Impact of Depression on Response to Comedy:
A Dynamic Facial Coding Analysis
Lawrence Ian Reed, Michael A. Sayette, and Jeffrey F. Cohn
University of Pittsburgh
Individuals suffering from depression show diminished facial responses to positive stimuli. Recent
cognitive research suggests that depressed individuals may appraise emotional stimuli differently than do
nondepressed persons. Prior studies do not indicate whether depressed individuals respond differently
when they encounter positive stimuli that are difficult to avoid. The authors investigated dynamic
responses of individuals varying in both history of major depressive disorder (MDD) and current
depressive symptomatology (N ? 116) to robust positive stimuli. The Facial Action Coding System
(Ekman & Friesen, 1978) was used to measure affect-related responses to a comedy clip. Participants
reporting current depressive symptomatology were more likely to evince affect-related shifts in expres-
sion following the clip than were those without current symptomatology. This effect of current symp-
tomatology emerged even when the contrast focused only on individuals with a history of MDD.
Specifically, persons with current depressive symptomatology were more likely than those without
current symptomatology to control their initial smiles with negative affect-related expressions. These
findings suggest that integration of emotion science and social cognition may yield important advances
for understanding depression.
Keywords: depression, facial expression, facial coding, comedy
Recent years have seen increased interest in applying affective
science research to the study of emotional disorders (Davidson,
2000). With respect to depression, research has examined how
individuals with mood disorders respond to positive and negative
stimuli (Gehricke & Shapiro, 2000). Such interest relates to early
behavioral approaches for treating depression that focused on
modifying one’s environment to create more rewarding experi-
ences (i.e., opportunities to increase positive affect; Lewinsohn &
Graf, 1973). This reasoning rests on the assumption that when
confronted with “positive” cues, depressed individuals will re-
spond appropriately. This assumption may not, however, be en-
Researchers using electromyography and facial coding tech-
niques have examined responses to stimuli intended to elicit pos-
itive emotion among individuals suffering from depression and
nondepressed controls. Typically, these studies expose participants
to stimuli that have reliably elicited positive affect in healthy
participants (e.g., Lang, Bradley, & Cuthbert, 1999). Following
stimulus presentation, immediate responses are recorded. These
studies find that individuals suffering from depression or dyspho-
ria are less likely than healthy controls to react to positive stimuli
with facial expressions associated with positive emotion (Gehricke
& Shapiro, 2000; Schwartz et al., 1976; Sloan, Bradley, Dimoulas,
& Lang, 2002), suggesting a diminished ability among affected
persons to experience positive emotion.
With few exceptions (Sloan et al., 2002), prior research has
assessed emotional reactions to positive stimuli without investi-
gating the development of the expression over time. In addition to
the morphology or configuration of these expressions, dynamic
aspects of the facial movement provide information about the
strength of the emotion and whether it is genuine (Ekman, 1993).
More important, a facial expression of emotion may be composed
of expressions occurring in a rapid sequence, conveying an emo-
tional experience different from what is conveyed by each separate
expression in the sequence (Ekman, 1993). Regarding depression,
an example might be a smile control. According to Keltner (1995),
smile controls involve facial actions that potentially counteract the
upward pull of the smile and/or obscure the smile. Accordingly,
researchers have begun to investigate these dynamic aspects of
emotional expression. Analysis of motion when viewing facial
expressions, for example, improves affect judgments. Specifically,
subtle facial expressions were identifiable when presented in dy-
namic displays, but not in static displays (Ambadar, Schooler, &
Cohn, 2005). This research dovetails with prior work revealing
that displays of embarrassment and amusement can be distin-
guished by examining temporal differences in their configurations
(Keltner & Buswell, 1997).
This basic research, which highlights the importance of dynamic
measures of facial expression, may prove useful for advancing
knowledge regarding mechanisms underlying depression. As
noted above, clinical investigations document that persons with
depression or dysphoria, on average, are less likely than nonde-
pressed individuals to express positive affect when presented with
Lawrence Ian Reed, Michael A. Sayette, and Jeffrey F. Cohn, Depart-
ment of Psychology, University of Pittsburgh.
This research was supported in part by National Institute of Mental
Health Grants P01 MH056193 and R01 MH051435. We thank Ellen Frank,
Charles George, and Karen Schmidt for their helpful comments. Rachel M.
Levenstein and Kasey M. Griffin provided assistance with FACS coding.
Correspondence concerning this article should be addressed to Law-
rence Ian Reed, Department of Psychology, 4315 Sennott Square, Univer-
sity of Pittsburgh, 210 S. Bouquet Street, Pittsburgh, PA 15260. E-mail:
Journal of Abnormal Psychology
2007, Vol. 116, No. 4, 804–809
Copyright 2007 by the American Psychological Association
0021-843X/07/$12.00 DOI: 10.1037/0021-843X.116.4.804
positive-mood inductions. Nevertheless, this mean difference does
not eliminate the possibility that some depressed participants
within the group still might display positive emotional reactions. It
is possible that in addition to a diminished ability to experience
positive affect, there also exists an active suppression of positive
affect in individuals suffering from depression. A primary aim of
the present study was to examine how individuals with a history of
depression (specifically major depressive disorder; MDD) who
either were or were not currently suffering from depressive symp-
tomatology respond to positive stimuli when positive responses are
difficult to avoid.
Although not without debate (cf. Block & Colvin, 1994; Taylor
& Brown, 1994), there is some evidence that individuals with
depression differ from those without depression in the way they
process self-relevant information. Nondepressed individuals ap-
pear to hold self-serving biases that buffer their perceptions and
appraisals in ways that enhance their self-concept (Taylor &
Brown, 1988). Moreover, when emotion information that is incon-
gruent with their self-concept cannot be avoided, there may be
ways that this information is marginalized (Taylor & Brown,
1988). In contrast, people suffering from depression may process
information about the self more negatively than do their nonde-
pressed counterparts (Abramson & Alloy, 1981; Golin, Terrell,
Weitz, & Drost, 1979). These studies suggest that depression may
alter the way that positive information is appraised, even when it
cannot be avoided.
Consistent with this position, recent work using a dot-probe task
suggests that individuals experiencing depression have difficulty
disengaging from negative stimuli. Gotlib, Krasnoperova, Yue,
and Joorman (2004) investigated attentional biases for sad faces
presented for varying lengths of time. Results suggested that
differences in stimulus duration may distinguish between initial
orientation and maintenance of responses. Their findings further
suggest that measuring affective responses dynamically will per-
mit examination of subtle shifts in affect over brief time intervals
that may characterize the reactions of individuals suffering from
In order to capture such momentary shifts in emotion, we used
an observational coding system to identify facial expressions
thought to be related to emotion (see Ekman & Rosenberg, 2005).
The most comprehensive of these coding systems is the Facial
Action Coding System (FACS; Cohn & Ekman, 2005). FACS is an
anatomically based system for measuring facial movement. Using
FACS, coders can code all possible facial displays—referred to as
action units (AUs) (Ekman & Friesen, 1978; Ekman, Friesen, &
Hager, 2002)—in order to provide an objective and reliable
method of measuring facial behavior over extremely rapid time
frames. Thus, use of FACS to examine responses to emotion cues
can provide key information that otherwise could go unnoticed if
relying exclusively on self-report measures. It is hypothesized that
these momentary shifts will be expressed in the face by initial
positive affect-related expressions, immediately followed by ex-
pressions of negative emotion in dysphoric individuals.
Previous studies also have used affect inductions that tended not
to reliably induce positive affect in depressed individuals. For
example, Gehricke and Shapiro (2000) reported mean happiness
ratings of less than 3 on a 0–9 scale. (More important, these low
ratings were consistent with the aim of that study, which was not
designed to examine expressions immediately following an initial
positive response.) In contrast, the present study required a stim-
ulus that would be especially effective in eliciting an immediate
positive response, even among persons experiencing symptoms of
depression. Prior studies examining the effect of depression on
positive affect have not been able to determine whether difficulty
responding to positive stimuli reflects a stable characteristic of
individuals with depression, or instead signals current depressive
symptomatology. By including groups of individuals with (a) a
history of MDD and current depressive symptomatology, (b) a
history of MDD without current depressive symptomatology, and
(c) no history of MDD (or other psychopathology) and no current
depressive symptomatology, we aimed to test the importance of
current depressive symptomatology versus general vulnerability to
MDD in responding to positive stimuli. Such an effort required a
sample that included individuals with a history of MDD who either
were currently symptomatic or currently asymptomatic.
In summary, this study tested the impact of a history of MDD,
as well as current depressive symptomatology, on emotional re-
sponses to a stimulus designed to induce positive emotion. By
using a robust positive stimulus, a sensitive and dynamic measure
of facial behavior, and by recruiting relatively large samples of
individuals with a history of MDD who were and who were not
currently symptomatic, this study aimed to observe a more com-
plex and dynamic pattern of emotion responding than found in
prior studies. We hypothesized that participants with a history of
MDD and currently symptomatic would be more likely to control
their smiles with particular negative expressions after hearing the
punch line of a comedy clip than would currently asymptomatic
Participants who enrolled prior to May 2003 in a longitudinal,
multidisciplinary program project examining risk factors for
childhood-onset mood disorders (see Miller et al., 2002) were
eligible for the present study. The present study focused on a
subset (67.1%; n ? 116: 30 men, 86 women) of this sample who
smiled in response to a comedy clip (see the Procedures section
for additional details). Included in the sample were individuals
with a history of MDD and individuals without a history of
psychopathology. Diagnostic information concerning history of
depression was obtained via the Structured Clinical Interview for
DSM–IV Patient Version (First, Spitzer, Gibbon, & Williams,
1994), adapted to include childhood diagnoses. To be classified
with a history of MDD, individuals had to receive a diagnosis of
this disorder using Diagnostic and Statistical Manual of Mental
Disorders, 3rd edition (DSM–III; American Psychiatric Associa-
tion, 1980), the 3rd edition revised (DSM–III–R; American Psy-
chiatric Association, 1987), or the 4th edition (DSM–IV; American
Psychiatric Association, 1994) criteria before entering the study.
Current depressive symptomatology was ascertained prior to film
clip viewing (described in the Procedures section) using the Beck
Depression Inventory (BDI; Beck, Steer, & Farbin, 1988). Follow-
ing guidelines for the BDI cut-off scores, distributed by The Center
for Cognitive Therapy, current depressive symptomatology was
defined as a BDI score ? 18, which corresponds to a moderate to
severe level of depressive symptomatology (Beck et al., 1988).
DEPRESSION AND RESPONSE TO COMEDY
More details on the recruitment and diagnosis determination pro-
cedure are found in Miller et al. (2002).
To examine the effects of history of MDD and current symp-
tomatology, participants were divided into three groups: (a) those
with a history of MDD who were currently symptomatic (history
? current; n ? 18), (b) those with a history of MDD who were not
currently symptomatic (history no current; n ? 39), and (c) those
without a history of psychopathology who were not currently
symptomatic (no history no current; n ? 59). (Because there was
only one individual in the project with elevated BDI scores but
without a history of MDD, this fourth group was omitted.) The
distribution of gender in each group was similar (74% female, 26%
male), as was ethnicity (80% Caucasian, 16% African American,
4% “other”; ps ? .10). There were no group differences regarding
gender or race (ps ? .10). Significant group differences were
found in age, F(2, 113) ? 4.36, p ? .015 (see Table 1). Age,
therefore, was included as a covariate in data analyses.
Participants were seated comfortably in a chair and asked to
make ratings on a series of film clips varying in emotional content.
Each participant then independently viewed a neutral film clip
(showing a train moving down a track), followed by a comedy clip,
and then followed by five other clips that are not included in this
study. The comedy clip was of the contemporary comedian Chris
Rock, entitled “Chris Rock: Bring the Pain,” selected on the basis
of criteria outlined by Gross and Levenson (1995) to reliably elicit
positive emotion. In this segment, Rock discusses a range of topics
such as food and relationships. We used the first 11 s of this 4-min
clip, which builds up to a clear, initial punch line. This resulted in
a stimulus that is well suited for assessing responses, as partici-
pants generally are in a neutral state at the start of the 11-s
segment. Because responses to this segment of the comedy clip are
the focus of this article, the remaining film clips are not described.
Video recordings of the participants’ facial behavior were digitized
for 11 s at 30 frames per second, which produced a set of 330
sequential 640 ? 480 pixel full color images for each participant.
The 11-s segment digitized for each participant began precisely 1 s
before the punch line and ended 10 s after the punch line. The
length of this digitized sequence allowed for analysis of sponta-
neous smiles that are similar in length to the average 4–6 s
reported for spontaneous smiles (Frank, Ekman, & Friesen, 1993).
The FACS (Ekman et al., 2002) was used to
measure facial behavior. To become a certified FACS coder, one
must complete a standardized exam and attain an agreement ratio
of at least .70 with criteria. This reliability has been shown to
generalize to research settings in which spontaneous emotion is
assessed (Sayette, Cohn, Wertz, Perrott, & Parrott, 2001).
FACS coding for the specified 11-s segment for smiles (as
defined by movement of the zygomatic major muscle: AU 12) was
completed by Lawrence Ian Reed (blind to history and symptom-
atology of participants). A number of different lower facial AUs
appearing following smile onsets have been identified as reflecting
smile controls (see Keltner, 1995). As noted by others (Ekman,
Friesen, & O’Sullivan, 1988), these AUs putatively modify the
meaning of the smile expression, such that the overall experience
may shift from reflecting joy to something different (Keltner &
Buswell, 1997). Smile controls are often low-frequency expres-
sions, and for the purpose of this study, we examined candidate
AUs that were present at least 10 times. These included move-
ments corresponding to contempt (“dimplers” AU 14) and sadness
(“lip corner depressors” AU 15; see Figure 1). Thus, the presence
of a smile control (the appearance of either AU14 or AU15 during
the smile) was dichotomously coded for each participant. Follow-
ing Keltner (1995), these expressions were coded immediately
following smile onset. Smile controls were only coded if they
occurred before smile offset (following Ekman et al., 1988).
Twenty percent of the participants were independently coded by a
comparison coder certified in FACS and quantified using kappa,
which corrects for chance agreement. Reliability for coding of
smile controls was acceptable (? ? .73).
Following both the neutral and happy
film clips, self-reported happiness ratings were recorded for each
participant on a 9-point Likert-type scale. (Although less relevant
to the present study’s aims, four other emotion terms—sadness,
anger, disgust, and fear ratings—were included in the project.)
Age, Self-Report, and Behavioral Findings by Group
No history no
(n ? 59)
No current (n ? 39) Current (n ? 18)
M SDM SDM SD
or omission by the participant, happiness ratings were missing for 3 participants for the neutral clip and 5 for
the comedy clip. The Group ? Clip interaction regarding happiness ratings was not significant, p ? .50, F(2,
105) ? 0.71. Means with nonoverlapping superscripts differ significantly at p ? .05 by Bonferroni correction.
Current symptoms defined as a Beck Depression Inventory (BDI) score ? 18. Due to technical reasons
REED, SAYETTE, AND COHN
Positive Affect Induction
We first examined whether the comedy clip was effective in
eliciting positive reactions (i.e., the occurrence of smiles) across all
three groups of participants. The majority (67.1%) of participants
across the three groups expressed smiles during the clip. This
percentage was similar among the three groups (58.1% history
current, 66.1% history no current, and 71.1% no history no cur-
rent), ?2(2, N ? 173) ? 1.77, p ? .41.
Self-reported happiness ratings reinforced the FACS findings.
[Due to technical reasons or omission by the participant, happiness
ratings were missing for 3 participants for the neutral clip and 5 for
the comedy clip.] More specifically, a 3 (group) ? 2 (rating
following neutral vs. following comedy clip) mixed ANOVA
revealed a main effect for clip, such that participants in all three
groups reported significantly more happiness following the com-
edy clip, F(1, 105) ? 136.18, p ? .001. A main effect of group
also was observed, with participants in the no-history-no-current
group reporting being happiest, F(2, 105) ? 9.61, p ? .001. There
was no Group ? Clip interaction (see Table 1). The effect of
watching the comedy clip on happiness was comparable in each
group. Endorsement of negative emotion during the neutral and
comedy clips was extremely low. On a 0–8 Likert-type scale, all
but 2 of the 24 mean scores were less than 1.0; the highest was
In summary, FACS coding and self-report measures both sug-
gested that the comedy clip was successful in providing a reliable
manipulation of positive affect across the three groups.
Shifts in Affect-Related Expressions
Using age as a covariate, a logistic regression model contrasting
the likelihood of observing a smile control in the three groups
revealed a significant effect for group, ?2(2, N ? 116) ? 11.71,
p ? .003. The effect of age was not significant, ?2(1, N ? 116) ?
0.13, p ? .72.
By comparing the history-no-current group with the no-history-
no-current group, we were able to contrast the effects of history of
MDD among those not currently symptomatic. Results indicated
that history-no-current individuals were no more likely to evince a
smile control than no-history-no-current participants (? ? ?0.42,
p ? .51).
By comparing the history ? current group with the history-no-
current group, we were able to contrast the effects of current
symptomatology among individuals with a history of MDD. his-
tory ? current in comparison with history no current were more
likely to use smile controls (? ? 1.71, p ? .009). The correspond-
ing odds ratio was 5.50, which suggests that history ? current
participants were 5.5 times more likely to show a smile control.
Point biserial correlations revealed that smile controls were
related to the BDI (r ? .32, p ? .001) as well as to self-report
ratings of sadness (r ? .23, p ? .018) and disgust (r ? .21, p ?
.025) following the comedy clip. Smile controls were unrelated to
self-report ratings of positive affect (r ? ?.14, p ? .16) following
the comedy clip.
The major finding of this study was that among individuals with
a history of MDD, those with current depressive symptomatology
were more likely than were asymptomatic individuals to express
smile controls during smiles. More specifically, individuals with
both a history of MDD and current depressive symptomatology
were more likely to express smile controls during smiles than were
individuals with a history of MDD without current depressive
To our knowledge, this is the first study using FACS to measure
dynamic affect-related shifts in a sample with a history of MDD
and current depressive symptomatology. These data reinforce prior
studies that found that individuals suffering from depression re-
sponded to positive stimuli with diminished responses. Our study
found that even when an initial smile is elicited, individuals with
current symptomatology are more likely than asymptomatic per-
sons to express negative AUs that are thought to control their
original positive response (Ekman et al., 1988). Although prior
work has not been able to disentangle the impact of a history of
MDD from an individual’s current state (as patients typically were
recruited on the basis of current diagnosis), the present study
provides preliminary evidences that it is the current state, rather
than some stable tendency related to a history of MDD, that is
driving the response to positive stimuli.
The momentary shifts in expression captured by FACS were not
observed in our postcomedy clip affect ratings. Nor did we find
group differences in initial positive facial responses. The similar
initial responses to the joke may have been due to the effectiveness
of our comedy clip in eliciting positive responses. Positive affec-
tive ratings in the present study were higher than those found in
previous research using film clips (e.g., Rottenberg, Kasch, Gross,
& Gotlib, 2002). Although we believe that this difference is due to
the effectiveness of our comedy clip, it remains possible that it is
a result of the way in which depression is defined in the present
study. Increases in positive affect ratings following the comedy
unilateral AU 14).
Example of sequence from neutral (action unit [AU] 0) to smile (AU 12) to smile control (AU 12 ?
DEPRESSION AND RESPONSE TO COMEDY
clip did not differ by group, suggesting similar response patterns
The FACS data reported here are consistent with the view that
depressive symptomatology can bias one toward negative inter-
pretations of cues. It also provides a complement to work suggest-
ing that currently depressed individuals have difficulty disengag-
ing from negative stimuli (Gotlib et al., 2004) and work suggesting
that depression severity might be associated with aversive re-
sponses to positive stimuli (Allen, Trinder, & Brennan, 1999). Our
data also are consistent with views of depression that consider
negative processing biases to be the result of well-learned routines,
or heuristics (Robinson, Goetz, Wilkowski, & Hoffman, 2006).
Alternatively, participants expressing smile controls simply may
have been responding with ambivalence. Future research to probe
these different sources of bias is indicated.
Together with other research (e.g., Allen et al., 1999; Gotlib et
al., 2004), our findings suggest that, under certain conditions,
individuals with current depressive symptomatology may actively
process information in order to create affectively congruent re-
sponses. This position differs from traditional conceptualizations
that link depression to a passive style of responding (Beck, 1967).
Accordingly, merely encouraging such patients to surround them-
selves with positive stimuli or situations may not be sufficient in
treatment if attention is not also paid to how these stimuli might be
appraised. Behavioral treatments for depression that involve be-
havioral activation recognize that individuals may not all find the
same stimuli reinforcing (Gable, Reis, & Elliot, 2000). With re-
spect to laboratory research, use of idiographic approaches to
assessing positive affect may prove valuable.
Given the important communicative function of facial expres-
sion (Fridlund, 1991), it is possible that the display of smile
controls by depressed individuals may alienate them from others.
This possibility may provide a clue toward understanding how the
behavior of depressed people may elicit hostility and rejection
from others (see Coyne, 1990).
A number of limitations to this study must be taken into account
in interpreting its findings. Because most of the participants in-
cluded in the sample were in their 20s, it is possible that some of
those classified without a history might develop MDD later in
adulthood. Regarding our sample, a no history ? current group
was not included in the analyses due to its small sample size. Thus,
we were unable to contrast the effects of current symptomatology
among individuals with a history of MDD by comparing a no
history ? current group with the history ? current group. More-
over, the majority of participants in the history-no-current (77%)
and history ? current (78%) groups were diagnosed with MDD
prior to age 16. Because early onset depression is associated with
poor adult outcome compared with later onset depression (e.g.,
Lewinsohn, Clarke, Seeley, & Rohde, 1994), future research is
needed to determine how well our findings generalize to individ-
uals with later onset depression.
In addition, use of a self-reported measure of depressive symp-
tomatology can be problematic. First, rather than strictly measur-
ing depressive symptomatology, the BDI may represent a more
global measure of negative affect. Second, any BDI cut-off score
is arbitrary. Thus, results could change with the use of different
cut-off scores. It also is possible that the BDI cut-off scores
function differently in those with and without a history of MDD.
More generally, it remains unclear whether the observed pattern of
smile controls would be maintained in response to other types of
emotion inductions (e.g., unpleasant and neutral stimuli).
The present data suggest that affective reactions in individuals
with a history of MDD with and without current symptomatology
are more likely to unfold differently over time than in asymptom-
atic “healthy” persons. Future research using other techniques
(e.g., brain imaging) may provide further support for the conclu-
sion that depression is associated with a complex controlling
response to positive information.
This research represents an initial attempt to use a dynamic
measure of affect-related expression to examine the unfolding of
emotion in individuals with depressive symptomatology. More
generally, this study highlights the utility of integrating basic
research in emotion with the study of affective disorders. Such
research holds promise for improving understanding of mecha-
nisms underlying depression, while also providing a fertile sample
for investigating emotional volatility. Microanalysis of such rap-
idly changing affect-related expressions may yield new insights
into theories of human emotion.
Abramson, L. Y., & Alloy, L. B. (1981). Depression, non-depression, and
cognitive illusions: A reply to Schwartz. Journal of Experimental Psy-
chology, 110, 436–447.
Allen, N. B., Trinder, J., & Brennan, C. (1999). Affective startle modula-
tion in clinical depression: Preliminary findings. Biological Psychiatry,
Ambadar, Z., Schooler, J. W., & Cohn, J. F. (2005). Deciphering the
enigmatic face: The importance of facial dynamics in interpreting subtle
facial expressions. Psychological Science, 16, 403–410.
American Psychiatric Association. (1980). Diagnostic and statistical man-
ual of mental disorders (3rd ed.). Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and statistical man-
ual of mental disorders (3rd ed., rev.). Washington, DC: Author.
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical
aspects. New York: Hoeber.
Beck, A. T., Steer, R. A., & Farbin, M. G. (1988). Psychometric properties
of the Beck Depression Inventory: Twenty-five years of evaluation.
Clinical Psychology Review, 8, 77–100.
Block, J., & Colvin, C. R. (1994). Positive illusions and well-being
revisited: Separating fiction from fact. Psychological Bulletin, 116, 28.
Cohn, J. F., & Ekman, P. (2005). Measuring facial action by manual
coding, facial EMG, and automatic facial image analysis. In J. A.
Harrigan, R. Rosenthal, & K. Scherer (Eds.), Handbook of nonverbal
behavior research methods in the affective sciences (pp. 9–64). New
York: Oxford University Press.
Coyne, J. C. (1990). Interpersonal processes in depression. In G. L. Keitner
(Ed.), Depression and families (pp. 31–54). Washington, DC: American
Davidson, R. J. (2000). Anxiety, depression, and emotion. New York:
Oxford University Press.
Ekman, P. (1993). Facial expression and emotion. American Psychologist,
Ekman, P., & Friesen, W. V. (1978). Facial Action Coding System. Palo
Alto, CA: Consulting Psychology Press.
Ekman, P., Friesen, W. V., & Hager, J. C. (2002). The Facial Action
Coding System. Salt Lake City, UT: Research Nexus, Network Research
Ekman, P., Friesen, W. V., & O’Sullivan, M. (1988). Smiles when lying.
Journal of Personality and Social Psychology, 54, 414–420.
REED, SAYETTE, AND COHN
Ekman, P., & Rosenberg, E. (2005). What the face reveals: Basic and Download full-text
applied studies of spontaneous expression using the facial action coding
system (FACS). New York: Oxford University Press.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1994).
Structured clinical interview for axis I DSM–IV disorders. SCID-I/
P(2.00). New York: Biometrics Research Department.
Frank, M. G., Ekman, P., & Friesen, W. V. (1993). Behavioral markers and
recognizability of the smile of enjoyment. Journal of Personality and
Social Psychology, 64, 83–93.
Fridlund, A. J. (1991). Sociality of solitary smiling: Potentiation by an
implicit audience. Journal of Personality and Social Psychology, 60,
Gable, S. L., Reis, H. T., & Elliot, A. (2000). Behavioral activation and
inhibition in everyday life. Journal of Personality and Social Psychol-
ogy, 78, 1135–1149.
Gehricke, J. G., & Shapiro, D. (2000). Reduced facial expression and social
context in major depression: Discrepancies between facial muscle ac-
tivity and self-reported emotion. Psychiatry Research, 95, 157–167.
Golin, S., Terrell, T., Weitz, J., & Drost, P. L. (1979). The illusion of
control among depressed patients. Journal of Abnormal Psychology, 88,
Gotlib, I. H., Krasnoperova, E., Yue, D. N., & Joorman, J. (2004). Atten-
tional biases for negative interpersonal stimuli in clinical depression.
Journal of Abnormal Psychology, 113, 127–135.
Gross, J., & Levenson, R. W. (1995). Emotion elicitation using films.
Cognition & Emotion, 9, 81–108.
Keltner, D. (1995). Signs of appeasement: Evidence for the distinct dis-
plays of embarrassment, amusement, and shame. Journal of Personality
and Social Psychology, 68, 441–454.
Keltner, D., & Buswell, B. N. (1997). Embarrassment: Its distinct form and
appeasement functions. Psychological Bulletin, 122, 250–270.
Lang, P. J., Bradley, M. M., & Cuthbert, B. M. (1999). International
affective picture system (IAPS): Instructional manual and affective rat-
ings. Gainesville: University of Florida, Center for Research in Psycho-
Lewinsohn, P. M., Clarke, G. N., Seeley, J. R., & Rohde, P. (1994). Major
depression in community adolescents: Age at onset, episode duration,
and time to recurrence. Journal of the American Academy of Child &
Adolescent Psychiatry, 33, 809–818.
Lewinsohn, P. M., & Graf, M. (1973). Pleasant activities and depression.
Journal of Consulting and Clinical Psychology, 41, 261–268.
Miller, A., Fox, N. A., Cohn, J. F., Forbes, E. E., Sherrill, J. T., & Kovacs,
M. (2002). Regional patterns of brain activity in adults with a history of
childhood-onset depression: Gender differences and clinical variability.
American Journal of Psychiatry, 159, 934–940.
Robinson, M. D., Goetz, M. C., Wilkowski, B. M., & Hoffman, S. J.
(2006). Driven to tears or to joy: Response dominance and trait-based
predictions. Personality and Social Psychology Bulletin, 32, 629–640.
Rottenberg, J., Kasch, K. L., Gross, J. J., & Gotlib, I. H. (2002). Sadness
and amusement reactivity differentially predict concurrent and prospec-
tive functioning in major depressive disorder. Emotion, 2, 135–146.
Sayette, M. A., Cohn, J. F., Wertz, J. M., Perrott, M. A., & Parrott, D. J.
(2001). A psychometric evaluation of the Facial Action Coding System
for assessing spontaneous expression. Journal of Nonverbal Behavior,
Schwartz, G. E., Fair, P. L., Mandel, M. R., Salt, P., Mieske, M., &
Klermen, G. L. (1976, April 30). Facial muscle patterning to affective
imagery in depressed and nondepressed subjects. Science, 192, 489–
Sloan, D. M., Bradley, M. M., Dimoulas, E., & Lang, P. J. (2002). Looking
at facial expressions: Dysphoria and facial EMG. Biological Psychology,
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social
psychological perspective on mental health. Psychological Bulletin, 102,
Taylor, S. E., & Brown, J. D. (1994). Positive illusions and well-being
revisited: Separating fact from fiction. Psychological Bulletin, 116,
Received May 16, 2006
Revision received April 13, 2007
Accepted April 16, 2007 ?
DEPRESSION AND RESPONSE TO COMEDY