When the Social Self Is Threatened: Shame,

Physiology, and Health

Sally S. Dickerson and Tara L. Gruenewald

University of California, Los Angeles
Margaret E. Kemeny
University of California, San Francisco

ABSTRACT Our program of research focuses on shame as a key

emotional response to ‘‘social self’’ threats (i.e., social evaluation or re-
jection). We propose that shame may orchestrate specific patterns of
psychobiological changes under these conditions. A series of studies dem-
onstrates that acute threats to the social self increase proinflammatory
cytokine activity and cortisol and that these changes occur in concert with
shame. Chronic social self threats and persistent experience of shame-re-
lated cognitive and affective states predict disease-relevant immunological
and health outcomes in HIV. Across our laboratory and longitudinal
studies, general or composite affective states (e.g., distress) are unrelated
to these physiological and health outcomes. These findings support a
stressor- and emotional response-specificity model for psychobiological
and health research.

Sally S. Dickerson, Department of Psychology and Social Behavior, University of

California, Irvine; Tara L. Gruenewald, Department of Medicine\Division of Ger-
iatrics, David Geffen School of Medicine, University of California, Los Angeles;
Margaret E. Kemeny, Health Psychology Program and Department of Psychiatry,
University of California, San Francisco. Preparation of this paper was supported in
part by a Health Psychology Training Grant Pre-doctoral Fellowship from the Na-
tional Institute of Mental Health, a National Science Foundation Graduate Fellow-
ship, a Ruth L. Kirchstein National Research Service Award, and funding from the
National Institute of Mental Health (MH42918). We thank Shelly Gable for her
thoughtful comments on previous versions of this paper. Correspondence concerning
this article should be addressed to: Sally S. Dickerson, Department of Psychology and
Social Behavior, University of California, Irvine, 3340 Social Ecology II, Irvine, CA
92697-7085. E-mail: sdickers@uci.edu.

Journal of Personality 72:6, December 2004.

Blackwell Publishing 2004
1192 Dickerson, Gruenewald, Kemeny

A voluminous literature demonstrates that stressors can influence

physiological and health outcomes. It has been commonly proposed
that affective responses may mediate these associations; key ques-
tions, however, remain about the nature of the links between stress-
ors, affect, physiology, and health.
The prevailing notion regarding these relationships can be de-
scribed as a generality model (Kemeny, 2003). This model follows
from the work of Hans Selye (1956), who argued that all stressors
activate a generalized physiological ‘‘stress’’ response. Therefore, a
variety of stressors (e.g., physical, psychological) would elicit a com-
mon set of physiological effects. More recently, affect has been in-
corporated into the generality model, in that stressful life events are
thought to trigger the experience of distress, and distress then alters
physiological processes that increase vulnerability to disease. Be-
cause generalized distress reactions are examined rather than distinct
emotional states (e.g., fear, shame, sadness), this model is ‘‘nonspe-
cific’’ in terms of the relationships posited between physiology, the
eliciting conditions of the stressor, and the emotional responses en-
gendered.
There is good reason, however, to suspect the existence of stressor
and emotional response specificity with regard to the stressor-affect–
health relationship. We have advocated the utility of an integrated
specificity model (Kemeny, 2003), which is grounded in several
basic premises. First, the nature of the stressor determines the
specific psychobiological responses evoked; stressors are not
interchangeable. Escaping a predator, coping with irrevocable
loss, or maintaining one’s acceptance in a social group all require
different sets of psychological, physiological, and behavioral
changes to appropriately react under these diverse circumstances.
Therefore, distinct patterns of psychobiological responses are
necessary to adaptively cope with threats to different goals (e.g.,
Weiner, 1992).
Second, emotions are thought to play a key role in orchestrating
these coordinated responses to different goal threats. Specific threats
elicit prototypical emotional states; for example, one can experience
fear when one’s safety or survival is threatened, sadness following
the death of a loved one, and shame in response to social rejection.
These emotions are associated with experiential, behavioral, and
physiological components that provide a coordinated, adaptive re-
sponse to these specific threats and/or opportunities (e.g., Ekman,
Shame, Physiology, and Health 1193

1999; Keltner & Gross, 1999; Levenson, 1994; Tooby & Cosmides,
1990; but see Russell, 2003,for an alternative perspective).
If different goal threats elicit specific patterns of emotional and
physiological changes, it follows that discrete emotions may have
distinct physiological correlates. Emerging data support this pre-
mise. For example, brain-imaging studies have demonstrated that
specific emotions are associated with different patterns of central
nervous system activity (e.g., Canli et al., 2001; Damasio et al., 2000;
Lane, Reiman, Ahern, & Schwartz, 1997). The induction of different
emotions can result in distinctive patterns of activation in peripheral
neural systems such as the autonomic nervous system (e.g., Ekman,
Levenson, & Friesen, 1983). These findings make sense when one
considers that emotions ‘‘do different things’’ and thus require dif-
ferentiated neurophysiological changes to support relevant actions
and shifts in goals.
This integrated specificity approach is often illustrated with
threats to the central goal of physical self-preservation (e.g., safety,
survival), which can trigger the emotion of fear. Fear is thought to
organize the behavioral and physiological changes necessary to ad-
dress this survival threat. These changes could include a shift in the
animal’s motivational state from finding food to finding an avenue
of escape, increases in vigilance to threat-related cues, and focused
attention on identifying and utilizing available resources. In concert
with these changes, the sympathetic nervous system can become ac-
tivated, increasing heart rate and respiratory rate in preparation for
physical exertion. While this pattern of psychobiological changes
may be optimal when physical self-preservation is threatened, threats
to other central goals (e.g., social acceptance/inclusion) may elicit
their own pattern of psychological, physiological, and behavioral
responses that would be adaptive under those specific conditions.
Our program of research has focused on the emotional, physio-
logical, and health consequence of threats to the central goal of pre-
serving one’s ‘‘social self.’’ Our social self preservation model
(Kemeny, Gruenewald, & Dickerson, 2004) proposes that threats
to one’s social self (i.e., threats to one’s social esteem, status, and
acceptance) are accompanied by a specific set of psychological and
physiological responses. These include increases in shame (and other
negative self-evaluative states), proinflammatory cytokine activity,
and cortisol. We argue that these changes are not simply epiphe-
nomena, but instead, that shame and accompanying physiology are
1194 Dickerson, Gruenewald, Kemeny

integral components of a coordinated psychobiological response to

threats to social self preservation, just as fear and its physiological
correlates are components of the response to threats to physical self
preservation.
Although threat-specific emotional and physiological changes
may be functional in an acute context, extreme or persistent expe-
rience of specific threats and corresponding patterns of psychobio-
logical responses may have negative health consequences. Individual
difference factors that increase one’s vulnerability to experience a
specific emotional response, or that amplify threat-induced emo-
tional reactions, may potentiate the adverse physiological effects of
threat experiences. For example, chronic or repeated exposure to
evaluative, rejecting conditions (i.e., social self threats), and accom-
panying increases in shame, proinflammatory cytokine activity, and
cortisol, could lead to negative health outcomes. This may be par-
ticularly likely among individuals with dispositions that increase
their vulnerability to experiencing shame-related cognitions and
emotions. Therefore, the specific set of emotional and physiological
responses that are persistently triggered, and the vulnerability factors
that may increase such responses, may determine the pathways
through which these health effects would occur. Examining specific
emotional and physiological responses to specific eliciting conditions
might elucidate coordinated patterns of psychobiological changes
that could be health-relevant.
In this article, we will present evidence that shame may be a key
affective component of a coordinated psychobiological response to
threats to the social self. We will first review evidence that shame is
elicited when the social self is threatened, and may be associated with
motivational and behavioral changes that are adaptive in these con-
texts. Next, we will present studies from our laboratory that dem-
onstrate that acute threats to the social self are associated with
changes in specific physiological parameters, which may occur in
concert with the emotion of shame. Finally, we will describe a series
of studies in the chronic disease model of HIV. This research dem-
onstrates that chronically experiencing social self threats or accom-
panying negative self-related cognitions and emotions predict
negative long-term immunological and health effects. In describing
our program of research, we hope to underscore how a specificity
approach can reveal distinctive relationships among eliciting condi-
tions, emotions, physiology, and health processes, leading to a more
Shame, Physiology, and Health 1195

comprehensive understanding of the pathways through which psy-

chological states can influence health and disease.

Threats to the Social Self and Shame

Threats to the social self are situations that provide the potential for
a loss of social esteem, social status, or social acceptance and are
characterized by potential or explicit rejection. Prototypical threats
to the social self are conditions in which an important aspect of the
self-identity is, or could be, negatively judged by others (i.e., ‘‘social-
evaluative threat’’; Dickerson & Kemeny, 2004). Social-evaluative
threat can occur in a variety of situations: in performance contexts
that require displays of valued attributes, traits, or abilities in the
presence of others (e.g., competence, intelligence), in rejection-laden
contexts where one is (or could be) judged unworthy of acceptance
or group membership, or in contexts where an uncontrollable char-
acteristic or an unwanted identity is made salient (e.g., stigmatizing
condition).
Shame may be a key affective response to social-evaluative threat,
based in part on evidence that this emotion is preferentially elicited
under these conditions.1 Shame is a self-conscious emotion, experi-
enced when a core aspect of the self is judged as defective, inferior or
inadequate (Gilbert, 1997; Tangney, 1995), and ‘‘is provoked by the
realization that others consider one’s self deficient’’ (Schott, 1979,
p. 1325). Negative self-evaluation is central to shame; this emotion is
associated with negative characterological self-related cognitions,
and its motivational and behavioral correlates are indicative of a
devalued or damaged self (e.g., submission, withdrawal). Therefore,
shame results when perceptions of negative social evaluation are
transformed into negative self-evaluation.
The premise that shame is predicated on social evaluation is not
new, but has had a prolific history within the scientific literature on
emotion. Early influential theorists, such as Darwin (1871/1899),
believed that this emotion ‘‘relates almost exclusively to the judg-
ment of others’’ (p. 114). The symbolic interactionists of the early

1. Other emotions in the shame family, such as embarrassment, are also likely to
be experienced in response to social self threat. Embarrassment, like shame, in-
volves flaws in self-presentation (Schott, 1979), and these emotions are likely to be
experienced together. However, following Gilbert (1997), we view shame as a
prototypical emotion elicited under these specific forms of social threat.
1196 Dickerson, Gruenewald, Kemeny

20th century followed in this tradition, stating ‘‘there is no sense of

‘I,’ as in pride or shame, without its correlative sense of you, or he, or
they’’ (Cooley, 1902/1983, p. 182). Others have also recognized that
negative social evaluation is fundamental to shame, arguing that a
characterological flaw or failure must be publicly revealed to elicit
this emotion (e.g., Ausubel, 1955; Benedict, 1946; Buss, 1980; Gehm
and Scherer, 1988). Even when the audience is not explicit, shame is
still thought to originate from social sources, and is elicited when
individuals evoke an image of a disapproving ‘‘imagined other’’ that
negatively evaluates the self (Lewis, 1971). This rich theoretical tra-
dition resonates with contemporary empirical evidence that negative
social evaluation and appraisals of interpersonal rejection elicit the
emotion of shame (e.g., Keltner & Buswell, 1996; Leary & Baum-
eister, 2000; Smith, Webster, Parrott, & Eyre, 2002; Tangney, 1995;
Tangney, Miller, Flicker, & Barlow, 1996).
The motivational state, nonverbal displays, and behaviors asso-
ciated with shame all support the premise that it is related to sub-
mission and withdrawal. When asked to recall shame-eliciting
events, individuals report wanting to ‘‘hide,’’ ‘‘escape,’’ ‘‘disappear
from view,’’ and ‘‘shrink into the floor,’’ during the experience, in-
dicating the desire to flee the social situation and conceal the ‘‘de-
fective’’ self from social scrutiny (Tangney et al., 1996; Wicker,
Payne, & Morgan, 1983). Shamed individuals also show body pos-
tures that signal a submissive strategy. The characteristic nonverbal
display associated with shame includes head down, slumped posture,
and eye gaze avoidance (Gilbert, 2000; Keltner, 1995), which are
similar to some of the displays that denote submission in primates.
Thus, the motivational state and behavioral displays associated with
shame indicate withdrawal and disengagement.
Theorists have argued that submissive displays and withdrawal
behaviors are social signals that function as an appeasement strategy
to reduce social conflict. Research in primates and other animals
demonstrates that submissive displays serve as social communication
that the animal is not going to contest resources or escalate conflict.
This could signal to the attacker to de-escalate and not induce se-
rious harm, which could facilitate control over aggression and main-
tain social cohesion (Gilbert, 2000; MacLean, 1990). Keltner and
colleagues (Keltner, 1995; Keltner, Young & Buswell, 1997) have
found that shame has similar appeasement functions in humans.
They argue that shame occurs when social rules are violated that
Shame, Physiology, and Health 1197

could potentially disrupt social relations and lead to a heightened

risk of conflict and aggression. Submissive displays indicative of
shame evoke emotions and behavior from the interaction partner(s)
that elicits cooperation and reductions in aggressive or punitive be-
havior (Keltner et al., 1997); therefore, submissive behavior could
help maintain social cohesion and reduce overt hostility in the face of
social threat.
Taken together, this theoretical and empirical evidence suggests
that shame may be more than a non-functional affective conse-
quence of social-evaluative conditions, but may be an integral com-
ponent of a coordinated response to social self threats. This emotion
is elicited under social-evaluative conditions and is sensitive to gra-
dations in negative interpersonal appraisals. Shame may play a role
in organizing motivational and behavioral changes that could be
functional in the face of an antagonistic social encounter, namely
promoting changes associated with withdrawal and disengagement
and nonverbal behavioral displays denoting submission or appease-
ment. Finally, it is possible that shame could orchestrate physiolog-
ical changes that may be adaptive in response to social self threats.

Physiological Responses to Social Self Threats

We propose that social self threats may provide one set of conditions
that can lead to increased activity of the cortisol and proinflamma-
tory systems, and further, that these changes may occur in concert
with the shame family of emotions. In the following section, we
provide an overview of these two systems, and review evidence that
they are elicited in response to social-evaluative threat and may be
co-activated with shame.

Cortisol

Cortisol, a hormone that is an end product of the activation of the

hypothalamic-pituitary-adrenocortical (HPA) axis, plays an impor-
tant regulatory role in normal physiological functioning. Hypo-
thalamic neurons release corticotropin-releasing hormone (CRH),
which in turn stimulates the anterior pituitary to release ad-
renocorticotropin hormone (ACTH). ACTH then triggers the re-
lease of cortisol from the adrenal cortex into the bloodstream, where
it exerts effects on a number of physiological systems. For example,
cortisol plays a primary role in metabolism; it releases energy stores
1198 Dickerson, Gruenewald, Kemeny

(primarily by elevating blood glucose levels), which can then provide

metabolic ‘‘fuel’’ for the body. Cortisol can regulate other physio-
logical systems; for example, cortisol can inhibit aspects of immune
functioning and permits other systems, like the sympathetic nervous
system, to function effectively.
In addition to its role in maintaining ongoing physiological proc-
esses, the HPA system can also be activated in response to certain
stressful circumstances. This is thought to be adaptive, presumably
because cortisol can mobilize energy resources and activate other
physiological systems that are necessary to overcome a threat (Lov-
allo & Thomas, 2000; Sapolsky, Romero, & Munck, 2000). Howev-
er, the HPA system does not appear to be activated indiscriminately
in response to all negative stimuli; instead, stress-induced elevations
in cortisol may be linked to the eliciting conditions or specific char-
acteristics of the stressor (Blascovich & Mendes, 2000; Dickerson &
Kemeny, 2004; Mason, 1968). There is growing evidence that one
class of events that can be potent elicitors of HPA responses is
threats to the social self.
In a meta-analytic review of 208 acute laboratory stressor studies,
we tested whether threats to the social self would be associated with
substantial changes in cortisol (Dickerson & Kemeny, 2004). We
proposed that stressors with social-evaluative threat, in which others
could negatively judge the self (e.g., evaluative audience present),
would elicit greater cortisol responses than tasks without this com-
ponent. We found strong support for this hypothesis; tasks in which
others could negatively evaluate performance were associated with
substantially greater cortisol responses (effect size d 5 0.67) com-
pared to tasks without this social-evaluative component (d 5 0.21).
These findings remained significant when controlling for other meth-
odological factors that predicted cortisol changes (e.g., time of day,
timing of cortisol assessment).
Additionally, cortisol responses were further heightened when so-
cial-evaluative threat was coupled with uncontrollability, in which
participants could not succeed despite their best efforts. Social-eval-
uative, uncontrollable conditions induce a potent threat to the social
self; not only can others evaluate performance, but failure is the
likely outcome. These conditions of exposed failure elicited the larg-
est cortisol responses observed in these laboratory studies (d 5 0.92);
this effect was nearly three times as large as tasks with one compo-
nent alone (i.e., only uncontrollability or only social-evaluative
Shame, Physiology, and Health 1199

threat; d 5 0.32 and 0.35, respectively). In addition, these social-

evaluative, uncontrollable tasks were associated with a slower re-
covery of cortisol to baseline levels. Overall, the meta-analysis pro-
vides strong support for the premise that social-evaluative contexts
that threaten the social self can elicit substantial, persistent changes
in the cortisol system.
A subsample of studies included in the meta-analysis reported
negative affective assessments pre- and poststressor. We calculated
effect sizes on these affective measures to test whether generalized
distress was associated with cortisol responses (Dickerson & Ke-
meny, 2004). Overall, the social-evaluative tasks were not more dis-
tressing than those without this evaluative component. Further,
changes in general negative affect or distress were unrelated to the
magnitude of cortisol responses. This provides support for the
premise that more generalized emotional states may not be related
to changes in the HPA system. As no studies in the meta-analysis
assessed shame, we could not test whether this more specific emotion
was associated with cortisol changes.
In a subsequent laboratory study, we experimentally tested the
premise that tasks with social-evaluative threat would be more likely
to activate the HPA and to induce the experience of shame compared
to tasks without this evaluative component (Gruenewald, Kemeny,
Aziz, & Fahey, in press). Participants were randomly assigned to
give a speech and complete a computerized math task, with elements
of uncontrollability, either alone or in front of an evaluative audi-
ence. There were no differences between the social evaluation and
nonsocial evaluation conditions on post-task ratings of difficulty or
anxiety. However, participants in the social-evaluative condition re-
ported significantly more shame and other self-conscious emotions
than those performing the same tasks alone. Additionally, it was
only the social-evaluative threat condition that induced robust in-
creases in cortisol from pre- to poststressor; those performing the
identical tasks without an audience present showed no changes on
average in cortisol from pre- to posttask. Consistent with our the-
oretical premise, increases in shame and cortisol occurred in concert
under social-evaluative conditions; the experience of shame and oth-
er self-conscious emotions may be the ‘‘active ingredient’’ that led to
cortisol responsivity within the social-evaluative condition.
We recently demonstrated a concordance between shame and co-
rtisol in a study examining the avoidance achievement motive ‘‘fear
1200 Dickerson, Gruenewald, Kemeny

of failure’’ (Dickerson & Gable, 2004). Because this trait has been
theoretically linked with the propensity to experience shame (e.g.,
Atkinson, 1957; Elliot & Thrash, 2004), we examined how fear of
failure influenced shame and cortisol following an evaluative per-
formance task. Consistent with hypotheses, we found that individ-
uals high on this trait experienced more shame and had higher
cortisol levels poststressor than those lower on this trait (Dickerson
& Gable, 2004). However, measures of general negative emotion or
other specific emotions (e.g., anxiety) could not differentiate those
high and low on fear of failure. These findings suggest that individ-
ual difference factors that affect one’s vulnerability to experience
shame also influence the cortisol system, and support the premise
that shame and cortisol may be co-activated under certain evaluative
conditions.
Other studies have found that the behavioral displays that ac-
company shame have been associated with cortisol changes. Chil-
dren who displayed more shame and embarrassment behaviors
during a series of laboratory tasks had greater cortisol changes dur-
ing the session than those with other nonverbal behavioral styles
(Lewis & Ramsay, 2002). Nonverbal shame behaviors observed in
the children included head down, slumped posture, and eye gaze
avoidance—behaviors quite similar to those that denote submission
in nonhuman primates and other social animals (Keltner, 1995;
Gilbert, 1997). It is particularly interesting that specific forms of so-
cial threat in animals (i.e., social defeat, subordination) are typically
associated with concurrent increases in cortisol (e.g., Sapolsky, 1993;
Shively, Laber-Laird, & Anton, 1997), and that the frequency of
submissive behavioral displays correlate with cortisol activity (Hall-
er, Kiem, & Makara, 1996; Shively et al., 1997). Threats to social
status and submissive behavior may provide the animal analogue to
social self threat and shame, and these processes may have been
maintained and elaborated in humans (Gilbert, 1997; Price, Sloman,
Gardner, Gilbert, Rohde, 1994).
Taken together, there is growing support that conditions that
threaten the social self (i.e., social evaluation and rejection) can elicit
cortisol responses. Several studies have shown a concordance be-
tween the conditions and vulnerability factors that specifically in-
crease the shame family of emotions and the conditions and
vulnerabilities that increase cortisol, while this relationship does
not exist for more general negative emotional states or other specific
Shame, Physiology, and Health 1201

emotions. These findings suggest that shame and cortisol could be

two parts of a coordinated response to social self threats.

Proinflammatory cytokine activity

Cytokines are mediators of the immune response. They are chemi-

cal communication molecules released from immune cells that alter
the function of other cells. Cytokines play an important role in cell-
to-cell communication among components of the immune system.
Produced by lymphocytes and monocytes/macrophages, cytokines
have a number of different functions including promoting cell
growth, inhibiting viral replication, and promoting local responses
in the tissue.
A subset of cytokines, called proinflammatory cytokines, initiate
and maintain the inflammatory response. Inflammation is a coordi-
nated bodily response to tissue injury or pathogen infection, which
involves immune cells and their products entering the infected
area, destroying the organism, and repairing injured tissue. Proin-
flammatory cytokines, such as tumor necrosis factor-a (TNFa,) in-
terleukin-6 (IL-6), and interleukin-1b (IL-1b), orchestrate the
activities of different types of immune cells involved in inflamma-
tion. They also are responsible for initiating the processes that lead
to signs of inflammation (e.g., redness, pain, edema).
There is a dynamic interaction between proinflammatory cyto-
kines and cortisol; however, the exact nature of this relationship is
complex. While cortisol can inhibit proinflammatory cytokine pro-
duction, proinflammatory cytokines can activate the HPA axis. Ad-
ditionally, both systems can be activated by other inputs, like the
sympathetic nervous system, so it is possible to have elevations in
both proinflammatory and HPA activity simultaneously.2 In fact,
this is the case under certain situations; the systems can be co-acti-
vated in response to infection or in response to physical threats. One
of the goals of our program of research is to test if threats to the
social self provide another set of conditions that can activate these
two systems in humans.

2. Recent studies have demonstrated that social stress in animals and humans can
lead to glucocorticoid resistance (Rohleder, Schommer, Hellhammer, Engel, &
Kirschbaum, 2001; Avitsur et al., 2001). Reduced downregulation of proinflam-
matory cytokine production in the presence of glucocorticoids following social
stress could explain simultaneous elevations in these systems.
1202 Dickerson, Gruenewald, Kemeny

While biological stimuli (e.g., pathogens, injury) are most com-

monly associated with activation of the proinflammatory cytokine
system, recent evidence from animal models and preliminary work in
humans suggests that certain psychological states and psychological
stressors can be linked to heightened inflammatory activity as well
(e.g., Watkins, Nguyen, Lee, & Maier, 1999). Evidence for a rela-
tionship between psychological states and inflammatory processes
stems primarily from studies of clinical depression. In some studies,
individuals with major depression have shown increased levels of
proinflammatory cytokines relative to non-depressed controls (e.g.
IL-1 and 6; for review, see Maes, 1999). These differences have
emerged even in carefully matched samples, which control for med-
ical condition, medication, health behaviors, and other confounding
factors (e.g., Miller, Stetler, Carney, Freedland, & Banks, 2002).
There is also evidence that pharmacologic treatment for depression
is associated with a reduction in proinflammatory cytokine levels in
clinically depressed patients (Kenis & Maes, 2002).
In animals, research has documented a link between stressor ex-
posure and activation of proinflammatory cytokines. Specifically,
these cytokines appear to be produced in the context of social
threats. The primary model of social threat utilized in this literature
is social disruption stress, in which a very aggressive intruder is
placed in the home cages of other animals, resulting in social defeat
and subordination in the home-caged animals. Animals who are so-
cially defeated show greater production of proinflammatory cyto-
kines (Il-6, Il-1b and TNFa) compared to control animals that were
not socially defeated (Stark et al., 2001; Stark et al., 2002; Avitsur,
Stark, & Sheridan, 2001; Quan et al., 2001). This form of social stress
can induce other immunological changes that are consistent with
greater inflammation, such as increased numbers of circulating gran-
ulocytes and levels of nerve growth factor (Avitsur et al., 2001; Ste-
fanski & Engler, 1998). Thus, specific forms of social threat (i.e.,
social subordination and defeat) cause increases in inflammatory
processes in animals.
Less research has focused on the relationship between stressors
and proinflammatory cytokine activity in humans. While some stud-
ies have investigated how stressful situations influence inflammatory
responses, recent narrative reviews have highlighted the inconsistent
effects in this area of research (Kronfol & Remick, 2000; Watkins
et al., 1999). However, some of the contradictory findings can be
Shame, Physiology, and Health 1203

reconciled when the social-evaluative nature of the task is consider-

ed. A number of different types of laboratory stressors have been
used in these studies; those that utilize social evaluative tasks (e.g.,
public speaking) have shown increases in proinflammatory cytokine
activity (e.g., Ackerman, Martino, Heyman, Moyna, & Rabin, 1998;
Altemus, Rao, Dhabhar, Din, & Granstein, 2001; Goebel, Mills,
Irwin, & Zeigler, 2000). Those that used other tasks (e.g., watching a
film, completing a Stroop task) have not consistently found increases
in inflammatory parameters (e.g., Dugue, Leppanen, Teopp, Fyhr-
quist, & Grasbeck, 1993; Peters et al., 1999; Zakowski, McAllister,
Deal, & Baum, 1992). While tentative, this suggests that social-eval-
uative conditions may have a greater capacity to elicit proinflam-
matory cytokines in humans than the other types of laboratory
stressors examined thus far.
We recently tested whether the negative self-related cognitions
and emotions that can accompany these social-evaluative threats are
associated with inflammatory processes (Dickerson, Kemeny, Aziz,
Kim, & Fahey, 2004). We randomly assigned participants to condi-
tions in which they wrote about an experience of self-blame or a
neutral topic on three separate days over a week period. Participants
in the self-blame condition wrote about a variety of stressful situa-
tions, including experiences of rejection and failing to live up to pa-
rental expectations. The self-blame induction elicited greater
increases in shame and guilt than other emotions measured, and
this increase was observed on each of the three experimental days. In
addition, the receptor for TNF-a (sTNF-RII), a marker of proin-
flammatory cytokine activity, significantly increased from pre- to
post-writing across the three days among the individuals in the
self-blame condition, while there were no changes for those in the
control condition. Furthermore, among the participants in the self-
blame condition, those who reported the greatest increases in shame
also showed the greatest increases in proinflammatory cytokine ac-
tivity, but guilt, anger, anxiety, sadness, and general negative emo-
tion were unrelated to this parameter. Together, these findings
suggest that the experience of self-blame and self-related emotions
can elicit changes in proinflammatory cytokine activity and that
there could be a specific relationship between shame and inflamma-
tory markers.
In the context of social self threat, this association between spe-
cific psychological states and proinflammatory cytokine activity may
1204 Dickerson, Gruenewald, Kemeny

function to promote disengagement and withdrawal. Recent evi-

dence has demonstrated a bidirectional relationship between be-
havior and inflammatory processes. Psychological states and other
stimuli can lead to alterations in proinflammatory cytokines, and
proinflammatory cytokines, in turn, can exert central effects that
influence cognition, affect, motivation, and behavior. For example,
these bidirectional relationships can be seen in the context of infec-
tious illness. In response to a pathogen, animals release proin-
flammatory cytokines, which can have local immune effects. Addi-
tionally, these cytokines can act on the brain to orchestrate processes
directly related to fighting infection (e.g., fever) and to induce a
constellation of behavioral changes called ‘‘sickness behavior.’’ This
behavioral pattern includes reductions in exploratory, sexual, and
social behavior, immobility and locomotor retardation, and decreas-
es in food and water intake—changes that are associated with dis-
engagement and withdrawal from the environment (Dantzer et al.,
2001; Hart, 1988). These behavioral changes are known to result
from the activity of proinflammatory cytokines because peripheral
injections of these proteins can induce sickness behaviors in healthy,
uninfected animals and humans (e.g., Dantzer et al., 2001; Yirmiya
et al., 1999). Therefore, the release of proinflammatory cytokines can
lead to motivational and behavioral changes reflecting disengage-
ment or withdrawal.
This type of behavioral disengagement may be adaptive under
certain conditions or in response to certain stimuli. Cytokine-in-
duced sickness behavior is believed to be an adaptive response to
infection because it reduces energy expenditure (by reducing activity)
so energy resources can be diverted to mounting an effective re-
sponse to the pathogen (Maier & Watkins, 1998). We have proposed
that uncontrollable, social-evaluative threats may provide another
set of conditions under which activation of the proinflammatory
cytokine network and subsequent behavioral disengagement may
also be adaptive (Kemeny et al., 2004). In contrast to controllable
situations of threat to the social self, where employing active be-
havioral responses could successfully circumvent adverse social out-
comes, uncontrollable threats to the social self provide a context
where active social responses may be ineffective and, in some cases,
even exacerbate conflict. Therefore, disengagement, submission, and
withdrawal may be an adaptive response to uncontrollable, social-
evaluative threat.
Shame, Physiology, and Health 1205

Research suggests that these uncontrollable, socially threatening

conditions may result in the release of proinflammatory cytokines
and elicit phenomenological and behavioral responses indicative of
withdrawal or disengagement. The central effects of proinflamma-
tory cytokines may facilitate these behavioral changes. Specifically,
these cytokines may support the withdrawal and disengagement be-
haviors that are observed in animals under social threat (e.g., the
subordinate animal) and in humans experiencing shame in response
to social self threats.

Summary: Psychological and physiological responses to social-evalu-

ative threat

The evidence presented demonstrates that social evaluative con-

texts—those in which an aspect of the self could be judged negatively
by others—may be accompanied by specific psychobiological chang-
es. A growing body of research in the self-related processes and
emotion literature indicates that shame may be experienced partic-
ularly in conditions characterized by negative social evaluation and
rejection. The cortisol and proinflammatory cytokine systems also
appear to be responsive to social-evaluative threat. While tentative,
there is support for the notion that the activation of these systems
under the very specific condition of threat to the social self may
hinge on the experience of shame and related emotions. Further-
more, it is possible that this pattern of psychobiological changes
represent an integrated response, which is adaptive in socially threat-
ening, uncontrollable conditions.

Social Self Threats, Self-Evaluations, and Health

The pattern of psychobiological responses detailed above may be
adaptive in short-term situations of uncontrollable social threat;
however, repeated or prolonged exposure to social self threats could
place individuals at risk for adverse psychological and physical
health outcomes. While little research has examined specific health
effects of chronic threat to the social self, the cognitive, affective, and
physiological sequelae of such threats may be important pathways to
adverse health outcomes. Chronic or persistent shame experiences
may have implications for mental health; shame is thought to be a
central component of a number of psychological conditions, includ-
ing depression (Lewis, 1971; Scheff, 2001), social anxiety (Gilbert &
1206 Dickerson, Gruenewald, Kemeny

Trower, 1990; Schwarzer, 1986), and suicidal ideation (Mokros,

1995). In addition, the tendency to easily experience shame, as well
as more frequent shame experiences, has been shown to predict de-
pressive symptoms cross-sectionally (Tangney, Wagner, & Gram-
zow, 1992) and prospectively (Andrews, Qian, & Valentine, 2002).
Chronic or repeated activation of the HPA and proinflammatory
cytokine network may increase susceptibility to disease or moderate
disease course (e.g., McEwen, 1998; Kronfol & Remick, 2000). El-
evated basal cortisol levels may increase susceptibility to disease in-
cidence and progression through suppression of some components of
immunity (Munck, Guyre, & Holbrook, 1984). Increased activation
of the cortisol and inflammatory systems has been implicated in a
number of disease conditions, including the metabolic syndrome
(Brunner et al., 2002; Ford, 2003) and major depression (Connor &
Leonard, 1998; Gold, Licinio, Wong, & Chrousos, 1995; Maes, 1999;
Maes et al., 1995). Accumulating evidence also associates heightened
inflammatory processes with the incidence and progression of car-
diovascular disease (Black & Garbutt, 2002; Danesh, Collins, Ap-
pleby, & Peto, 1998; Danesh et al., 2000; Pasic, Levy, & Sullivan,
2003), HIV infection (Kedzierska, Crowe, Turville, & Cunningham,
2003), and other chronic inflammatory diseases (e.g., rheumatoid
arthritis; Feldmann, Brennan, & Maini, 1996). Elevated levels of the
proinflammatory cytokine IL-6 in elderly individuals have been
shown to predict increased mortality over extended follow-up peri-
ods (Harris et al., 1999). Thus, there are multiple pathways through
which these physiological systems, which we propose are activated
under conditions of threat to the social self and the experience of
shame, may influence health outcomes.
In a series of studies, we have examined how individual differences
in sensitivity to social self threats, and the self-related cognitions and
emotions that can accompany these threats, predict immunological
and health outcomes in a specific disease model—that of HIV in-
fection. This is an ideal model in which to examine physical health
correlates of repeated or prolonged threats to the social self. The
adults we have studied have a life-threatening, stigmatizing disease,
and, in most of our samples, are gay or bisexual, a stigmatizing sex-
ual orientation. Therefore, they are subject to chronic threat to the
social self and must navigate their way through a world that is often
unaccepting, and sometimes explicitly hostile, regarding a core, un-
controllable aspect of their identity. Additionally, HIV/AIDS is
Shame, Physiology, and Health 1207

characterized by a lengthy but variable progression and known im-

munologic and virologic markers prognostic of accelerated disease
course and mortality (e.g., CD4 T-cell decline, increased viral load;
Fahey et al., 1990; see also Cole & Kemeny, 2001, for overview); this
also makes this an excellent model in which to examine stress-affect–
physiology relationships.
In one line of research, we have found that individuals who are
particularly sensitive to social self threats, especially social rejection,
are more likely to evidence poorer HIV-specific immunological and
health outcomes. In a 9-year longitudinal study, Cole and colleagues
found that individuals who were particularly sensitive to rejection
based on their homosexual identity showed faster CD4 T-cell de-
clines (an important marker of HIV progression) and faster times to
AIDS diagnosis and death, compared to those lower on this trait
(Cole, Kemeny, & Taylor, 1997). The results were quite striking;
highly rejection-sensitive individuals died on average 2 years before
those less sensitive to negative evaluation and rejection. These effects
were observed in a carefully selected group of HIV-positive individ-
uals who were healthy at baseline and when controlling for a variety
of confounding factors (baseline immune status, demographics,
health behaviors, medication usage, etc.).
In a separate sample of HIV-positive gay men, rejection sensitivity
predicted elevated HIV viral load and poorer virologic and immu-
nological outcomes over time (Cole, Kemeny, Zach, Fahey, & Nalib-
off, in press). The sample included men who were not on the highly
effective highly active antiretroviral therapy (HAART) medication
regimen at the baseline point of the study. All were assessed for im-
munological and virological status at baseline and one year later,
during which time they initiated HAART therapy. Those high on
rejection sensitivity showed a weaker response to the medication; the
beneficial reductions in viral load and increases in CD4 T-cells were
significantly less pronounced in the rejection-sensitive men. These
findings suggest that a heightened sensitivity to negative social eval-
uation is associated with physiological states that may have pro-
found consequences for disease progression and mortality.
We have also found parallel associations between perceptions
of rejection and immunological declines in a multi-ethnic sample
of HIV-positive women (Lewis, Kemeny, Myers, & Wyatt, 2004).
In this study, specific components of depression (as measured by
the CES-D) were predicted to be differentially associated with
1208 Dickerson, Gruenewald, Kemeny

immunological outcomes. Consistent with theory, the interpersonal

rejection factor predicted CD4 T-cell declines over a 2-year period,
whereas other components of depression, such as vegetative symp-
toms or depressed affect, did not (controlling for relevant confound-
ing factors). In other words, strong endorsement of items such as ‘‘I
felt people disliked me’’ or ‘‘people were unfriendly,’’ was associated
with poorer immunological outcomes, whereas strong endorsement
of items such as ‘‘I felt sad’’ or ‘‘my sleep was restless’’ was unrelated
to CD4 T-cell decline. Taken together, these investigations indicate
that perceptions of social rejection—or a heightened sensitivity to
social rejection—are associated with immunological states indicative
of poorer disease prognosis, as well as HIV health outcomes and
mortality.
Several other studies from our research group have demonstrated
that the negative self-related cognitions that can be elicited in re-
sponse to social self threats are associated with immune declines.
HIV-positive gay or bisexual men who characteristically blamed
themselves for negative events (i.e., made negative characterological
attributions) had faster declines in CD4 T-cells than those without
this self-blaming attributional style over a 1 1/2-year follow-up pe-
riod (Segerstrom et al., 1996). In a separate sample of HIV-positive,
gay men, the self-reproach aspect of depression predicted CD4 T-cell
declines (Kemeny & Dean, 1995), whereas other components that
lacked a self-blame component (e.g., depressed mood, vegetative
symptoms) were not associated with immunological change. These
studies suggest that these specific cognitions, centered on self-blame,
are associated with negative immunological outcomes in the context
of HIV.
In all of our studies, the findings remain significant when con-
trolling for a host of biobehavioral variables (e.g., demographics,
health behaviors) as well as more general personality variables (e.g.,
neuroticism). Furthermore, general affective states, such as negative
affectivity and depressed mood, do not explain the relationship be-
tween sensitivity to social self threats, negative self-related cognit-
ions, and HIV-related outcomes. This is parallel to the results
obtained in our experimental studies, in which distress or general
negative emotion were unrelated to physiological parameters. How-
ever, as reviewed above, rejection sensitivity, perceptions of rejec-
tion, negative characterological attributions, and self-reproach can
all elicit shame (e.g., Leary & Baumeister, 2000; Weiner, 1985); thus,
Shame, Physiology, and Health 1209

it is possible shame may underlie associations between these con-

structs and HIV outcomes.
To more specifically examine the relationship between shame and
immunological outcomes in HIV, we tested whether chronically ex-
periencing different specific negative emotions was associated with
immune declines among HIV-positive gay and bisexual men. We
found that persistent experience of shame and guilt surrounding HIV
infection predicted CD4 declines over a 7-year follow-up, while other
HIV-specific and general emotions, such as anxiety, anger, sadness,
and negative affectivity, did not (Weitzman, Kemeny, & Fahey,
2004). Ancillary analyses indicated that this relationship hinged on
persistent feelings of shame. These findings support the premise that
shame, experienced in response to possession of a stigmatizing con-
dition, is an important predictor of disease-relevant immunological
change. It is plausible that shame may be an important affective
component of previous findings linking individual differences factors
and negative self-related cognitions to immunological parameters
as well.
Taken together, these findings suggest that how individuals per-
ceive and respond to social self threats are associated with disease
progression and mortality in HIV-infected individuals. The immu-
nological and virological changes assessed in our investigations are
natural consequences of the course of HIV infection. However, our
research suggests that individual difference factors and cognitive re-
sponses relevant to social-evaluative threat, as well as shame-related
affective responses, may moderate the course of HIV disease pro-
gression. Thus, this work supports the potential importance of social
self threats, and their cognitive and affective consequences, on phys-
ical health.

Conclusion
Our research is guided by our theoretical model that posits that
conditions characterized by social evaluation or rejection, or those
that threaten the social self, elicit a coordinated psychobiological
response. We have focused on shame as a key affective component of
this response, which may orchestrate specific patterns of physiolog-
ical and behavioral changes under these conditions. Support for this
model stems from our laboratory studies which demonstrate that
acute social self threats increase proinflammatory cytokine activity
1210 Dickerson, Gruenewald, Kemeny

and cortisol, and these changes occur in concert with shame. In other
work, we have shown that chronic threats to the social self and per-
sistent feelings of shame-related cognitions and emotions predict
disease-relevant immunological and health outcomes in the chronic
disease model of HIV. These two areas of research link together be-
cause extensive research has documented that the proinflammatory
cytokines IL-1b, TNF-a and IL-6 stimulate HIV replication and
predict disease progression (Kedzierska et al., 2003).
Across our laboratory and longitudinal studies, we have found
associations between shame and related self-evaluative states and
physiological and health outcomes; however, more general affective
states, such as distress, have been unrelated to these parameters. This
is consistent with an integrated specificity approach (Kemeny, 2003),
in which specific eliciting conditions are associated with distinct
patterns of emotional and physiological changes. The shame
family of emotions is vastly understudied in psychobiological
research, and only a handful of studies have examined its physio-
logical or health correlates. However, our findings suggest that
shame may be important for elucidating pathways through which
certain stressful conditions influence physiological responses and
health outcomes.

REFERENCES
Ackerman, K. D., Martino, M., Heyman, R., Moyna, N. M., & Rabin, B. S.
(1998). Stressor-induced alteration of cytokine production in multiple sclerosis
patients and controls. Psychosomatic Medicine, 60, 484–491.
Altemus, M., Rao, B., Dhabhar, F. S., Din, W., & Granstein, R. D. (2001). Stress-
induced changes in skin barrier function in healthy women. Journal of Inves-
tigative Dermatology, 117, 309–317.
Andrews, B., Qian, M., & Valentine, J. D. (2002). Predicting depressive symptoms
with a new measure of shame: The Experience of Shame Scale. British Journal
of Clinical Psychology, 41, 29–42.
Atkinson, J. W. (1957). Motivational determinants of risk-taking behavior. Psy-
chological Review, 64, 359–372.
Atkinson, J. W. (Ed.) (1958). Motives in fantasy, action, and society. Princeton,
NJ: Van Nostrand.
Ausubel, D. P. (1955). Relationships between shame and guilt in the socialization
process. Psychological Review, 67, 378–390.
Avitsur, R., Stark, J. L., & Sheridan, J. F. (2001). Social stress induces
glucocorticoid resistance in subordinate animals. Hormones and Behavior, 39,
247–257.
Shame, Physiology, and Health 1211

Benedict, R. (1946). The chrysanthemum and the sword. Boston: Houghton

Mifflin.
Black, P. H., & Garbutt, L. D. (2002). Stress, inflammation and cardiovascular
disease. Journal of Psychosomatic Research, 52 (1), 1–23.
Blascovich, J., & Mendes, W. B. (2000). Challenge and threat appraisals: The
role of affective cues. In J. Forgas (Ed.), Feeling and thinking: The role of
affect in social cognition (pp. 59–82). Cambridge: Cambridge University
Press.
Brunner, E. J., Hemingway, H., Walker, B. R., Page, M., Clarke, P., & Juneja, M.,
et al. (2002). Adrenocortical, autonomic, and inflammatory causes of the
metabolic syndrome: Nested case-control study. Circulation, 106 (21),
2659–2665.
Buss, A. H. (1980). Self-consciousness and social anxiety. San Francisco: W. H.
Freeman and Company.
Canli, T., Zhao, Z., Desmond, J. E., Kang, E., Gross, J., & Gabrieli, J. D. (2001).
An fMRI study of personality influences on brain reactivity to emotional
stimuli. Behavioral Neuroscience, 115 (1), 33–42.
Cole, S. W., & Kemeny, M. E. (2001). Psychosocial influences on the progression
of HIV infection. In R. Ader, D. Felton, & N. Cohen (Eds.), Psychoneuroim-
munology (3rd ed., pp. 583–612). San Diego: Academic Press.
Cole, S. W., Kemeny, M. E., Fahey, J. L., Zack, J. A., & Naliboff, B. (in press).
Psychological risk factors for HIV pathogenesis: Mediation by the autonomic
nervous system. Biological Psychiatry.
Cole, S. W., Kemeny, M. E., & Taylor, S. E. (1997). Social identity and physical
health: Accelerated HIV progression in rejection-sensitive gay men. Journal of
Personality and Social Psychology, 72, 320–335.
Connor, T. J., & Leonard, B. E. (1998). Depression, stress and immunological
activation: the role of cytokines in depressive disorders. Life Sciences, 62 (7),
583–606.
Cooley, C. N. (1902). Human nature and the social order. New York: Scribner.
Damasio, A. R., Grabowski, T. J., Bechara, A., Damasio, H., Ponto, L. L. B., &
Parvizi, J., et al. (2000). Subcortical and cortical brain activity during the feel-
ing of self-generated emotions. Nature Neuroscience, 3, 1049–1056.
Danesh, J., Collins, R., Appleby, P., & Peto, R. (1998). Association of fibrinogen,
C-reactive protein, albumin, or leukocyte count with coronary heart disease:
meta-analyses of prospective studies. Journal of the American Medical Asso-
ciation, 270 (18), 1477–1482.
Danesh, J., Whincup, P., Walker, M., Lennon, L., Thomson, A., & Appleby, P.,
et al. (2000). Low grade inflammation and coronary heart disease: Pro-
spective study and updated meta-analysis. British Medical Journal, 321,
199–204.
Dantzer, R., Bluthe, R., Castanon, N., Cauvet, N., Capuron, L., & Goodall, G.,
et al. (2001). Cytokine effects on behavior. In R. Ader, D. L. Felten, & N.
Cohen (Eds.), Psychoneuroimmunology (3rd ed., pp. 583–612). New York: Ac-
ademic Press.
Darwin, C. (1899). The descent of man (2nd ed.). London: Murray (Original work
published in 1871).
1212 Dickerson, Gruenewald, Kemeny

Dickerson, S. S., & Gable, S. L. (2004, January). Emotional and physiological

effects of avoidance motives and goals following an acute stressor. Poster
presented at the annual meeting of the Society for Personality and Social
Psychology, Austin, TX.
Dickerson, S. S., & Kemeny, M. E. (2004). Acute stressors and cortisol responses:
A theoretical integration and synthesis of laboratory research. Psychological
Bulletin, 130, 355–391.
Dickerson, S. S., Kemeny, M. E., Aziz, N., Kim, K. H., & Fahey, J. L. (2004).
Immunological effects of induced shame and guilt. Psychosomatic Medicine, 66
(1), 124–131.
Dugue, B., Leppanen, E. A., Teppo, A. M., Fyhrquist, F., & Grasbeck, R. (1993).
Effects of psychological stress on plasma interleukins-1 beta and 6, C-reactive
protein, tumour necrosis factor alpha, anti-diuretic hormone, and serum co-
rtisol. Scandinavian Journal of Clinical Laboratory Investigations, 53, 555–561.
Ekman, P. (1999). Basic emotions. In T. Dalgleish & M. Power (Eds.). Handbook
of cognition and emotion. Sussex, UK: John Wiley and Sons, Ltd.
Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic nervous system
activity distinguishes among emotions. Science, 221, 1208–1210.
Elliot, A. J., & Thrash, T. M. (2004). The intergenerational transmission of fear of
failure. Personality Social Psychology Bulletin, 30, 957–971.
Fahey, J. L., Taylor, J. M., Detels, R., Hofmann, B., Melmed, R., & Nishanian,
P., et al. (1990). The prognostic value of cellular and serologic markers in in-
fection with human immunodeficiency virus type 1. The New England Journal of
Medicine, 322 (3), 166–172.
Feldmann, M., Brennan, F. M., & Maini, R. N. (1996). Role of cytokines in
rheumatoid arthritis. Annual Review of Immunology, 14, 397–440.
Ford, E. S. (2003). The metabolic syndrome and C-reactive protein, fibrinogen,
and leukocyte count: findings from the Third National Health and Nutrition
Examination Survey. Atherosclerosis, 168, 351–358.
Gehm, T. L., & Scherer, K. R. (1988). Relating situation evaluation to emotion
differentiation: Nonmetric analysis of cross-cultural questionnaire data. In K.
R. Scherer (Ed.)., Facets of emotion: Recent research (pp. 61–77). Hillsdale, NJ:
Erlbaum.
Gilbert, P. (1997). The evolution of social attractiveness and its role in shame,
humiliation, guilt, and therapy. British Journal of Medical Psychology, 70,
113–147.
Gilbert, P. (2000). Varieties of submissive behavior as forms of social defense:
Their evolution and role in depression. In L. Sloman & P. Gilbert (Eds.), Sub-
ordination and defeat: An evolutionary approach to mood disorders and their
therapy. Mahwah, NJ: Lawrence Erlbaum Associates.
Gilbert, P., & Trower, P. (1990). The evolution and manifestation of social anx-
iety. In W. R. Crozier (Ed.) Shyness and embarrassment: Perspectives from so-
cial psychology (pp. 144–177). New York: Cambridge University Press.
Goebel, M. U., Mills, P. J., Irwin, M. R., & Zeigler, M. G. (2000). Interleukin-6
and tumor necrosis factor-a production after acute psychological stress, exer-
cise, and infused isoproterenol: Differential effects and pathways. Psychoso-
matic Medicine, 62, 591–598.
Shame, Physiology, and Health 1213

Gold, P. W., Licinio, J., Wong, M. L., & Chrousos, G. P. (1995). Corticotropin
releasing hormone in the pathophysiology of melancholic and atypical depres-
sion and in the mechanism of action of antidepressant drugs. Annals Of The
New York Academy Of Sciences, 771, 716–729.
Gruenewald, T. L., Kemeny, M. E., Aziz, N., & Fahey, J. L. (in press). Acute
threat to the social self: Shame, social self-esteem, and cortisol activity.
Psychosomatic Medicine.
Haller, J., Kiem, D. T., & Makara, G. B. (1996). The physiology of social conflict
in rats: what is particularly stressful? Behavioral Neuroscience, 110, 353–359.
Harris, T., Ferrucci, L., Tracy, R., Corti, M., Wacholder, S., & Ettinger, W. J.,
et al. (1999). Association of elevated interleukin-6 and C-reactive protein
levels with mortality in the elderly. American Journal of Medicine, 106,
506–512.
Hart, B. L. (1988). Biological basis of the behavior of sick animals. Neuroscience
and Biobehavioral Reviews, 12, 123–137.
Kedzierska, K., Crowe, S. M., Turville, S., & Cunningham, A. L. (2003). The
influence of cytokines, chemokines and their receptors on HIV-1 replication in
monocytes and macrophages. Review of Medical Virology, 13, 39–56.
Keltner, D. (1995). Signs of appeasement: Evidence for the distinct displays of
embarrassment, amusement, and shame. Journal of Personality and Social
Psychology, 68, 441–454.
Keltner, D., & Buswell, B. N. (1996). Evidence for the distinctness of embarrass-
ment, shame, and guilt: A study of recalled antecedents and facial expressions
of emotion. Cognition & Emotion, 10, 155–171.
Keltner, D., & Gross, J. J. (1999). Functional accounts of emotion. Cognition and
Emotion, 13, 467–480.
Keltner, D., Young, R. C., & Buswell, B. N. (1997). Appeasement in human
emotion, social practice, and personality. Aggressive Behavior, 23, 359–374.
Kemeny, M. E. (2003). The psychobiology of stress. Current Directions in Psy-
chological Science, 12 (4), 124–129.
Kemeny, M. E., & Dean, L. (1995). Effects of AIDS-related bereavement on HIV
progression among New York City gay men. AIDS Education and Prevention,
7, 36–47.
Kemeny, M. E., Gruenewald, T. L., & Dickerson, S. S. (2004). Social-self pres-
ervation system: The interface of social threats, self-evaluation, and health.
Manuscript in preparation.
Kenis, G., & Maes, M. (2002). Effects of antidepressants on the production of
cytokines. International Journal of Neuropsychpharmacology, 5, 401–412.
Kronfol, Z., & Remick, D. G. (2000). Cytokines and the brain: Implications for
clinical psychiatry. American Journal of Psychiatry, 157, 683–694.
Lane, R. D., Reiman, E. M., Ahern, G. L., & Schwartz, G. E. (1997). Neuroan-
atomical correlates of happiness, sadness and disgust. American Journal of
Psychiatry, 154, 926–933.
Leary, M. R., & Baumeister, R. F. (2000). The nature and function of self-esteem:
Sociometer theory. Advances in Experimental Social Psychology, 32, 1–62.
Lewis, H. B. (1971). Shame and guilt in neurosis. New York: International Uni-
versities Press.
1214 Dickerson, Gruenewald, Kemeny

Lewis, M., & Ramsay, D. (2002). Cortisol response to embarrassment and shame.
Child Development, 73, 1034–1045.
Lewis, T. T., Kemeny, M. E., Myers, H. F., & Wyatt, G. E. (2004). Perceived
interpersonal rejection and CD4 decline in a community sample of women in-
fected with HIV. Manuscript in preparation.
Levenson, R. W. (1994). The search for autonomic specificity. In P. Ekman & R.
Davidson (Eds.) The nature of emotion:Fundamental questions (pp. 252–257).
New York: Oxford University Press.
Lovallo, W. R., & Thomas, T. L. (2000). Stress hormones in psychophysiological
research: Emotional, behavioral and cognitive implications. In J. T. Cacioppo,
L. G. Tassinary, & G. G. Berntson (Eds.), Handbook of psychophysiology (pp.
342–367). Cambridge: Cambridge University Press.
MacLean, P. D. (1990). The triune brain in evolution. New York: Plenum.
Maes, M. (1999). Major depression and activation of the inflammatory response
system. In R. Dantzer, E. Wollman, & R. Yirmiya (Eds.), Cytokines, stress and
depression (pp. 25–46). New York: Kluwer Academic.
Maes, M., Van Gastel, A., Block, P., Martin, M., Cosyns, P., & Scharpe, S., et al.
(1995). An augmented escape of androstenedione from suppression
by dexamethasone in melancholia: Relationships to intact ACTH and
cortisol nonsuppression. Journal of Affective Disoders, 34, 291–300.
Maier, S. F., & Watkins, L. R. (1998). Cytokines for psychologists: Implications
of bi-directional immune-to-brain communication for understanding behavior,
mood and cognition. Psychological Review, 105, 83–107.
Mason, J. W. (1968). A review of psychoendocrine research on the pituitary-ad-
renal cortical system. Psychosomatic Medicine, 30, 576–607.
McEwen, B. S. (1998). Protective and damaging effects of stress mediators. New
England Journal of Medicine, 338, 171–179.
Miller, G. E., Stetler, C. A., Carney, R. M., Freedland, K. E., & Banks, W. A.
(2002). Clinical depression and inflammatory risk markers for coronary heart
disease. American Journal of Cardiology, 90, 1279–1283.
Mokros, H. B. (1995). Suicide and shame. American Behavioral Scientist, 38 (8),
1091–1103.
Munck, A., Guyre, P. M., & Holbrook, N. J. (1984). Physiological functions of
glucocorticoids in stress and their relation to pharmacological actions. Endo-
crine Reviews, 5 (1), 25–44.
Pasic, J., Levy, W. C., & Sullivan, M. D. (2003). Cytokines in depression and
heart failure. Psychosomatic Medicine, 65 (2), 181–193.
Peters, M. L., Godaert, G. L. R., Balleiux, R. E., Brosschot, J. F., Sweep, F. C. G.
J., & Swinkels, L. M. J. W., et al. (1999). Immune responses to experimental
stress: Effects of mental effort and uncontrollability. Psychosomatic Medicine,
61, 513–524.
Price, J., Sloman, L., Gardner, R., Gilbert, P., & Rohde, P. (1994). The social
competition hypothesis of depression. British Journal of Psychiatry, 164,
309–315.
Quan, N., Avitsur, R., Stark, J. L., He, L., Shah, M., & Caligiuri, M., et al. (2001).
Social stress increases the susceptibility to endotoxic shock. Journal of Neuro-
immunology, 115, 36–45.
Shame, Physiology, and Health 1215

Rohleder, N., Schommer, N. C., Hellhammer, D. H., Engel, R., & Kirschbaum, C.
(2001). Sex differences in glucocorticoid sensitivity of proinflammatory cytokine
production after psychosocial stress. Psychosomatic Medicine, 63, 966–972.
Russell, J. A. (2003). Core affect and the psychological construction of emotion.
Psychological Review, 110 (1), 145–172.
Sapolsky, R. M. (1993). Endocrinology alfresco: Psychoendocrine studies of wild
baboons. Recent Progress in Hormone Research, 48, 437–468.
Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorti-
coids influence stress responses? Integrating permissive, suppressive, stimula-
tory, and preparative actions. Endocrine Reviews, 21, 55–89.
Scheff, T. J. (2001). Shame and community: social components in depression.
Psychiatry, 64 (3), 212–224.
Schott, S. (1979). Emotion and social life: A symbolic interactionist analysis.
American Journal of Sociology, 84, 1317–1334.
Schwarzer, R. (1986). Self-related cognitions in anxiety and motivation: An in-
troduction. In R. Schwarzer (Ed.) Self-related cognitions in anxiety and moti-
vation (pp. 1–17). Hillsdale, NJ: Erlbaum Associates.
Shively, C. A., Laber-Laird, K., & Anton, R. F. (1997). Behavior and physiology
of social stress and depression in female cynomolgus monkeys. Biological Psy-
chiatry, 41, 871–882.
Segerstrom, S. C., Taylor, S. E., Kemeny, M. E., Reed, G. M., & Visscher, B. R.
(1996). Causal attributions predict rate of immune decline in HIV-seropositive
gay men. Health Psychology, 15, 485–493.
Seyle, H. (1956). The stress of life. New York: McGraw-Hill.
Smith, R. H., Webster, J. M., Parrott, W. G., & Eyre, H. L. (2002). The role of
public exposure in moral and nonmoral shame and guilt. Journal of Personality
and Social Psychology, 83, 138–159.
Stark, J. L., Avitsur, R., Hunzeker, J., Padgett, D. A., Campbell, K. A., & Beck,
F. M., et al. (2001). Social stress induces glucocorticoid resistance in macroph-
ages. American Journal Physiology: Regulatory, Integrative, Comparative,
Physiology, 280 (6), R1799–1805.
Stark, J. L., Avitsur, R., Hunzeker, J., Padgett, D. A., & Sheridan, J. F. (2002).
Interleukin-6 and the development of social disruption-induced glucocorticoid
resistance. Journal of Neuroimmunology, 124, 9–15.
Stefanski, V., & Engler, H. (1998). Effects of acute and chronic social stress on
blood cellular immunity in rats. Physiology and Behavior, 64, 733–741.
Tangney, J. P. (1995). Recent advances in the empirical study of shame and guilt.
American Behavioral Scientist, 38, 1132–1145.
Tangney, J. P., Miller, R. S., Flicker, L., & Barlow, D. H. (1996). Are shame,
guilt, and embarrassment distinct emotions? Journal of Personality and Social
Psychology, 70, 1256–1269.
Tangney, J. P., Wagner, P., & Gramzow, R. (1992). Proneness to shame, prone-
ness to guilt, and psychopathology. Journal of Abnormal Psychology, 101 (3),
469–478.
Tooby, J., & Cosmides, L. (1990). The past explains the present: Emotional ad-
aptations and the structure of ancestral environment. Ethology and Sociobiol-
ogy, 11, 375–424.
1216 Dickerson, Gruenewald, Kemeny

Watkins, L. R., Nguyen, K. T., Lee, J. E., & Maier, S. F. (1999). Dynamic regu-
lation of proinflammatory cytokines. In R. Dantzer, E. Wollman, & R. Yirmiya
(Eds.), Cytokines, stress, and depression. New York: Kluwer Academic.
Weiner, B. (1985). An attributional theory of achievement motivation and emo-
tion. Psychological Review, 92, 548–573.
Weiner, H. (1992). Perturbing the organism: The biology of stressful experience.
Chicago: University of Chicago Press.
Weitzman, O., Kemeny, M. E., & Fahey, J. L. (2004). HIV-related shame and guilt
predict CD4 decline. Manuscript submitted for publication.
Wicker, F. W., Payne, G. C., & Morgan, R. D. (1983). Participant descriptions of
guilt and shame. Motivation and Emotion, 7, 25–39.
Yirmiya, R., Weidenfeld, J., Pollak, U., Morag, M., Morag, A., & Avitsur, R.,
et al. (1999). Cytokines, ‘‘depression due to a general medical condition,’’ and
antidepressant drugs. In R. Dantzer, E. Wollman, & R. Yirmiya (Eds.),
Cytokines, stress and depression (pp. 283–316). New York: Kluwer Academic.
Zakowski, S. G., McAllister, C. G., Deal, M., & Baum, A. (1992). Stress, reac-
tivity, and immune function in healthy men. Health Psychology, 11, 223–232.