The Importance of Being Flexible:The Ability to Both Enhance and Suppress Emotional Expression Predicts Long-Term Adjustment
George A. Bonanno, Anthony Papa, Kathleen O扤eill, Maren Westphal, and Karin Coifman
Teachers College, Columbia University
Address correspondence to George A. Bonanno, Clinical Psychology Program, 525 West 120th
St., Box 218, Teachers College, Columbia University, New York, NY 10027; e-mail:
[email protected].
ABSTRACT
Researchers have documented the consequences of both expressing and suppressing emotionusing between-subjects designs. It may be argued, however, that 英國dissertation網successful adaptation dependsnot so much on any one regulatory process, but on the ability to flexibly enhance or suppressemotional expression in accord with situational demands. We tested this hypothesis among NewYork City college students in the aftermath of the September 11th terrorist attacks. Subjects?
performance in a laboratory task in which they enhanced emotional expression, suppressedemotional expression, and behaved normally on different trials was examined as a prospectivepredictor of their adjustment across the first two years of college. Results supported the
flexibility hypothesis. A regression analysis controlling for initial distress and motivation andcognitive resources found that subjects who were better able to enhance and suppress theexpression of emotion evidenced less distress by the end of the second year. Memory deficitswere also observed for both the enhancement and the suppression tasks, suggesting that bothprocesses require cognitive resources.
Is it better to express or conceal one's emotions? Although scholars have debated thisquestion for centuries, research over the past few decades suggests that both expressing andsuppressing the expression of emotion can serve adaptive ends, but also that both behaviors may
extract some cost (Bonanno, 2001; Gross, 1998b). How can these competing findings bereconciled? Recent research on coping has indicated that the crucial element in successfuladaptation is not so much which particular strategies are used, but rather whether copingstrategies are applied flexibly in a manner that corresponds with the nature of the stressor (e.g.,Cheng, 2001). In a similar vein, emotion theorists have increasingly argued that whether oneexpresses or suppresses emotional expression is not as important for adjustment as is the abilityto flexibly express or suppress emotional expression as demanded by the situational context
(Barrett & Gross, 2001; Bonanno, 2001; Consedine, Magai, & Bonanno, 2002; Parrott, 1993;
Westphal & Bonanno, 2004).
It is widely accepted that emotions are not unidimensional phenomena, but rather
manifest themselves through multiple response channels, including emotional experience,
expression, and physiology. Each of these components is thought to serve distinct adaptive ends#p#分頁標題#e#
and to be subject to self-regulatory processes (Bonanno, 2001; Gross, 1998b). The expression of
http://www.400190.com/dissertation_writing/sociology/emotion serves multiple adaptive functions, including communicating and regulating internal
states (Ekman & Davidson, 1993; Izard, 1990; Zajonc, Murphy, & Inglehart, 1989) and
developing and maintaining social interactions (Darwin, 1872; Ekman, 1993; Keltner, 1995).
There are times, however, when expressing emotion may be harmful (Gross & Mu駉z, 1995;
Kennedy-Moore & Watson, 2001). For example, the chronic expression of negative emotion (in
particular, anger) is a risk factor in cardiovascular disease (Adler & Matthews, 1994). At times, it
may be more adaptive to suppress rather than express emotion (Consedine et al., 2002; Parrott,
1993). The ability to hide one抯 feelings is useful in myriad social situations (e.g., not showing
fear while presenting a speech, minimizing the expression of anger during conflict mediation)
and the idea that this ability is important resonates with evolutionary considerations of the
ubiquity and survival value of deception (de Waal, 1989; Trivers, 1985). In extremely adverse
conditions, the ability to modulate display of negative emotions can foster the recovery of
normal functioning (Bonanno & Keltner, 1997), help maintain and expand social networks
(Coyne, 1976; Harber & Pennebaker, 1992), and facilitate close personal relationships (Levenson
& Gottman, 1983).
Despite its potential usefulness, expressive suppression, like emotional expression, may
extract serious costs if employed indiscriminately or chronically. To measure emotional
suppression, Gross and Levenson (1993, 1997) developed a between-subjects paradigm in which
some subjects were instructed to conceal all outward signs of emotion and then exposed to
emotional stimuli. Using this paradigm, Gross and Levenson found that although suppressing
subjects reported the same level of negative experience as control subjects, they exhibited
heightened sympathetic activity (Gross, 1998a; Gross & Levenson, 1993, 1997), had poorer
memory for the emotional stimuli (Richards & Gross, 2000), and engendered reduced rapport
and willingness to affiliate in dyadic partners (Butler et al., 2003).
Although the confluence of this evidence clearly suggests that successful adaptation
depends on the ability to both enhance and suppress emotional expression and to do so flexibly
in accord with situational demands, this idea has never been tested directly. The current
investigation was designed to redress this deficit by examining subject抯 performance in a
laboratory study of expressive regulation as a prospective predictor of their adjustment following
a potentially stressful life transition. More specifically, we modified the between-subjects#p#分頁標題#e#
suppression paradigm developed by Gross and Levenson (1993) in two important ways. First, to
examine the ability of individuals to flexibly regulate emotional expression both upward and
downward, we developed a within-subjects task that included conditions for the enhancement
and the suppression of emotional expression, as well as a control condition in which participants
were instructed to behave as they would normally. Our inspiration for this design was a recent
within-subjects study of enhancement and suppression of emotional experience (Jackson,
Malmstadt, Larson, & Davidson, 2000). By manipulating expressive regulation within subjects,
we were also able to examine whether the memory deficits associated with suppression might
also be observed following expressive enhancement. Richards and Gross (2000) suggested this
deficit was due to the cognitive load associated with suppressing. We assumed that expressive
enhancement would also tax cognitive resources and therefore produce a similar memory deficit.
Second, we examined the ability to regulate emotional expression upward and downward
as a prospective predictor of long-term adjustment in a sample of New York City college
students. Many students experience at least some increase in distress during the transition to
college life (Gerdes & Mallinckrodt, 1994). The students in the current study began college in
New York City just prior to the September 11th terrorist attacks, and this transition was likely to
have been particularly difficult for students with poor emotion-regulation skills. On the basis of
the assumption that the ability to flexibly enhance and suppress the expression of emotion
promotes adaptation, we predicted that greater ability to perform these tasks soon after beginning
college would be correlated with better adjustment by the end of the second year in college.
METHOD
Subjects and Procedure
Within 1 month after beginning college, 101 New York City undergraduates (67 female,
34 male; mean age = 18.05 years, SD = 0.59) completed a 32-item version of the Symptom
Checklist-90-Revised (Derogatis, 1983) as a measure of initial (T1) distress. One to 3 months
later, these subjects participated individually in an expressive-regulation experiment. Long-term
adjustment was assessed by repeating the distress measure at the end of the second academic
year (T2). Fourteen subjects (14%) dropped out of the study before T2, leaving 87 subjects for
longitudinal analyses. Subjects who dropped out did not differ from remaining subjects on any
variable measured in this study (p > .15). Subjects were paid $250 for each year they
participated.
For the expressive-regulation task, subjects were seated before a desktop computer and
filmed from a one-way mirror positioned above their line of vision. They were instructed in how
to interact with software that displayed blocked sequences of five digitized picture stimuli#p#分頁標題#e#
selected from the International Affective Picture System (Lang, Bradley, & Cuthbert, 1995).
College-student norms (Lang et al., 1995) were used to balance stimuli for valence and arousal
across blocks. Within each block, each stimulus was presented for 10 s, with 4 s between stimuli.
For practice, subjects viewed randomly presented blocks of positive or negative stimuli, and
following each block rated the degree to which they felt 搉egative emotion (e.g., anger,
revulsion, sadness, distress),?by typing a number between 1 (no negative emotion) and 7
(extreme negative emotion), and then the degree to which they felt 損ositive emotion (e.g.,
happiness, joy, amusement, interest),?using a similar scale.
Following practice trials, subjects were told that there was another subject in the adjacent
room who would also take part in the experiment (another subject was not actually present in the
adjacent room); that they would not see the other person, but the other person would sometimes
be able to view them on a video monitor; that they would always be informed when the monitor
was on and when it was off; and that the other person could not hear them or see the picture
stimuli but would attempt to guess their emotions for each block of stimuli. The instructions
further explained that when the experiment began, the computer would (a) sometimes ask
subjects to enhance their expression of emotion so the observer could more easily guess what
they were feeling, (b) sometimes ask them to suppress their expression of emotion so the
observer could not easily guess what they were feeling, and (c) sometimes inform them that the
monitor was turned off and that the observer would be unable to see them, in which case they
should behave as they would normally. Subjects were then shown three paragraphs, one
describing each condition.
The paragraph for the expression condition was as follows:
The monitor is on. Shortly, you will be presented with a set of images. Please view each
image carefully. While viewing the images, please do your best to express as fully as possible
the emotions you feel while viewing the images. Remember that the person viewing you on the
monitor can only see your head and part of your upper torso, and cannot hear you. It is important
for the sake of this study that you do your best to communicate what you are feeling. So please
do the best you can to behave in such a way that the person viewing you on a monitor will be
able to guess what you are feeling while viewing the images. When you have viewed each
image, you will be asked to rate the emotional reactions you had to the images.
The suppression condition was described as follows:
The monitor is on. Shortly, you will be presented with a set of images. Please view each
image carefully. While viewing the images, please do your best to suppress as fully as possible#p#分頁標題#e#
any expression of the emotions you feel while viewing the images. Remember that the person
viewing you on the monitor can only see your head and part of your upper torso, and cannot hear
you. It is important for the sake of this study that you do your best to conceal what you are
feeling. So please do the best you can to behave in such a way that the person viewing you on a
monitor will not be able to guess what you are feeling while viewing the images. When you
have viewed each image, you will be asked to rate the emotional reactions you had to the images.
The third paragraph described the monitor-off condition:
The monitor is off. Shortly, you will be presented with a set of images. Please view each
image carefully. The person in the other room will not be able to see you while you view this
set of images. Simply view the images in any way you would naturally do so. When you have
viewed each image, you will be asked to rate the emotional reactions you had to the images.
Subjects were informed that one of the instruction paragraphs would precede each block
of stimuli, and that emotion ratings would follow each block of stimuli. Six blocks of
experimental trials (enhancement, suppression, or control instruction using positive or negative
stimuli) were then presented in random order.
At the completion of all six trials, subjects were administered a 10-min filler task
consisting of sixth-grade-level math and word problems, with the instruction to complete as
many problems as possible. After the filler task, subjects were given an unexpected memory test
consisting of questions about the first four picture stimuli in each block of trials. Four questions
were asked about each picture, for a total of 96 questions: Half pertained to emotional details and
half pertained to nonemotional details.
Observer Ratings of Emotional Expression
Three master's-level psychology students who were blind to the goals and hypothesis of
the study rated videotapes of subjects?performance for emotional expression. Observers used the
same positive and negative scales as the subjects used. Onset and offset of each block of trials
were indicted by an auditory signal, and observers had no knowledge of the subject抯 instructions
for any given block. Overall observer agreement was high (intraclass correlation coefficient =
.91) and did not differ significantly by expression condition or stimulus valence. Final scores for
observer-rated expression were calculated by averaging across three raters.
RESULTS
Preliminary analyses of subjective and observer-rated emotion indicated that only the
ratings matching the valence of the stimuli (e.g., negative ratings following negative stimuli)
produced meaningful effects, and that the opposite-valence ratings (e.g., positive ratings
following negative stimuli) were relatively low and did not vary across conditions. Accordingly,#p#分頁標題#e#
only the matching-valence ratings were used in subsequent analyses. There was a significant
main effect of gender on emotion, F(1, 99) = 9.35, p < .01. Overall average emotion (subjective
and observer rated) was greater for female (M = 2.16, SD = 0.37) than male (M = 1.96, SD =
0.54) participants. Gender was not involved in any significant interaction effects and did not
moderate distress or any of the expressive-regulation findings reported (p > .10).
Expressive-Regulation Task
Analyses of the emotion ratings supported the validity of the expression manipulation. A
repeated measures analysis of variance for rating source (subject, observer), expression condition
(enhancement, suppression, no monitor), and stimulus valence (positive, negative) revealed a
significant main effect for stimulus valence, F(1, 98) = 29.75, p < .001; ratings overall were
higher for negative emotion (M = 4.09, SD = 0.84) than for positive emotion (M = 3.72, SD =
0.75). The main effect of source was also significant, F(1, 98) = 450.30, p < .001; subjective
ratings were higher (M = 4.96, SD = 1.03) than observer ratings (M = 2.84, SD = 0.7). Most
important, the qualifying Condition X Source interaction, F(2, 97) = 156.08, p < .001, supported
the efficacy of the manipulation: Subjective ratings did not differ significantly across conditions,
F(2, 99)= 0.33, whereas observer ratings did, F(2, 98) = 224.20, p < .001 (see Fig. 1). In further
accord with the manipulation, Student-Newman-Keuls tests (p < .05) showed that subjects
expressed significantly greater emotion in the enhancement condition (M = 4.27, SD = 1.20) than
the control, no-monitor condition (M = 2.73, SD = 1.12) and significantly greater emotion in the
control condition than the suppression condition (M = 1.54, SD = 0.53).
Memory for Emotion Stimuli
We anticipated that the additional cognitive load required to regulate emotional
expression would reduce memory for stimuli presented in both the enhancement and suppression
conditions. Recognition accuracy was above chance in all conditions. As predicted, however,
there was a significant main effect of condition, F(2, 99) = 5.51, p < .01, and recognition was
significantly weaker in the enhancement (M = 51.8%, SD = 13.0) and suppression (M = 54.0%,
SD = 13.0) conditions than in the control condition (M = 57.5%, SD = 13.7).
Adjustment
Despite the temporal proximity to the September 11th terrorist attacks, distress for the
sample at T1 (M = 0.75, SD = 0.53) was within the range typically observed among college
students (e.g., Todd, Deane, & McKenna, 1997). There was a trend toward lower distress at T2
(M = 0.67, SD = 0.49), t(86) = 1.37, p = .17. However, distress scores at T1 and T2 were only
moderately correlated (r = .36, p < .001), indicating considerable variability across time.#p#分頁標題#e#
Correlations between distress and raw emotion variables in each condition are presented
in Table 1. None of the subjective emotion variables were associated with distress at either time
point. T1 distress was inversely correlated with the expression of positive emotion in both the
enhancement and the suppression conditions (i.e., highly distressed subjects were less able to
enhance and better able to suppress the expression of positive emotion). T2 distress was
inversely correlated with the expression of both positive and negative emotion in the
enhancement condition and with the expression of negative emotion in the control condition.
Thus, individual correlations, similar to those that would be generated from a between-subjects
design, indicated only that subjects who expressed more emotion tended to have less distress
over time.
Expressive Flexibility as a Predictor of Change in Long-Term Adjustment
Our primary hypothesis was that the ability to both enhance and suppress the expression
of emotion would predict reduced distress over time. To test this hypothesis, we created separate
variables for expressive-enhancement ability and expressive-suppression ability by calculating
for each subject the difference between the total levels of emotion (positive and negative)
expressed in the enhancement and control conditions and the difference between the total levels
of emotion expressed in the suppression and control conditions. The enhancement and
suppression ability scores were then summed to create an overall flexibility score (i.e., higher
overall scores indicated greater ability to both enhance and suppress).
To examine the predicted relationships of enhancement and suppression ability to
improved long-term adjustment, we conducted a series of hierarchical regressions using T2
distress as the dependent variable. The first step in each analysis included distress at T1 as a
control for initial levels of distress. The T1 distress variable also controlled for the possibility
that deficits in cognitive resources and motivation among highly distressed individuals
influenced their performance in the expressive-regulation task. As an additional control for this
factor, the initial step in each analysis also included the number of problems completed during
the 10-min filler task (M = 35.39, SD = 8.04). This variable was mildly inversely correlated with
T1 distress (r = -.19, p < .05) and mildly positively correlated with expressive ability (r = .20, p
= .05). The regression analyses are summarized in Table 2.
The initial regression step combining T1 distress and number of filler problems was
significant, accounting for 11% of the variance in T2 distress. Adding the two variables
representing expressive-enhancement and expressive-suppression ability on a subsequent step
explained an additional 7% of the variance in T2 distress and significantly increased the overall#p#分頁標題#e#
R2 of the equation to .18, F(2, 76) = 3.13, p < .05. In the next analysis, we replaced the
enhancement and suppression variables with the summed flexibility score, which also explained
an additional 7% of the variance in T2 distress, F(1, 77) = 6.33, p < .05. Together, these analyses
indicate that the abilities to enhance and suppress the expression of emotion each independently
contributed to long-term adjustment regardless of level of adjustment prior to the experiment or
cognitive resources and motivation. Thus, people who possess both of these abilities (i.e.,
expressive flexibility) will tend to have the best long-term adjustment, and people low in both
abilities will tend to have the poorest long-term adjustment.
A possible alternative explanation for these findings is that generally people tend to show
greater ability in one form of expressive regulation than the other and that the combined score,
rather than representing flexibility, may represent extreme polarity (i.e., extremely high scores on
one ability). Indeed, although we expected expressive enhancement and expressive suppression
to be mildly positively correlated, these variables were actually moderately inversely correlated
(r = -.34, p < .001). To examine this issue further, we created an additional summary variable
representing polarity in expressive regulation. The polarity variable was calculated as the
absolute value of the discrepancy between expressive-enhancement and expressive-suppression
ability; thus, higher polarity scores represent more extreme asymmetry in expressive regulation.
Entering the polarity variable instead of the flexibility variable in a third regression analysis did
not explain any additional variance in T2 distress and did not increase R2 beyond its value in the
first step of the analysis, F(1, 77) = 0.33, n.s. Thus, extreme ability in one form of expressive
regulation over the other does not predict change in long-term adjustment.
DISCUSSION
This study provides the first direct empirical support for the assumption that successful
adaptation is linked to the ability to flexibly enhance or suppress emotional expression. To test
this assumption, we used a within-subjects manipulation to examine expressive enhancement and
suppression as a prospective predictor of distress among New York City undergraduates
beginning college in the immediate aftermath of the September 11th terrorist attacks (T1) and
again prospectively one and a half years later (T2). Although the sample on average did not show
elevated distress, there was considerable variability in distress scores from T1 to T2. Regardless
of their level of adjustment, subjects experienced the emotion-evoking stimuli in much the same
way. However, as predicted by the expressive-flexibility hypothesis, the abilities to enhance and
to suppress the expression of emotion each independently predicted reduced T2 distress, over#p#分頁標題#e#
and above the effects accounted for by T1 distress and the number of problems completed during
the filler task (included as a measure of individual differences in cognitive resources and
motivation). Moreover, combining expressive-enhancement and -suppression abilities into an
overall flexibility score also predicted reduced distress, whereas a polarity score, reflecting
asymmetry between expressive and suppressive abilities, was unrelated to adjustment.
At a more general level, the findings of the current study complement and extend the
corpus of studies that have examined emotion regulation using between-subjects designs. As in
the previous studies, expressive suppression did not influence the subjective experience of
emotion but attenuated the overt expression of emotion and reduced memory for emotional
stimuli. When we instructed the same subjects to enhance the expression of emotion, although
the experience of emotion was again not influenced, subjects did show greater overt displays of
emotion relative to the control condition and also exhibited a memory deficit comparable to that
observed for the suppression task, presumably because both tasks increase cognitive load.
There were several limitations to the study. First, we used only one kind of emotional
stimuli, a set of standardized emotion-evoking picture stimuli, and one type of social action,
communicating or concealing one抯 emotions from an observer in another room. Although this
approach enabled us to vary stimuli and experimental condition on every block of trials, it will
be important for future studies of expressive flexibility to explore other types of stimuli (e.g.,
evocative films) and other types of social interaction (e.g., dyadic interactions), and to examine
whether suppression and enhancement produce different effects for specific emotions. The
current study was also limited by its use of college students. The methodological and conceptual
advantages of manipulating emotion regulation within subjects, demonstrated here and in other
studies (e.g., Jackson et al., 2000), suggest the value of exploring this paradigm in future studies
using more varied samples.
Within the context of these limitations, the experimental measurement of expressive
flexibility presents several new avenues for investigating expressive regulation in relation to
psychological and physical health. Much of the contemporary research on emotion regulation
and adjustment has focused on the frequency with which people express or suppress emotion
and, consequently, has relied heavily on self-reports of emotional behavior (e.g., Gross & John,
1997). Although this work has led to considerable advances in understanding of these behaviors,
there are limits to when and how well people can accurately report on their emotional life over
time (Robinson & Clore, 2002). This concern is to a large extent obviated by the type of withinsubjects#p#分頁標題#e#
experimental paradigm employed in the current study, which focused on regulatory
ability rather than self-reported frequency. Future studies might examine how the experimental
measurement of expressive flexibility compares to self-report measures of both emotional
experience and expression, or to other related processes such as linguistic flexibility (Campbell
& Pennebaker, 2003). It should also be fruitful to compare how expressive flexibility relates to
more broadly defined positive emotion concepts, such as emotional intelligence (Barrett &
Gross, 2001; Salovey, 2001).
The flexibility paradigm also offers a relatively objective, standardized method for
examining the relation between emotion regulation and adjustment across different populations
and situations. For example, expressive flexibility might be examined as a predictor variable
among individuals exposed to highly stressful events or in comparative studies of emotional
difficulties across different forms of psychopathology (Kring & Bachorowski, 1999). In a similar
vein, the measurement of expressive flexibility offers a novel perspective on the health costs
associated with expressive suppression. Although emotional suppression can serve adaptive
ends, considerable evidence suggests that chronic suppression holds adverse consequences for
physical health (Gross & Levenson, 1997; Robinson & Pennebaker, 1991; Sapolsky, 1998). The
英國dissertation網findings of the current study suggest that the health costs of emotional suppression may emerge
only among individuals who fail to show expressive flexibility (Salovey, 2001). The withinsubjects
design used in the current study should make it possible to explore this possibility in
future research.
REFERENCES
Adler, N., & Matthews, K. (1994). Health psychology: Why do some people get sick and
some stay well? Annual Review of Psychology, 45, 229-259.
Barrett, L.F., & Gross, J.J. (2001). Emotional intelligence: A process model of emotion
representation and regulation. In T.J. Mayne & G.A. Bonanno (Eds.), Emotions: Current issues
and future directions (pp. 286-311). New York: Guilford Press.
Bonanno, G.A. (2001). Emotion self-regulation. In T.J. Mayne & G.A. Bonanno (Eds.),
Emotions: Current issues and future directions (pp. 251-285). New York: Guilford Press.
Bonanno, G.A., & Keltner, D. (1997). Facial expressions of emotion and the course of
conjugal bereavement. Journal of Abnormal Psychology, 106, 126-137.
Butler, E.A., Egloff, B., Wilhelm, F.H., Smith, N.C., Erickson, E.A., & Gross, J.J.
(2003). The social consequences of expressive suppression. Emotion, 3, 48-67.
Campbell, R.S., & Pennebaker, J.W. (2003). The secret life of pronouns: Flexibility in
writing style and physical health. Psychological Science, 14, 60-65.#p#分頁標題#e#
Cheng, C. (2001). Accessing coping flexibility in real-life and laboratory settings: A
multimethod approach. Journal of Personality and Social Psychology, 80, 814-833.
Consedine, N., Magai, C., & Bonanno, G.A. (2002). Moderators of the emotion
inhibition-health relationship: A review and research agenda. Review of General Psychology, 6,
204-238.
Coyne, J.C. (1976). Depression and the responses of others. Journal of Abnormal
Psychology, 85, 31-52.
Darwin, C. (1872). The expression of emotion in man and animals. London: Murray.
Derogatis, L.R. (1983). SCL-90-R administration, scoring & procedures manual - II for
the R(evised) version. Towson, MD: Clinical Psychometric Research.
de Waal, F. (1989). Peacemaking among primates. Cambridge, MA: Harvard.
Ekman, P. (1993). Facial expressions and emotion. American Psychologist, 48, 384-392.
Ekman, P., & Davidson, R.J. (1993). Voluntary smiling changes regional brain activity.
Psychological Science, 4, 342-345.
Gerdes, H., & Mallinckrodt, B. (1994). Emotional, social, and academic adjustment of
college students: A longitudinal study of retention. Journal of Counseling and Development, 72,
281-288.
Gross, J.J. (1998a). Antecedent and response-focused emotion regulation: Divergent
consequences for experience, expression, and physiology. Journal of Personality and Social
Psychology, 74, 224-237.
Gross, J.J. (1998b). The emerging field of emotion regulation: An integrative review.
Review of General Psychology, 2, 271-299.
Gross, J.J., & John, O.P. (1997). Revealing feelings: Facets of emotional expressivity in
self-reports, peer ratings, and behavior. Journal of Personality and Social Psychology, 72, 435-
448.
Gross, J.J., & Levenson, R.W. (1993). Emotional suppression: Physiology, self-report,
and expressive behavior. Journal of Personality and Social Psychology, 64, 970-986.
Gross, J.J., & Levenson, R.W. (1997). Hiding feelings: The acute effects of inhibiting
negative and positive emotion. Journal of Abnormal Psychology, 106, 95-103.
Gross, J.J., & Mu駉z, R.F. (1995). Emotion regulation and mental health. Clinical
Psychology: Science and Practice, 2, 151-164.
Harber, K.D., & Pennebaker, J.W. (1992). Overcoming traumatic memories. In S.-A.
Christianson (Ed.), The handbook of emotion and memory: Research and theory (pp. 359-387).
Hillsdale, NJ: Erlbaum.
Izard, C.E. (1990). Facial expressions and the regulation of emotions. Journal of
Personality and Social Psychology, 58, 487-498.
Jackson, D.C., Malmstadt, J.R., Larson, C.L., & Davidson, R.J. (2000). Suppression and
enhancement of emotional responses to unpleasant pictures. Psychophysiology, 37, 515-522.
Keltner, D. (1995). The signs of appeasement: Evidence for the distinct displays of
embarrassment, amusement, and shame. Journal of Personality and Social Psychology, 68, 441-#p#分頁標題#e#
454.
Kennedy-Moore, E., & Watson, J.C. (2001). How and when does emotional expression
help? Review of General Psychology, 5, 187-212.
Kring, A.M., & Bachorowski, J. (1999). Emotions and psychopathology. Cognition and
Emotion, 13, 575-599.
Lang, P.J., Bradley, M.M., & Cuthbert, B.N. (1995). International Affective Picture
System (IAPS): Technical manual and affective ratings (Technical Report A-4). Gainesville:
University of Florida, Center for Research in Psychophysiology.
Levenson, R.W., & Gottman, J.M. (1983). Marital interaction: Physiological linkage and
affective exchange. Journal of Personality and Social Psychology, 45, 587-597.
Parrott, W.G. (1993). Beyond hedonism: Motives for imbibing good moods and for
maintaining bad moods. In D.M. Wegner & J.W. Pennebaker (Eds.), Handbook of mental control
(pp. 278-308). Englewood Cliffs, NJ: Prentice-Hall.
Richards, J.M., & Gross, J.J. (2000). Emotion regulation and memory: The cognitive
costs of keeping one抯 cool. Journal of Personality and Social Psychology, 79, 410-424.
Robinson, M.D., & Clore, G.L. (2002). Belief and feeling: Evidence for an accessibility
model of emotional self-report. Psychological Bulletin, 128, 934-960.
Robinson, R.J., & Pennebaker, J.W. (1991). Emotion and health: Toward an integrative
approach. International Review of Studies on Emotion, 1, 347-367.
Salovey, P. (2001). Applied emotional intelligence: Regulating emotions to become
healthy, wealthy, and wise. In J. Ciarrochi, J.P. Forgas, & J.D. Meyer (Eds.), Emotional
intelligence in everyday life: A scientific inquiry (pp. 168-184). New York: Psychology Press.
Sapolsky, R.M. (1998). Why zebras don抰 get ulcers. New York: W.H. Freeman.
Todd, D.M., Deane, F.P., & McKenna, P.A. (1997). Appropriateness of SCL-90-R
adolescent and adult norms for outpatient and nonoutpatient college students. Journal of
Counseling Psychology, 44, 294-301.
Trivers, R.L. (1985). Social evolution. Boston: Addison-Wesley.
Westphal, M., & Bonanno, G.A. (2004). Emotional self-regulation. In M. Beauregard
(Ed.), Consciousness, emotional self-regulation, and the brain (pp. 1-34). Philadelphia: John
Benjamins.
Zajonc, R.B., Murphy, S.T., & Inglehart, M. (1989). Feeling and facial efference:
Implications of the vascular theory of emotion. Psychological Review, 96, 395-416.
(Received 3/13/03; Revision accepted 11/10/03)
Fig. 1. Subjects?ratings of their own (subjective) emotion and observers?ratings of emotion
expressed by subjects in the three expressive-regulation conditions. Higher ratings indicate
greater emotion.
Table 1. Distress at T1 and T2 in relation to raw values for subjective emotion and expressed
emotion
Subjective emotion
Enhanced expression ?negative
T1 Distress
-.14
T2 distress#p#分頁標題#e#
-.12
Enhanced expression ?positive -.06 -.15
Suppressed expression ?negative .08 -.02
Suppressed expression ?positive -.01 .05
Control ?negative -.07 -.10
Control ?positive -.04 -.14
Expression of emotion
Enhanced expression ?negative
T1 Distress
-.13
T2 distress
-.28*
Enhanced expression ?positive -.22* -.40***
Suppressed expression ?negative .08 .01
Suppressed expression ?positive -.24* -.02
Control ?negative -.04 -.20+
Control ?positive .09 -.07
+ = p < .10; * = p < .05; *** = p < .001.
Table 2. Hierarchical regression analyses predicting T2 distress.
R2 Beta Partial F change
1. T1 distress .11 .26* .34 F(2,78)=5.08**
Number of filler problems completed -.03 -.12
2. Expressive enhancement ability .18 -.24* -.23 F(2,76)=3.13*
Expression suppression ability -.23* -.23
R2 Beta Partial F change
1. T1 distress .11 .26* .34 F(2,78)=5.08**
Number of filler problems completed -.03 -.12
2. Expressive flexibility
(enhancement + suppression)
.18 -.27* -.28 F(1,77)=6.33*
R2 Beta Partial F change
1. T1 distress .11 .26* .34 F(2,78)=5.08**
Number of filler problems completed -.03 -.12
2. Expressive polarity
英國dissertation網( |enhancement - suppression| )
.11 -.02 -.02 F(1,77)=0.33
* = p <.05; ** = p<.01