufficiently powerful and unconditioned to be thought of as a “Bayesian machine,” making
suggest that the brain’s first and most powerful “bets” at any given time about what resources
approach to affect regulation is via social prox- to deploy, and at what level of effort (Addis,
imity and interaction. This is most obvious in Wong, & Schacter, 2007; Bar, 2007).
infancy, where very basic physiological needs
are regulated first via affect expression, leading The social baseline model proposes that social
to a dynamic of regulating affect per se, where species are hard-wired to assume relatively
caregivers become the primary agent through close proximity to conspecifics, because they
which infants regulate affective responding have adopted social proximity and interaction
(Hofer, 2006). For the infant, this is occurring as a strategy for reducing energy expenditure
in a context of rapid and expansive neural de- relative to energy consumption. This implies
velopment—possibly a “critical period” during that the absence of conspecifics, in defying this
which a number of expectations about the na- baseline assumption, functionally adds to the
ture of the infant’s future environment are be- perceived cost of interacting with the environ-
ing formed. A great deal of this development is ment—especially in threatening contexts (an
occurring in the prefrontal cortex, a region of implication discussed explicitly by Bowlby,
the brain powerfully implicated in self- 1969/1982). In other words, the social baseline
regulation of affect. Because the prefrontal cor- model proposes that social isolation is, for a
tex is underdeveloped in infancy, the caregiver social organism, akin to donning a heavy back-
effectively serves as a kind of surrogate pre- pack, altering the real and perceived demands
frontal cortex, a function that attachment fig- associated with its environment. There are at
ures likely continue to serve for each other to least two ways in which the presence of con-
varying degrees throughout life. specifics may reduce, for social organisms, the
actual and perceived cost of engagement with
What I will call the social baseline model sug- the environment. I will call these strategies risk
gests that social affect regulation was long ago distribution and load sharing.
adopted as an efficient and cost- effective
means of regulating affect. It draws on the Risk distribution. The first way in which social
principle of economy of action, which states species, including humans, benefit from close
that organisms must, over time, consume more social proximity is via the simple distribution of
energy than they expend if they are to survive risk in the environment. Many species benefit
to reproduce (Proffitt, 2006). Because all bodily from living in groups, and simple risk distribu-
activities—including tion strategies are likely to be plesiomorphic, or
relatively ancient in evolutionary terms. Al-
neural activities—expend energy, energy ex- though group living comes at a cost at the level
penditure must be managed. Proffitt (2006) has of resource consumption, the benefits may
19
outweigh those costs sufficiently to create con- thought to be particularly costly to deploy
ditions under which group cohesion ultimately (Galliot & Baumeister, 2007). Evidence for this
promotes the survival of each individual in the derives from studies of cognitive depletion as a
group. Risk distribution speaks to the amount consequence of effortful attention and self-
of risk a given individual carries as a function control. In this work, individuals who are asked
of the degree to which he or she is alone, and it to engage in tasks requiring self-regulation are
can manifest in many ways (Krebs & Davies, subsequently less capable of similar tasks.
1993). For example, the larger the group, the Moreover, engaging in these tasks has been ob-
more individuals there are to scan for possible served to result in temporary depletions in
signs of danger. Similarly, a given individual is blood glucose concentration (Galliot & Bau-
at substantially reduced risk of personal danger meister, 2007).
(e.g., predation) when group size increases. A
similar example among warm-blooded species The social baseline model predicts that the pre-
may be the thermal advantage of huddling to- frontal cortex, and many of the regulatory
gether. Some social species utilize group size to processes it supports, may be particularly af-
maximize their performance as predators, and fected by the presence of an attachment figure,
this, too, can be a form of risk distribution, for especially in the context of a threat. Here, the
if predation (especially of large target animals) advantage of close proximity extends far be-
is maximized in groups of predators, the risk yond simple models of risk distribution: Over
that any one predator will perish from starva- and above the dilution of risk via large num-
tion is minimized. bers, a trusted and interdependent associate can
be counted on to engage in a number of
From the perspective of the social baseline health- and safety-enhancing behaviors on
model, it is important that the brains of social one’s behalf. Such behaviors may include the
species appear to be capable of assessing the identification and acquisition of resources, vigi-
distribution of risk and making Bayesian deci- lance for environmental threats, caring for one’s
sions about the cost-effectiveness of affective needs, and nurturing of one’s offspring. These
behavior at any given time. Practically speaking, allegiances—these attachments—serve to dis-
the presence or absence of conspecifics pro- tribute the cost of many of life’s metabolically
vides, at the lowest level of social proximity, a expensive activities, not least being the regula-
heuristic for deploying potentially costly re- tion of one’s own negative affect. Simply put,
sources. For example, in the presence of oth- affect regulation is possible, but more difficult,
ers, individuals may work less hard at being in isolation. I refer to this second level of social
vigilant for—or even fleeing—predators. These regulation as load sharing, and I believe it is an
activities, which may be yoked to perceived essential component of attachment relation-
bodily resources (Proffitt, 2006), are deployed ships throughout the life span. Load sharing is
only as needed. The resources that are saved by likely to be apomorphic, or relatively advanced
close social proximity are either simply con- in evolutionary terms, having arisen as a strat-
served or used for other valuable purposes. egy relatively recently. Human brains are highly
sensitive to the load sharing significance of
Load sharing. Risk distribution processes are not close attachment bonds, and adjust their efforts
likely to have strong effects on processing at accordingly. For example, individuals in close,
the cortical level, especially in the prefrontal trusted relationships will invest less effort in
regions supporting attention, working memory, down-regulating their negative affect, leaving
and the self-regulation of affect. Interestingly, them less responsive to threat cues and other
such prefrontally mediated activities are
20
signs of possible harm (Coan et al., 2006b; scanner were confronted with the threat of a
Edens et al., 1992; Mikulincer & Florian, 1998; mild electric shock under each of three condi-
Robles & Kiecolt-Glaser, 2003). Thus, the so- tions: while alone, while holding a stranger’s
cial brain is designed in part to distribute affect hand, and while holding their spouse’s hand.
regulation activities to attachment figures. As Women in the highest quality relationships
with the metabolic benefits of risk distribution, showed the lowest degree of threat-related
this should produce major metabolic resource brain activation, limiting their response to rela-
savings. tively automatic regulation of threat perception
via structures such as the ventromedial PFC.
Unlike risk distribution strategies, which are When the marital relationship was of relatively
primarily sensitive to numbers alone, load shar- poor quality, however, the
ing, especially in adult attachment relationships,
likely develops as the brains of individuals in a number of problems confronting the woman’s
relationship become conditioned to one an- brain under threat increased to include atten-
other, especially in the context of coping with tion to bodily sensory afferents, presumably
threats. Over time, individuals in attachment related to the threat of shock (right anterior
relationships literally become part of each insula), task salience (superior frontal gyrus),
other’s emotion regulation strategy. This is not and release of regulatory stress hormones (hy-
metaphorical, but literal, even at the neural pothalamus).
level. For example, an individual who has been
alone for a long period of time may have Presumably, the regulatory benefits associated
learned to exercise his prefrontal cortex in the with attachment figures in both the higher and
service of regulating his threat responses. The lower quality relationships reflected the load
social baseline model predicts that upon estab- sharing function of attachment relationships.
lishing an attachment relationship, the individ- As the nature of the hand-holding partner
ual’s perception of the degree to which his en- switched from attachment figure to stranger,
vironment is threatening or dangerous will however, yet more problems presented them-
change, decreasing the frequency with which he selves, with additional threat-related brain acti-
exercises his prefrontal cortex in the service of vations triggered to solve them. For example,
emotion regulation. Note that this is because threat-related vigilance increased (e.g., via the
his brain assumes a decrease in the need for superior colliculus), effortful emotion-
emotion regulation. With sufficient experience regulation strategies were employed (e.g., via
in the relationship, the level of interdependence right dorsolateral PFC), and areas were re-
associated with emotion-regulation needs can cruited that indicated increased threat-related
become strong. Indeed, a grim reminder of this avoidance motivation (e.g., caudate/nucleus
occurs when one or the other member of an accumbens).
attached pair is suddenly absent due to death or Still, the brain was less active while holding the
divorce, leaving the partner severely dysregu- hand of a stranger than while alone, presuma-
lated (Bowlby, 1980; Sbarra, 2006); see Fraley & bly reflecting the effects of risk distribution via
Shaver, Chapter 3, this volume). fairly minimal social proximity. While alone, the
An example of this dynamic of increasing need brain appears to get busy solving yet more per-
for self- regulation as a function of distance ceived problems, adding to those already enu-
from an attachment figure can be found in the merated somatic preparations for threat re-
previously mentioned study by Coan et al. sponding, increasing bodily arousal (e.g.,
(2006b), in which married women in an MRI through the ventral ACC) and coordinating vis-
21
ceral and musculoskeletal responses (e.g., poste- about how to utilize one’s own neural resources
rior cingulate, supramarginal gyrus, postcentral in the presence or absence of strangers and
gyrus). attachment figures. A secure attachment style
presumably disposes a person to make bets
It is important to emphasize that social affect closely in accordance with the idealized picture
regulation appears to be a relatively bottom-up described above. By contrast, avoidant and anx-
process, as opposed to one’s solo affect regula- ious strategies may encourage individuals to
tion, which is more top down. When engaging make greater use of their own resources, even
in self-regulation, a person is likely to need to in the presence of social support, or to place
engage in costly, effortful cognitive and atten- themselves outside the reach of social support
tional strategies in the service of inhibiting ei- in the hopes of avoiding additional costs (e.g.,
ther somatic responses or structures supporting having to regulate others as well as self), thus
the identification of threat cues. This effortful again requiring one to rely on one’s own
regulation of affect relies to a great degree the emotion-regulation strategies.
prefrontal cortex. In this way, self- regulation
frequently occurs in the context of an affective At present, the social baseline model, as well as
response that has already occurred. By contrast, this Bayesian conceptualization of attachment
social affect regulation may often affect the style, is predominantly a matter of conjecture.
perception of threat in the first place, thereby As I stated at the outset, however, we have to
decreasing the need for threat responding and begin somewhere in the move from evolution-
leaving the prefrontal cortex with relatively little ary, behavioral, observational, self-report, and
or nothing to regulate. Thus, social affect regu- interview-based analyses of attachment proc-
lation could be said to be more efficient, or less esses to analyses based on the methods pro-
costly, than self-regulation strategies, such as vided by rapidly developing neuroscience. I ex-
the suppression of emotional responses, the pect that future neuroscientific studies of at-
cognitive reappraisal of threatening situations, tachment will provide additional clues as to the
and even popular strategies such as meditation. nature of social affect regulation in the brain.
The extent to which this is true awaits further
investigation. Recommendations and
Conclusions
Attachment styles as Bayesian priors. Of course, the
preceding discussion offers only a simplified, In this chapter I have sought to synthesize a
idealized model of social affect regulation, and broad array of studies in the service of intro-
one highly dependent on situational contingen- ducing the reader to the current state of the
cies. It is likely that superimposed on all of the neurosciences as they pertain to research on
processes described above are trait-like assump- attachment, and to propose a plausible model
tions about the function and metabolic cost of of how what is known about the social brain
social factors in regulating the perception of and affect regulation may eventually be com-
threat cues and, hence, of affect. Accordingly, bined with attachment theory. This effort nec-
one way to conceptualize attachment styles and essarily included discussions of the neural con-
internal working models is as prior probabilities stituents of attachment, from neural systems
in a Bayesian decision-making process, where supporting emotion and motivation to those
the goal is to predict the regulatory cost- supporting emotion- regulation, filial bonding,
effectiveness of attachment figures. In this way, familiarity, proximity seeking, and individual
attachment styles come to represent strategies, differences in attachment style. What follows is
based on prior experience, for making decisions a partial list of recommendations for research-
22
ers excited about pursuing the neuroscience of tive regulation of affect via social channels was
attachment. (Other models and suggestions can done only with women, and it may not general-
be found in other chapters in the present vol- ize to men.
ume, especially those by Simpson & Belsky;
Fox & Hane; and Polan & Hofer.) Pursue animal models of attachment style. To date,
virtually no studies exist of attachment styles in
Use designs that combine social contact with non-human animals, despite growing evidence
emotional provocations. Studies of the neural that other personality dimensions are evident in
systems underlying attachment should combine non-human animals (Gosling & John, 1999).
the presence or absence of attachment cues For example, King and Figueredo (1997) pro-
(e.g., proximity to attachment figures) with vided strong evidence that the “big five” per-
laboratory situations that elicit emotional re- sonality structure and distribution is very simi-
sponses, including threats to either the partici- lar in humans and chimpanzees, and the anxiety
pant’s attachment system or to the participant and avoidance dimensions of attachment style
directly. Many theorists have proposed that the are somewhat related to the big-five traits of
attachment behavioral system is activated dur- neuroticism and agreeableness, respectively
ing threats to the individual or to the individ- (Noftle & Shaver, 2006). Other personality
ual’s attachment bond, but few studies of at- traits shared to one degree or another with
tachment processes at the neural level have ac- humans have been observed in species as di-
tually designed studies with this in mind. verse as gorillas, hyenas, domesticated dogs,
Moreover, no work to date has sought to iden- cats, donkeys, pigs, rats, octopi, and even gup-
tify how social contact influences neural re- pies (Gosling & John, 1999). Attempts to study
sponses to positive affect elicitations. attachment styles in non-human animals would
constitute a badly needed step toward bridging
Be sensitive to sex differences. Little is known about the gaps between the human and animal litera-
how the sex of an individual under study af- tures addressing attachment behavior.
fects activity in the attachment behavioral sys-
tem, or the neural constituents of attachment. Allow for systemic effects in research designs.
Self- reported sex differences have been noted Most attachment style research identifies effects
in behavioral studies, however. For example, of a given participant’s attachment style on that
women are more likely to endorse items indi- person’s own attachment behavior. One ques-
cating a preoccupied attachment strategy (char- tion of great interest is the degree to which the
acterized by worry that the partner will leave attachment style of one member of a dyad af-
them), whereas men are more likely to endorse fects the behavior of the other member. (See J.
a dismissive-avoidant strategy (characterized by Feeney, Chapter 21, this volume for examples.)
discomfort with interpersonal closeness) (Bar- For instance, Coan et al. (2005) presented evi-
tholomew & Horowitz, 1991). And many stud- dence that the husband’s preoccupation score
ies have found that women are most bothered corresponded with increased neural threat reac-
by their male partners’ avoidance, whereas men tivity throughout the wife’s brain if she was
are most bothered by their female partners’ holding the hand of a stranger (while her pos-
anxiety (Mikulincer & Shaver, 2007). Others sibly jealous husband looked on). These sorts
have reported sex differences in relationship of effects are likely to be numerous and are of
stability as a function of attachment styles, sug- great interest to any neuroscience of attach-
gesting that attachment styles may interact in ment. (Such findings also suggest that the ef-
important ways with gender roles (Kirkpatrick fects of context are likely to be profound.)
& Davis, 1994). Our own work on the norma-
23
Seek to understand contextual and situational Pursue clinical implications. As reviewed briefly
influences. Nearly a half-century of research above, and by scores of other scholars in recent
makes clear that personality is most stable decades (Cacioppo et al., 2002; Coyne et al.,
within classes of situations as opposed to 2001; Flinn & England, 1997; Harrison, Wil-
across situations (Mischel, Shoda, & Mendoza- liams, Berbaum, Stem, & Leeper, 2000; House,
Denton, 2002). The question can reasonably be Landis, & Umberson, 1988; Kawachi, 2001;
asked: Is a given individual secure in her rela- Kim & McKenry, 2002; Robles & Kiecolt-
tionship with her spouse to the same degree as Glaser, 2003; Uchino et al., 1996; Uvnaes-
she is in her relationship with her best friend Moberg, 1998), social relationships hold major
mother, or sister? Moreover, does her attach- implications for health and well-being. As the
ment style manifest in the same way to a threat neural mechanisms supporting these effects
to her relationship as it does to her personal become better known, it may be possible to
sense of bodily harm? Would she have en- implement clinical interventions that not only
dorsed the same level of security during her emphasize the forging and maintenance of
last relationship as she does in her current one? close relationships, but that also focus on the
Some studies suggest that within-person varia- use of social affect regulation for clinical pur-
tion in attachment style across different rela- poses.
tionships is substantial (La Guardia, Ryan,
Couchman, & Deci, 2000). This is likely to be For example, it may be possible to use certain
especially true at the neural level, where meas- relationship interventions (see Johnson, Chap-
ures can be very sensitive to small changes in ter 33, this volume) to transform couples that
context. do not show a strong social regulation effect on
neural threat responding into those that do.
Implement longitudinal designs. One extremely Johnson (2002) has already used attachment-
important problem for the neuroscience of related marital interventions to help with the
attachment is delineating the process by which treatment of post-traumatic stress disorder.
two individuals progress from not being at-
tached to being attached (see Zeifman & It warrants emphasis here that most stress-
Hazan, Chapter 20, this volume for a discus- reduction techniques involve highly individual-
sion of this issue). What is the rate at which ized activities (e.g., cognitive behavioral therapy,
this typically occurs? How is this affected by mindfulness meditation) that may be less effi-
attachment style? With special relevance to the cient or more costly than using social networks
present chapter, which neural structures associ- or attachment relationships in the implementa-
ated with emotional responding, motivation, tion of affect-regulation strategies. Few or no
and emotion regulation are particularly sensitive interventions are designed with this specifically
to this process? For example, at what point, or in mind, and even those that are rarely if ever
with what kinds of interpersonal experiences, offer training in how to allow oneself to be
does a stranger who regulates the brain’s auto- soothed by another person.
nomic and musculoskeletal response to threat Finally, the careful delineation of neural sys-
become a partner who regulates additional neu- tems underlying attachment stands to expand
ral processes related to effortful affect regula- our basic understanding of a wide variety of
tion and threat vigilance? Longitudinal studies disorders that implicate social processes. The
may also address questions of within-subject potential exists for this work to inform research
variation in attachment style over both time and on disorders ranging from autism to fragile X
relationships. syndrome, Williams syndrome, depression, so-
24
cial anxiety, schizophrenia, and virtually all of References
the personality disorders (most or all of which
are more or less defined in terms of social be- Addis, D. R., Wong, A. T., & Schacter, D. L. (2007). Remember-
haviors). ing the past and imagining the future: common and dis-
tinct neural substrates during event construction and
Differentiate behavioral from neural systems. A major elaboration. Neuropsychologia, 45, 1363-1377.
challenge to future neuroscientists interested in
Adolphs, R., Baron-Cohen, S., & Tranel, D. (2002). Impaired
the study of attachment will be the temptation
recognition of social emotions following amygdala dam-
to think of the attachment behavioral system as
age. Journal of Cognitive Neuroscience, 14, 1264- 1274.
a unitary neural construct, which it almost cer-
tainly is not. A host of neural processes, each Adolphs, R., & Tranel, D. (2003). Amygdala damage impairs
with its own unique problems to solve, con- emotion recognition from scenes only when they contain
tribute to what we have come to call the at- facial expressions. Neuropsychologia, 41, 1281- 1289.
tachment behavioral system, and the attach-
Adolphs, R., Tranel, D., & Damasio, A. R. (1998). The human
ment behavioral system may indeed be little
amygdala in social judgment. Nature, 393, 470-474.
more than a convenient rubric for describing
the collective social activities of social bonding Agmo, A., Barreau, S., & Lemaire, V. (1997). Social motivation
and social affect regulation. On the other hand, in recently weaned rats is modified by opiates. Developmen-
the attachment behavioral system may repre- tal Neuroscience, 19, 505-520.
sent an emergent property of its constituent
neural components that is, under some condi- Aharon, I., Etcoff, N., Ariely, D., Chabris, C. F., O'Connor, E.,
& Breiter, H. C. (2001). Beautiful faces have variable re-
tions and in some situations, relatively irreduci-
ward value: fMRI and behavioral evidence. Neuron, 32,
ble.
537-551.
Collaborate. The neuroscience of attachment Ainsworth, M. D. S. (1989). Attachments beyond infancy.
represents uncommonly fertile ground for a American Psychologist, 44, 709-716.
wide variety of researchers, from neuroscien-
tists to psychologists, biologists, physicians, Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S.
epidemiologists, and others. Individuals from (1978). Patterns of attachment: A psychological study of
diverse scientific traditions stand to contribute the strange situation. Hillsdale, N.J: Erlbaum.
many essential pieces to this fundamentally im-
Allen, J. P., McElhaney, K. B., Land, D. J., Kuperminc, G. P.,
portant puzzle. Because this area is so necessar- Moore, C. M., O'Beirne-Kelley, H., & Liebman, S. M.
ily multidisciplinary, researchers interested in (2003). A Secure Base in Adolescence: Markers of At-
these and related questions will do well to ex- tachment Security in the Mother-Adolescent Relationship.
plore contacts in related disciplines as their par- Child Development, 74, 292-307.
ticular research questions call for it (Cacioppo
et al., 2007). It is for precisely this reason that Aragona, D. J., Liu, Y., Yu, Y. J., Curtis, J. T., Detwiler, J. M.,
collaborations are increasingly the norm among Insel, T. R., & Wang, Z. (2006). Nucleus accumbens do-
the social, cognitive, and affective neurosci- pamine differentially mediates the formation and mainte-
nance of monogamous pair bonds. Nature Neuroscience, 9,
ences. Such collaborations enrich the science,
134-139.
and often richly reward the scientists who take
part. When such efforts are focused on a ques- Balfour, M. E., Yu, L., & Coolen, L. M. (2004). Sexual behavior
tion as fundamentally important as the neuro- and sex-associated environmental cues activate the meso-
science of attachment, it is expected that col- limbic system in male rats. Neuropsychopharmacology, 29, 718-
laborative efforts will be embraced with great 730.
enthusiasm.
25
Bar, M. (2007). The proactive brain: using analogies and asso- Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation:
ciations to generate predictions. Trends in Cognitive Sci- Anxiety and anger. New York, NY: Basic Books.
ence, 11, 372.
Bowlby, J. (1979). The Making and Breaking of Affectional
Barber, B. L., Eccles, J. S., & Stone, M. R. (2001). Whatever Bonds. New York: Routledge.
happened to the jock, the brain, and the princess? Young
adult pathways linked to adolescent activity involvement Bowlby, J. (1980). Attachment and Loss: Vol. 3. Separation:
and social identity. Journal of Adolescent Research, 16(5), Sadness and Depression. New York, NY: Basic Books.
429-455.
Bozarth, M. A., & Wise, R. A. (1996). Toxicity associated with
Bard, K. A. (1983). The effect of peer separation in young long-term intravenous heroin and cocaine self- administra-
chimpanzees (Pan troglodytes). American Journal ofPri- tion in the rat. Journal of the American Medical Associa-
matology, 5, 25-37. tion, 254, 81-83.
Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles Brasted, P. J., Bussey, T. J., Murray, E. A., & Wise, S. P. (2003).
among young adults: A test of a four-category model. Role of the hippocampal system in associative learning
Journal of Personality and Social Psychology, 61, 226- beyond the spatial domain. Brain, 126, 1202-1223.
244.
Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report
Benuzzi, F., Pugnaghi, M., Meletti, S., Lui, F., Serafini, M., measurement of adult attachment: An integrative over-
Baraldi, P., & Nichelli, P. (2007). Processing the socially view. In J. A. Simpson & W. S. Rholes (Eds.), Attachment
relevant parts of faces. Brain Research Bulletin, 74, 344- theory and close relationships. (pp. 46-76). New York, NY:
356. Guilford Press.
Berardi, N., & Maffei, L. (1999). From visual experience to Brewer, M. B., & Caporeal, L. R. (1990). Selfish genes vs. selfish
visual function: roles of neurotrophins. Journal of Neurobi- people: Sociobiology as origin myth. Motivation & Emo-
ology, 41, 119-126. tion, 14(237-243).
Berridge, K. C. (2007). The debate over dopamine's role in Briones, T. L., Klintsova, A. Y., & Greenough, W. T. (2004).
reward: the case for incentive salience. Psychopharmacology, Stability of synaptic plasticity in the adult rat visual cortex
191, 391-431. induced by complex environment exposure. Brain Re-
search, 1018, 130-135.
Berscheid, E. (2003). The human's greatest strength: Other
humans. In U. M. Staudinger (Ed.), A psychology of human Buchheim, A., Erk, S., George, C., Kächele, H., Ruchsow, M.,
strengths: Fundamental questions and future directions for a positive Spitzer, M., Tilo, K., & Henrik, W. (2006). Measuring at-
psychology (pp. 37-47). Washington D.C.: American Psycho- tachment representation in an fMRI environment: A pilot
logical Association. study. Psychopathology, 39, 144-152.
Boccia, M. L., Reite, M., & Laudenslager, M. (1989). On the Burgdorf, J., & Panksepp, J. (2001). Tickling induces reward in
physiology of grooming in a pigtail macaque. Physiology adolescent rats. Physiology & Behavior, 72, 167-173.
& Behavior, 45, 667-670.
Buss, D. M., & Kenrick, D. T. (1998). Evolutionary social psy-
Borman-Spurrell, E., Allen, J. P., Hauser, S. T., Carter, A., & chology. In D. T. Gilbert, S. T. Fiske & G. Lindzey (Eds.),
Cole-Detke, H. C. (1995). Assessing adult attachment: A The handbook of social psychology (pp. 982- 1026). New
comparison of interview-based and self-report methods. York: McGraw Hill.
Unpublished Manuscript, Yale University, New Haven,
Cacioppo, J. T., Amaral, D. G., Blanchard, J. J., Cameron, J. L.,
CT.Unpublished manuscript.
Carter, C. S., Crews, D., Fiske, S., Heatherton, T., Johnson,
Bowlby, J. (1969/1982). Attachment and loss: Vol. 1 (2nd. ed.). M. K., Kozak, M. J., Levenson, R. W., Lord, C., Miller, E.
New York, NY: Basic Books. K., Ochsner, K., Raichle, M. E., Shea, M. T., Taylor, S. E.,
Young, L. J., & Quinn, K. J. (2007). Social neuroscience:
26
Progress and implications for mental health. Perspectives neural systems underlying response to threat. Psychophysiol-
on Psychological Science, 2, 99–123. ogy, 42, S44.
Cacioppo, J. T., Hawkley, L. C., Crawford, L. E., Ernst, J. M., Coan, J. A., Schaefer, H. S., & Davidson, R. J. (2006b). Lending
Burleson, M. H., Kowalewski, R. B., Malarkey, W. B., Van a hand: Social regulation of the neural response to threat.
Cauter, E., & Berntson, G. G. (2002). Loneliness and Psychological Science, 17, 1032-1039.
Health: Potential Mechanisms. Psychosom Med, 64(3),
407-417. Coyne, J. C., Rohrbaugh, M. J., Shoham, V., Sonnega, J. S., Nick-
las, J. M., & Cranford, J. A. (2001). Prognostic importance
Cahill, L., Prins, B., Weber, M., & McGaugh, J. L. (1994). Beta- of marital quality for survival of congestive heart failure.
adrenergic activation and memory for emotional events. American Journal of Cardiology, 88, 526- 529.
Nature, 371, 702-704.
Darwin, C. (1872/1998). The Expression of Emotions in Man
Cancedda, L., Putignano, E., Sale, A., Viegi, A., Berardi, N., & and Animals. Oxford: Oxford University Press.
Maffei, L. (2004). Acceleration of visual system develop-
ment by environmental enrichment. Journal of Neurosci- Davidovitz, R., Mikulincer, M., Shaver, P. R., Ijzak, R., & Pop-
ence, 24, 4840-4848. per, M. (2007). Leaders as attachment figures: Their at-
tachment orientations predict leadership-related mental
Carter, C. S. (2003). Developmental consequences of oxytocin. representations and followers’ performance and mental
Physiology & Behavior, 79, 383-397. health. Journal of Personality and Social Psychology, 93,
632-650.
Carter, C. S., & DeVries, A. C. (1999). Stress and soothing : An
endocrine perspective. In M. Lewis & D. Ramsay (Eds.), Davidson, R. J., & Irwin, W. (1999). The functional neuroanat-
Soothing and Stress (pp. 3-18). Mahwah, NJ: Lawrence omy of emotion and affective style. Trends in Cognitive
Erlbaum Associates. Sciences, 3, 11-21.
Carter, C. S., & Keverne, E. B. (2002). The neurobiology of Dawson, G., Ashman, S. B., Hessl, D., Spieker, S., Frey, K.,
social affiliation and pair bonding. Hormones, Brain & Panagiotides, H., & Embry, L. (2001). Autonomic and
Behavior, 1, 299-337. brain electrical activity in securely- and insecurely- at-
tached infants of depressed mothers. Infant Behavior and
Chugani, H. T. (1998). A critical period of brain development: Development, 24, 135-149.
studies of cerebral glucose utilization with PET. Preven-
tive Medicine, 27, 184-188. DeCasper, A. J., & Fifer, W. P. (1980). Of human bonding:
newborns prefer their mothers' voices. Science, 208, 1174-
Coan, J. A., & Allen, J. J. B. (2003). The state and trait nature of 1176.
frontal EEG asymmetry in emotion. In K. Hugdahl & R.
J. Davidson (Eds.), The Asymmetrical Brain (pp. 565-615). Depue, R. A., & Collins, P. F. (1999). Neurobiology of the
Cambridge, MA: MIT Press. structure of personality: dopamine, facilitation of incen-
tive motivation, and extraversion. Behavioral and Brain
Coan, J. A., & Allen, J. J. B. (2004). Frontal EEG Asymmetry as Sciences, 22, 491-517.
a Moderator and Mediator of Emotion. Biological Psy-
chology, 67, 7-49. Depue, R. A., & Morrone-Strupinsky, J. V. (2005). A neurobe-
havioral model of affiliative bonding: Implications for
Coan, J. A., Allen, J. J. B., & McKnight, P. E. (2006a). A capabil- conceptualizing a human trait of affiliation. Behavioral
ity model of individual differences in frontal EEG asym- and Brain Sciences, 28, 313– 395.
metry. Biological Psychology, 72, 198-207.
Diamond, L. M. (2001). Contributions of psychophysiology to
Coan, J. A., Schaefer, H. S., & Davidson, R. J. (2005). Marital research on adult attachment: Review and recommenda-
adjustment and interpersonal styles moderate the effects tions. Personality and Social Psychology Review, 5, 276-
of spouse and stranger hand holding on activation of 296.
27
Edens, J. L., Larkin, K. T., & Abel, J. L. (1992). The effect of Hamann, S. B., Ely, T. D., Grafton, S. T., & Kilts, C. D. (1999).
social support and physical touch on cardiovascular reac- Amygdala activity related to enhanced memory for pleas-
tions to mental stress. Journal of Psychosomative Re- ant and aversive stimuli. Nature neuroscience, 2, 289-293.
search, 36, 371-382.
Hansen, S., Bergvall, A. H., & Nyiredi, S. (1993). Interaction
Ellenbogen, M. A., Schwartzman, A. E., Stewart, J., & Walker, with pups enhances dopamine release in the ventral stria-
C. D. (2006). Automatic and effortful emotional informa- tum of maternal rats: a microdialysis study. Pharmacology,
tion processing regulates different aspects of the stress Biochemistry & Behavior, 45, 673-676.
response. Psychoneuroendocrinology, 31, 373- 387.
Hansen, S., Harthon, C., Wallin, E., Löfberg, L., & Svensson,
Flinn, M. V., & England, B. G. (1997). Social economics of K. (1991). The effects of 6-OHDA-induced dopamine
childhood glucocorticoid stress response and health. depletions in the ventral or dorsal striatum on maternal
Amerrican Journal of Physical Antropology, 102, 33- 53. and sexual behavior in the female rat. Pharmacology, Bio-
chemistry & Behavior, 39, 71-77.
Fraley, R. C., & Shaver, P. R. (2000). Adult romantic attach-
ment: Theoretical developments, emerging controversies, Harrison, L. L., Williams, A. K., Berbaum, M. L., Stem, J. T., &
and unanswered questions. Review of General Psychol- Leeper, J. (2000). Physiologic and behavioral effects of
ogy, 4, 132-154. gentle human touch on preterm infants. Research in Nurs-
ing & Health, 23(6), 435-446.
Franceschini, M. A., Thaker, S., Themelis, G., Krishnamoorthy,
K. K., Bortfeld, H., Diamond, S. G., Boas, D. A., Arvin, Hebb, D. O. (1949). The Organization of Behavior. New York:
K., & Grant, P. E. (2007). Assessment of infant brain Wiley.
development with frequency-domain near-infrared spec-
troscopy. Pediatric Research, 61, 546-551. Heinrichs, M., Meinlschmidt, G., Neumann, I., Wagner, S.,
Kirschbaum, C., Ehlert, U., & Hellhammer, D. H. (2001).
Gainer, H., & Wray, S. (1994). Cellular and molecular biology of Effects of suckling on hypothalamic-pituitary- adrenal
oxytocin and vasopressin. In E. Knobil & J. Neill (Eds.), axis responses to psychosocial stress in postpartum lactat-
The Physiology of Reproduction. New York: Raven Press. ing women. The Journal of Clinical Endocrinology &
Metabolism, 86, 4798-4804.
Galliot, M. T., & Baumeister, R. F. (2007). The physiology of
willpower: Linking blood glucose to self-control. Personal- Herschkowitz, N. (2000). Neurological bases of behavioral
ity and Social Psychology Review, 11, 303-327. development in infancy. Brain & Development, 22, 411-
416.
Gillath, O., Bunge, S. A., Shaver, P. R., Wendelken, C., &
Mikulincer, M. (2005). Attachment-style differences and Hofer, M. A. (1995). Hidden regulators: Implications for a new
ability to suppress negative thoughts: Exploring the neural understanding of attachment, separation, and loss. In S.
correlates. NeuroImage, 28, 835-847. Goldberg, R. Muir & J. Kerr (Eds.), Attachment theory:
Social, developmental, and clinical perspectives. Hillsdale,
Gosling, S. D., & John, O. P. (1999). Personality dimensions in NJ: Analytic Press, Inc.
non-human animals: A cross-species review. Current Di-
rections in Psychological Science, 8, 69-75. Hofer, M. A. (2006). Psychobiological roots of early Attach-
ment. Current Directions in Psychological Science, 15, 84-
Graves, F. C., Wallen, K., & Maestripieri, D. (2002). Opioids 88. Hofer, M. A., & Sullivan, R. M. (2001). Toward a neu-
and attachment in rhesus macaque (Macaca mulatta) abu- robiology of attachment. In C. A. Nelson & M. Luciana
sive mothers. Behavioral Neuroscience, 116, 489- 493. (Eds.), Handbook of Developmental Cognitive Neurosci-
ence (pp. 599–616). Cambridge: MIT Press.
Gust, D. A., Gordon, T. P., Brodie, A. R., & McClure, H. M.
(1996). Effect of companions in modulating stress associ- Holloway, K. S., Cornil, C. A., & Balthazart, J. (2004). Effects of
ated with new group formation in juvenile rhesus ma- central administration of naloxone during the extinction
caques. Physiology & Behavior, 59, 941-945. of appetitive sexual responses. Behavioural Brain Re-
search, 153, 567-572.
28
House, J. S., Landis, K. R., & Umberson, D. (1988). Social rela- Kim, H. K., & McKenry, P. C. (2002). The relationship between
tionships and health. Science, 241(4865), 540-545. marriage and psychological well-being: A Longitudinal
analysis. Journal of Family Issues, 23(8), 885-911.
Hubel, D. H., & Wiesel, T. N. (1970). The period of suscepti-
bility to the physiological effects of unilateral eye closure King, J. E., & Figueredo, A. J. (1997). The five-factor model
in kittens. The Journal of Physiology, 206, 419-436. plus dominance in chimpanzee personality. Journal of
Research in Personality, 31, 271-271.
Insel, T. R., & Fernald, R. D. (2004). How the brain processes
social information: Searching for the social brain. Annual Kirkpatrick, L. A., & Davis, K. E. (1994). Attachment style,
Review of Neuroscience, 27, 697-722. gender, and relationship stability: A longitudinal analysis.
Journal of Personality and Social Psychology, 66, 502-512.
Insel, T. R., & Winslow, J. T. (1998). Serotonin and neuropep-
tides in affiliative behaviors. Biological Psychiatry, 44, 207- Kiyokawa, Y., Kikusui, T., Takeuchi, Y., & Mori, Y. (2004).
219. Partner’s stress status influences social buffering effects in
rats. Behavioral Neuroscience, 118, 798–804.
Izzo, A., Rotondi, M., Perone, C., Lauro, C., Manzo, E., Casilli,
B., Rasile, M., & Amato, G. (1999). Inhibitory effect of Krebs, J. R., & Davies, N. B. (1993). An Introduction to Behav-
exogenous oxytocin on ACTH and cortisol secretion dur- ioural Ecology (3rd ed.). Malden, Ma: Blackwell.
ing labour. Clinical and Experimental Obstetrics and Gy-
necology, 26, 221-224. La Guardia, J. G., Ryan, R. M., Couchman, C. E., & Deci, E. L.
(2000). Within-person variation in security of attachment:
Jacob, S., & McClintock, M. K. (2000). Psychological state A self-determination theory perspective on attachment,
mood effects of steroidal chemosignals in women and need fulfillment, and well-being. Journal of Personality and
men. Hormones and Behavior, 37, 57-78. Social Psychology, 79, 367-384.
Jamner, L. D., & Leigh, H. (1999). Repressive/defensive coping, Laird, J. D., & Strout, S. (2007). Emotional behaviors as emo-
endogenous opioids and health: how a life so perfect can tional stimuli. In J. A. Coan & J. J. B. Allen (Eds.), The
make you sick. Psychiatry Research, 85, 17-31. Handbook of Emotion Elicitation and Assessment (pp.
54-64). New York: Oxford University Press.
Johnson, S. M. (2002). Emotionally Focused Couple Therapy
with Trauma Survivors: Strengthening Attachment Bonds. Le Doux, J. E. (2000). Emotion circuits in the brain. Annual
New York: Guilford. Review of Neuroscience, 23, 155-184.
Kawachi, I. B., L.F. ( 2001). Social Ties and Mental Health Le Moal, M., & Simon, H. (1991). Mesocorticolimbic dopamin-
ergic network: Functional and regulatory roles. Physiologi-
Journal of Urban Health: Bulletin of the New York Academy cal Reviews, 71, 155-234.
of Medicine, Volume 78, Number 3, 1, pp. 458-467(410).
Lewis, M., & Ramsay, D. S. (1999). Effect of maternal soothing
Kemeny, M. E. (2003). The psychobiology of stress. Current on infant stress response. Child Development, 70(1), 11-
Directions in Psychological Science, 12, 124-129. 20.
Kennedy, P. J., & Shapiro, M. L. (2004). Retrieving memories Lim, M. M., Hammock, E. A., & Young, L. J. (2004). The role
via internal context requires the hippocampus. The Jour- of vasopressin in the genetic and neural regulation of
nal of neuroscience, 24, 6979-6985. monogamy. Journal of Neuroendocrinology, 16, 325- 332.
Keverne, E. B., Martensz, N. D., & Tuite, B. (1989). Beta- en- Lim, M. M., & Young, L. J. (2006). Neuropeptidergic regulation
dorphin concentrations in cerebrospinal fluid of monkeys of affiliative behavior and social bonding in animals.
are influenced by grooming relationships. Psychoneuro- Hormones and Behavior, 50, 506-517.
endocrinology, 14, 155-161.
Lorberbaum, J., Newman, J., Dubno, J., Horwitz, A., Nahas, Z.,
Teneback, C., Bloomer, C., Bohning, D., Vincent, D.,
Johnson, M., Emmanuel, N., Brawman-Mintzer, O., Book,
29
S., Lydiard, R., Ballenger, J., & George, M. (1999). Feasibil- troglodytes) to a novel environment. Folia Primatologica,
ity of using fMRI to study mothers responding to infant 47, 128-142.
cries. Depression and Anxiety, 10, 99-104.
Miller, L. C., Bard, K. A., Juno, C. J., & Nadler, R. D. (1990).
Lorenz, K. (1935). Der kumpan in der umwelt des vogels. Behavioral responsiveness to strangers in young chimpan-
Journal of Ornithology, 83, 137-413. zees (Pan troglodytes). Folia Primatologica, 55, 142-155.
Marlier, L., & Schaal, B. (2005). Human newborns prefer hu- Mineka, S., & Suomi, S. J. (1978). Social separation in monkeys.
man milk: conspecific milk odor is attractive without Psychological Bulletin, 85, 1376-1400.
postnatal exposure. Child Development, 76, 155-168.
Mischel, W., Shoda, Y., & Mendoza-Denton, R. (2002). Situa-
McEwan, B. S. (2007). Physiology and neurobiology of stress tion- behavior profiles as a locus of consistency in per-
and adaptation: central role of the brain. Physiological sonality. Current Directions in Psychological Science, 11,
Reviews, 87, 873-904. 50-54.
Melo, A. I., Lovic, V., Gonzalez, A., Madden, M., Sinopoli, K., Moriceau, S., & Sullivan, R. M. (2005). Neurobiology of infant
& Fleming, A. S. (2006). Maternal and littermate depriva- attachment. Developmental Psychobiology, 47, 230-242.
tion disrupts maternal behavior and social- learning of
food preference in adulthood: tactile stimulation, nest Nakamura, S., & Sakaguchi, T. (1990). Development and plas-
odor, and social rearing prevent these effects. Develop- ticity of the locus coeruleus: a review of recent physio-
mental Psychobiology, 48, 209-219. logical and pharmacological experimentation. Progress in
Neurobiology, 34, 505-526.
Meredith, M. (2001). Human vomeronasal organ function: A
critical review of best and worst cases. Chemical Senses, Newcomb, M. D., & Bentler, P. M. (1988). Impact of adoles-
26, 433-445. cent drug use and social support on problems of young
adults: A longitudinal study. Journal of Abnormal Psy-
Mikulincer, M., & Florian, V. (1998). The relationship between chology, 97(1), 64.
adult attachment styles and emotional and cognitive reac-
tions to stressful events. In J. A. Simpson & W. S. Rholes Niesink, R. J., Vanderschuren, L. J., & van Ree, J. M. (1996).
(Eds.), Attachment Theory and Close Relationships (pp. Social play in juvenile rats after in utero exposure to mor-
143-165). New York: Guilford. phine. Neurotoxicology, 17, 905-912.
Mikulincer, M., & Shaver, P. R. (2007). Attachment in adult- Nocjar, C., & Panksepp, J. (2007). Prior morphine experience
hood: Structure, dynamics, and change. New York: Guil- induces long-term increases in social interest and in appe-
ford Press. titive behavior for natural reward. Behavioural Brain Re-
search, 181, 191-199.
Mikulincer, M., Shaver, P. R., & Pereg, D. (2003). Attachment
theory and affect regulation: The dynamics, Noftle, E. E., & Shaver, P. R. (2006). Attachment dimensions
and the big five personality traits: Associations and com-
development, and cognitive consequences of attachment- parative ability to predict relationship quality. Journal of
related strategies. Motivation & Emotion, 27, 77-102. Research in Personality, 40, 179-208.
Milad, M. R., Quinn, B. T., Pitman, R. K., Orr, S. P., Fischl, B., Norris, C. J., Coan, J. A., & Johnstone, I. T. (2007). Functional
& Rauch, S. L. (2005). Thickness of ventromedial prefron- magnetic resonance imaging and the study of emotion. In
tal cortex in humans is correlated with extinction memory. J. A. Coan & J. J. B. Allen (Eds.), The Handbook of Emo-
Proceedings of the National Academy of Sciences, 102, tion Elicitation and Assessment (pp. 440-459). New York:
10706-10711. Oxford University Press.
Miller, L. C., Bard, K. A., Juno, C. J., & Nadler, R. D. (1986). Ochsner, K. N., Bunge, S. A., Gross, J. J., & Gabrieli, J. D.
Behavioral responsiveness of young chimpanzees (Pan (2002). Rethinking feelings: an FMRI study of the cogni-
30
tive regulation of emotion. Journal of Cognitive Neuro- Rolls, E. T. (2007b). The representation of information about
science, 14, 1215-1229. faces in the temporal and frontal lobes. Neuropsychologia,
45, 124-143.
Ochsner, K. N., & Gross, J. J. (2005). The cognitive control of
emotion. Trends in Cognitive Science, 9, 242-249. Ruis, M. A. W., te Brake, J. H. A., Buwalda, B., De Boer, S. F.,
Meerlo, P., Korte, S. M., BlokhuisH, J., & Koolhaas, J. M.
Öhman, A. (2005). The role of the amygdala in human fear: (1999). Housing familiar male wildtype rats together re-
Automatic detection of threat. Psychoneuroendocrinol- duces the long-term adverse behavioural and physiological
ogy, 30, 953-958. effects of social defeat. Psychoneuroendocrinology, 285–
300.
Panksepp, J., Nelson, E., & Siviy, S. (1994). Brain opioids and
mother-infant social motivation. Acta paediatrica, 397, 40- Sapolsky, R. M. (1998). Why Zebras Don't Get Ulcers. New York,
46. NY: Holt.
Pessoa, L., Kastnerb, S., & Ungerleider, L. G. (2002). Atten- Sbarra, D. A. (2006). Predicting the onset of emotional recov-
tional control of the processing of neutral and emotional ery following nonmarital relationship dissolution: Survival
stimuli. Cognitive Brain Research 15, 31-45. analyses of sadness and anger. Personality and Social Psychol-
ogy Bulletin, 32, 298-312.
Pfaus, J. G., Kippin, T. E., & Centeno, S. (2001). Conditioning
and sexual behavior: A review. Hormones and Behavior, Schaal, B., Marlier, L., & Soussignan, R. (1998). Olfactory func-
40, 291-321. tion in the human fetus: evidence from selective neonatal
responsiveness to the odor of amniotic fluid. Behavioral
Phelps, E. A., & LeDoux, J. E. (2005). Contributions of the
Neuroscience, 112, 1438-1449.
amygdala to emotion processing: from animal models to
human behavior. Neuron, 48, 175-187. Schauwers, K., Gillis, S., Daemers, K., De Beukelaer, C., De
Ceulaer, G., Yperman, M., & Govaerts, P. J. (2004). Nor-
Posner, M. I., & Rothbart, M. K. (2007). Research on attention
mal hearing and language development in a deaf- born
networks as a model for the integration of psychological
child. Otology & Neurotology, 25, 924-929.
science. Annual Review of Psychology, 58, 1-23.
Schultz, W., Dayan, P., & Montague, P. R. (1997). A neural sub-
Proffitt, D. R. (2006). Embodied perception and the economy
strate of prediction and reward. Science, 275, 1593- 1599.
of action. Perspectives on Psychological Science, 1, 110-
122. Sierra-Mercado, D. J., Corcoran, K. A., Lebrón-Milad, K., &
Quirk, G. J. (2006). Inactivation of the ventromedial pre-
Quirk, G. J., & Beer, J. S. (2006). Prefrontal involvement in the
frontal cortex reduces expression of conditioned fear and
regulation of emotion: convergence of rat and human
impairs subsequent recall of extinction. The European
studies. Current Opinion in Neurobiology, 16, 723-727.
Journal of Neuroscience, 24, 1751-1758.
Reichardt, L. F. (2006). Neurotrophin-regulated signalling
Skuse, D., Albanese, A., Stanhope, R., Gilmour, J., & Voss, L.
pathways. Philosophical transactions of the Royal Society
(1996). A new stress-related syndrome of growth failure
of London, 361, 1545-1564.
and hyperphagia in children, associated with reversibility
Robles, T. F., & Kiecolt-Glaser, J. K. (2003). The physiology of of growth-hormone insufficiency. Lancet, 348, 353-358.
marriage: Pathways to health. Physiology & Behavior, 79,
Sullivan, R. M. (2003). Developing a Sense of Safety: The Neu-
409-416.
robiology of Neonatal Attachment. Annals of the New
Rolls, E. T. (2007a). Emotion elicited by primary reinforcers York Academy of Sciences, 1008, 122-131.
and following stimulus-reinforcement association learning.
Teicher, M. H., Samson, J. A., Polcari, A., & McGreenery, C. E.
In J. A. Coan & J. J. B. Allen (Eds.), The Handbook of
(2006). Sticks, stones, and hurtful words: relative effects of
Emotion Elicitation and Assessment (pp. 137-157). New
various forms of childhood maltreatment. American
York: Oxford University Press.
Journal of Psychiatry, 163, 993-1000.
31
Terranova, M. L., Cirulli, F., & Laviola, G. (1999). Behavioral dala responsivity to masked fearful eye whites. Science,
and hormonal effects of partner familiarity in periadoles- 306, 2061.
cent rat pairs upon novelty exposure. Psychoneuroendo-
crinology, 24, 639–656. Whalen, P. J., Rauch, S. L., Etcoff, N. L., McInerney, S. C., Lee,
M. B., & Jenike, M. A. (1998). Masked presentations of
Thomas, K. M., Drevets, W. C., Whalen, P. J., Eccard, C. H., emotional facial expressions modulate amygdala activity
Dahl, R. E., Ryan, N. D., & Casey, B. J. (2001). Amygdala without explicit knowledge. The Journal of Neuroscience,
response to facial expressions in children and adults. Bio- 18, 411-418.
logical Psychiatry, 49, 309-316.
Wiedenmayer, C. P., Magarinos, A. M., McEwen, B. S., & Barr,
Topal, J., Miklosi, A., Csanyi, V., & Doka, A. (1998). Attach- G. A. (2003). Mother lowers glucocorticoid levels of
ment behavior in dogs (canis familiaris): A new application preweaning rats after acute threat. Annals of the New
of Ainsworth's (1969) strange situation test. Journal of York Academy of Sciences, 1008, 304-307.
Comparative Psychology, 112, 219-229.
Williamson, A. (2006). Using self-report measures in neurobe-
Tzschentke, T. M., & Schmidt, W. J. (2000). Functional relation- havioral toxicology: Can they be trusted? Neurotoxicol-
ship among medial prefrontal cortex, nucleus accumbens, ogy, 28, 227-234.
and ventral tegmental area in locomotion and reward.
Critical reviews in neurobiology, 14, 131- 142. Windle, R. J., Shanks, N., Lightman, S. L., & Ingram, C. D.
(1997). Central oxytocin administration reduces stress-
Uchino, B. N., Cacioppo, J. T., & Kiecolt-Glaser, J. K. (1996). induced corticosterone release and anxiety behavior in
The relationship between social support and physiological rats. Endocrinology, 138, 2829-2834.
processes: A review with emphasis on underlying mecha-
nisms and implications for health. Psychological Bulletin, 119, Wismer-Fries, A. B., Ziegler, T. E., Kurian, J. R., Jacoris, S., &
488-531. Pollak, S. D. (2005). Early experience in humans is associ-
ated with changes in neuropeptides critical for regulating
Uvnaes-Moberg, K. (1998). Oxytocin may mediate the benefits social behavior. Proceedings of the National Academy of
of positive social interaction and emotions. Psychoneuroen- Science, 102, 17237-17240.
docrinology, 23, 819-835.
Young, L. J., & Wang, Z. (2004). The neurobiology of pair
Weaver, A., & de Waal, F. B. M. (2003). The mother–offspring bonding. Nature Neuroscience, 7, 1048-1054.
relationship as a template in social development: Recon-
ciliation in captive brown capuchins (cebus apella). Journal
of Comparative Psychology, 117, 101- 110.
Weaver, I. C. G., Cervoni, N., Champagne, F. A., D'Alessio, A.

C., Sharma, S., Seckl, J. R., Dymov, S., Szyf, M., & Meaney,
M. J. (2004). Epigenetic programming by maternal behav-
ior. Nature Neuroscience, 7, 847-854.
Weiss, S. J. (1990). Effects of differential touch on nervous

system arousal of patients recovering from cardiac disease.
Heart Lung, 19, 474-480.
Whalen, P. J. (in press). The uncertainty of it all. Trends in Cogni-

tive Science.
Whalen, P. J., Kagan, J., Cook, R. G., Davis, F. C., Kim, H.,
Polis, S., McLaren, D. G., Somerville, L. H., McLean, A.
A., Maxwell, J. S., & Johnstone, T. (2004). Human amyg-
32
View publication stats

Toward A Neuroscience of Attachment

Uploaded by

Copyright:

Available Formats

Toward A Neuroscience of Attachment

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Toward A Neuroscience of Attachment

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Toward a Neuroscience of Attachment

Chapter · August 2008

Neural Mechanisms involved in Emotion, Depression, and/or Anxiety View project

Takeover Transitions in Automated Vehicles(Safe-SCAD) View project

The user has requested enhancement of the downloaded file.

FROM THE HANDBOOK OF ATTACHMENT:

THEORY, RESEARCH, AND CLINICAL IMPLICATIONS

Pages 241 - 265

The Guilford Press, NY

Weaver, I. C. G., Cervoni, N., Champagne, F. A., D'Alessio, A.

Weiss, S. J. (1990). Effects of differential touch on nervous

Whalen, P. J. (in press). The uncertainty of it all. Trends in Cogni-

View publication stats

You might also like