Working memory capacity and the self-regulation of emotional expression and experience.

Abstract

This research examined the relationship between individual differences in working memory capacity and the self-regulation of emotional expression and emotional experience. Four studies revealed that people higher in working memory capacity suppressed expressions of negative emotion (Study 1) and positive emotion (Study 2) better than did people lower in working memory capacity. Furthermore, compared to people lower in working memory capacity, people higher in capacity more capably appraised emotional stimuli in an unemotional manner and thereby experienced (Studies 3 and 4) and expressed (Study 4) less emotion in response to those stimuli. These findings indicate that cognitive ability contributes to the control of emotional responding.

Full text

Page 1

PERSONALITY PROCESSES AND INDIVIDUAL DIFFERENCES
Working Memory Capacity and the Self-Regulation of Emotional
Expression and Experience
Brandon J. Schmeichel
Texas A&M University
Rachael N. Volokhov and Heath A. Demaree
Case Western Reserve University
This research examined the relationship between individual differences in working memory capacity and
the self-regulation of emotional expression and emotional experience. Four studies revealed that people
higher in working memory capacity suppressed expressions of negative emotion (Study 1) and positive
emotion (Study 2) better than did people lower in working memory capacity. Furthermore, compared to
people lower in working memory capacity, people higher in capacity more capably appraised emotional
stimuli in an unemotional manner and thereby experienced (Studies 3 and 4) and expressed (Study 4) less
emotion in response to those stimuli. These findings indicate that cognitive ability contributes to the
control of emotional responding.
Keywords: cognitive control, emotion regulation, emotional expression, working memory
Emotion regulation is an essential element of psychological
well-being. This is exemplified by the role of faulty emotion
regulation in several forms of psychopathology (American Psychi-
atric Association, 1994). For example, mood and anxiety disorders
have been linked to poor emotion regulation. In one relevant study,
students with generalized anxiety disorder reported poorer ability
to control emotions relative to students without the disorder (Men-
nin, Heimberg, Turk, & Fresco, 2005). In another study, adults
who reported difficulty with emotion regulation also reported more
anxiety, more worry, and more agoraphobic thoughts relative to
other adults (Kashdan, Zvolensky, & McLeish, 2008).
What contributes to success at emotion regulation? The present
investigation examined the contributions of cognitive ability to
emotion regulation. More precisely, four studies tested the hypoth-
esis that individual differences in working memory capacity pre-
dict success at the voluntary regulation of emotional expression
and emotional experience.
Emotion Regulation
Emotion regulation entails efforts to influence the experience,
expression, or duration of an emotional response. According to an
influential process model of emotion regulation, emotion unfurls
over time, and efforts to regulate emotion vary in form and
effectiveness according to when the regulatory efforts occur
(Gross, 1998). Response-focused emotion regulation occurs after
an emotion has arisen and entails efforts to suppress or to amplify
emotional responses. Research suggests that response-focused reg-
ulation influences the external expression of emotion but has little
or no effect on the internal experience of emotion (see Gross,
2002). One experiment, for example, found that participants could
suppress facial expressions of emotion while they viewed scenes
of burn victims and surgical procedures, but expressive suppres-
sion had no effect on subjective responses to the gruesome
scenes—suppressers and expressers were equally disgusted (Gross
& Levenson, 1993). Likewise, participants in another experiment
capably exaggerated facial expressions of emotion while viewing
an unpleasant film stimulus, but expressive exaggeration had no
apparent effect on inner emotional experience (Schmeichel, De-
maree, Robinson, & Pu, 2006).
By contrast, antecedent-focused emotion regulation occurs be-
fore an emotion has arisen or very early in the emotion-generative
process. One form of antecedent-focused emotion regulation en-
tails efforts to reinterpret (or preinterpret) the meaning of an
emotional stimulus. For example, a person may attempt to appraise
an emotional event in neutral or unemotional terms. When used
successfully, neutral appraisals diminish the impact of an emo-
tional stimulus and preempt a full-blown emotional response
(Gross, 2002). Hence, unlike suppression, neutral appraisals may
minimize both the experience and the expression of emotion. In
one study, for example, appraising photographs of badly wounded
Brandon J. Schmeichel, Department of Psychology, Texas A&M Uni-
versity; Rachael N. Volokhov and Heath A. Demaree, Department of
Psychology, Case Western Reserve University.
Portions of this research were supported by a National Research Service
Award to Brandon J. Schmeichel (MH 069139). Thank you to Jackie
Anderson, Lauren Antista, Eric Barber, Kristin Berglund, Anne Bradbury,
Lakeyia Butler, Chris Castanon, Kristi Gold, Allison Greeson, Sophia Hu,
Scott Jacobs, Cathy Laterza, Cody Philips, Lauren Preyss, Lauren Saporito,
Lauren Seeds, and Sommer Shelley for helping to conduct this research.
Thanks also to Seth Gitter, Eden Schmeichel, and the Social/Emotion
Research Group at Texas A&M University for providing feedback on
earlier versions of this article.
Correspondence concerning this article should be addressed to Brandon
J. Schmeichel, Department of Psychology, Texas A&M University, Col-
lege Station, TX 77843-4235. E-mail: schmeichel@tamu.edu
Journal of Personality and Social Psychology, 2008, Vol. 95, No. 6, 1526–1540
Copyright 2008 by the American Psychological Association 0022-3514/08/$12.00DOI: 10.1037/a0013345
1526

Page 2

men in neutral or nonemotional terms helped participants to ex-
perience and to express less negative emotion in response to the
photographs, whereas expressive suppression helped to reduce the
expression but not the experience of negative emotion (Richards &
Gross, 2000, Study 2).
A number of personality characteristics contribute to success at
emotion regulation. Self-esteem is one such characteristic. People
higher in self-esteem are more adept at savoring or prolonging
positive emotional responses relative to people lower in self-
esteem (Wood, Heimpel, & Michela, 2003). Conscientiousness is
also associated with success at emotion regulation. In one illumi-
nating study, participants were insulted by a stranger (Jensen-
Campbell, Knack, Waldrip, & Campbell, 2007). Higher levels of
conscientiousness predicted less anger and less aggression in re-
sponse to the insult, consistent with the idea that participants high
in conscientiousness successfully prevented a strong emotional
response. Resting or baseline levels of activation in the prefrontal
cortex also predict success at emotion regulation. More specifi-
cally, greater relative left-sided activation of the prefrontal cortex
at baseline has been linked to more effective down-regulation of
emotional responses to aversive events (Jackson et al., 2003) and
to greater psychological well-being (Urry et al., 2004).
We propose that individual differences in cognitive ability also
predict success at emotion regulation. Working memory capacity
is a cognitive ability factor that plays a key role in the regulation
of cognitive processes. We tested the hypothesis that working
memory capacity also facilitates the self-regulation of emotional
expressions and experiences.
Working Memory Capacity
Working memory is perhaps best understood in relation to
short-term memory. Short-term memory refers to the simple main-
tenance of information in memory over the course of a few
seconds or minutes, whereas working memory refers to the capac-
ity to maintain information while simultaneously processing other
information or engaging other cognitive operations. Following
Engle (2001) and others (e.g., de Fockert, Rees, Frith, & Lavie,
2001; Smith & Jonides, 1999), we defined working memory ca-
pacity as the ability to sustain goal-relevant information processing
in the presence of alternative goals or other distractions.
Several tasks have been devised to measure individual differ-
ences in working memory capacity. The most prominent of these
is the operation span (OSPAN) task, which requires the test taker
to perform a short-term memory test while solving mathematical
equations (Turner & Engle, 1989). The better the person encodes
and recalls target words (goal-relevant information) despite also
having to solve mathematical equations (goal-irrelevant informa-
tion), the higher the person’s working memory capacity.
Most research on individual differences in working memory
capacity has examined its relationship to other cognitive or atten-
tional tasks. This research confirms working memory’s central role
in cognitive control (Engle, Kane, & Tuholski, 1999). For exam-
ple, one set of experiments assessed performance on the Stroop
color-word interference task, which measures the ability to over-
ride an automatic response in favor of a more controlled response.
Participants higher in working memory capacity performed better
(i.e., suffered less interference from the automatic response) than
people lower in capacity did (Kane & Engle, 2003). Working
memory has also been linked to the control of visual attention. One
study found that working memory capacity contributed to the
ability to focus one’s eyes away from a salient visual stimulus
(Kane, Bleckley, Conway, & Engle, 2001; see also de Fockert et
al., 2001). Another study found that working memory capacity
predicted the ability to ignore the “unattended” message in a
dichotic-listening task (Conway, Cowan, & Bunting, 2001). All of
these findings converge on the view that working memory capacity
serves to sustain goal-relevant processing despite the presence of
competing response tendencies or distractions.
Relatively less research has examined the relationship between
working memory capacity and emotional processes or responses
(see Unsworth, Heitz, & Engle, 2005). The research that has been
done in this area has tended to examine the influence of emotional
states on working memory capacity. For example, some forms of
stress and anxiety may reduce working memory capacity (Ashcraft
& Kirk, 2001; Beilock & Carr, 2005; Darke, 1988; Klein & Boals,
2001; Schmader & Johns, 2003). Other research has examined the
extent to which emotional stimuli (rather than emotional states)
influence capacity. One study found evidence of lower working
memory capacity among individuals with an avoidant attachment
style but only when the measure of capacity contained emotionally
charged attachment-related words. When the measure contained
neutral words or emotional words unrelated to attachment, work-
ing memory capacity was not associated with attachment style
(Edelstein, 2006). In some instances, then, emotional states and
stimuli reduce working memory capacity (cf. Kensinger & Corkin,
2003).
The Present Research
Whereas previous research has examined the effects of emo-
tional states and emotional stimuli on the operation of working
memory, the present research examined working memory’s con-
tribution to the voluntary regulation of emotional expression and
emotional experience. We hypothesized that working memory
capacity would facilitate both response-focused and antecedent-
focused forms of emotion regulation, such that higher capacity
predicts more thorough suppression of emotional expression and
more successful use of neutral appraisals to preempt expressive
and experiential responses to emotional stimuli.
Our first objective was to assess the relationship between work-
ing memory capacity and expressive suppression. Emotionally
charged stimuli elicit expressive behavior quickly and automati-
cally (e.g., Dimberg, Thunberg, & Grunedal, 2002). To suppress
emotional expressions, then, one must override expressive im-
pulses and replace them with a more controlled response, namely
a stone-faced or stoic expression (e.g., Gross & Levenson, 1993;
Jackson, Malmstadt, Larson, & Davidson, 2000; Vohs &
Schmeichel, 2003). Insofar as working memory capacity sustains
goal-relevant responses in the presence of competing responses or
distractions, higher capacity should be associated with superior
ability to maintain a neutral facial expression while attending to
stimuli that trigger expressive impulses.
The ability to hide emotions is potentially useful across a wide
range of social situations, and so superior abilities to suppress
emotional expressions are likely to confer several advantages. For
instance, the ability to hide anger while being chastised by one’s
boss may help one to retain gainful employment, and the ability to
1527
WORKING MEMORY AND EMOTION REGULATION

Page 3

hide excitement while holding a pair of aces may help one to win
a sizable pot at the poker table. However, expressive suppression
typically does not alter subjective feelings in response to emotion-
ally evocative stimuli (e.g., Gross, 1998), including the film stim-
uli used in the present studies (Demaree, Robinson, Pu, & Allen,
2006; Demaree, Schmeichel, et al., 2006). Thus, we anticipated
little or no relationship between working memory and subjective
emotional experience under instructions to suppress emotional
expressions.
Our second goal was to assess the relationship between working
memory capacity and effectiveness of adopting neutral appraisals
of emotional stimuli. To appraise an otherwise emotional stimulus
in detached or unemotional terms, a person must generate and
sustain “cool” thoughts to counteract the impact of a “hot” emo-
tional stimulus. We reasoned that working memory capacity could
be devoted to sustaining goal-relevant responses (i.e., unemotional
thoughts) in the presence of competing responses (i.e., emotional
responses). Given that neutral appraisals of emotional stimuli help
to minimize both emotional expression and subjective emotional
experience (e.g., Gross, 1998), we predicted that people with
higher working memory capacity would both experience and ex-
press less emotion under instructions to adopt neutral appraisals of
emotional stimuli compared to people with lower working memory
capacity.
Previous research on the neural basis of emotion regulation has
provided support for the hypothesis that working memory contrib-
utes to the effective use of neutral appraisals. Research using
functional magnetic resonance imaging has observed that, com-
pared to simply attending to emotional stimuli, attempting to
appraise emotional stimuli in nonemotional terms increases acti-
vation of both the lateral prefrontal cortex and the medial prefron-
tal cortex (Ochsner, Bunge, Gross, & Gabrieli, 2002; Ochsner et
al., 2004). These same areas of the brain are crucial for effective
cognitive control (Miller & Cohen, 2001), so we expected that
people who are particularly adept at cognitive control (i.e., those
higher in working memory capacity) would also be adept at
regulating their emotions using neutral appraisals.
Study 1: Working Memory and the Suppression of
Disgust Expressions
Study 1 tested the hypothesis that working memory capacity con-
tributes to the self-regulation of emotional expression. Participants
completed a test of working memory capacity and then watched an
emotional film clip under instructions to suppress all expressions of
emotion as they watched. We predicted that participants higher in
working memory capacity would express less emotion relative to
participants lower in working memory capacity.
We also measured and statistically controlled for individual differ-
ences in the tendency to express negative emotion. Trait negative
expressivity has been linked to facial displays of negative emotion in
response to film clips similar to the clip used in Study 1 (see Gross &
John, 1997), so we sought to ascertain whether working memory
capacity predicted the ability to suppress negative emotional expres-
sions above and beyond individual differences in the tendency to
express negative emotion.
Method
Participants.
ticipated in a laboratory study concerning memory and emotion.
The average age of participants was 19.09 years (SD ? 1.04;
range ? 18–23). In this study and the subsequent studies, partic-
ipants earned partial credit toward a course requirement for their
participation.
Procedure.
Participants first completed a questionnaire and a
working memory test. The Berkeley Expressivity Questionnaire
(BEQ; Gross & John, 1995, 1997) measured dispositional tenden-
cies to express positive and negative emotions. Most relevant for
Study 1 was the BEQ subscale tapping Negative Expressivity,
which included items such as “Whenever I feel negative emotions,
people can easily see exactly what I am feeling,” and “No matter
how nervous or upset I am, I tend to keep a calm exterior”
(reverse-scored; ? ? .59). Participants indicated their agreement
with each item on the BEQ using a scale from 1 (strongly dis-
agree) to 7 (strongly agree). The average score on the six-item
Negative Expressivity subscale in the present sample was 20.89
(SD ? 5.17).
After completing the BEQ, participants completed a widely used
and well-validated measure of working memory capacity known as
the OSPAN task (Turner & Engle, 1989). The OSPAN task mea-
sures participants’ abilities to maintain information in memory
while performing concurrent cognitive operations. One aspect of
the OSPAN task requires participants to solve mathematical equa-
tions. For example, participants saw “(9 ? 3) – 1 ? 2” and had to
indicate whether the given answer was correct. (In the example,
the correct answer was no.) The second aspect of the OSPAN task
requires participants to read and recall target words. One target
word (e.g., house) is presented after each mathematical equation.
Thus, participants read an equation, evaluated whether it was
correct, read a target word, and then advanced to the next equation,
the next target word, and so on. Participants saw sets of two, three,
four, or five equation/word pairings before being prompted to
recall all the target words in the set. Participants worked through
16 sets totaling 56 equation/word pairings in all, presented in the
same order for each participant. Consistent with past work, we
operationalized working memory capacity as the number of words
comprising only word sets recalled in full (e.g., Schmader &
Johns, 2003; see Conway et al., 2005). Working memory capacity
scores in the present sample ranged from 0 to 34, and the average
capacity score was 15.87 (SD ? 6.23).1
During the second half of the study, participants watched a film
clip that has been shown to increase negative emotions (particu-
larly disgust; Demaree, Schmeichel, Robinson, & Everhart, 2004;
Schmeichel et al., 2006). The 2-min clip depicted gruesome scenes
of animal mutilation and slaughter from the film Faces of Death
(James, Scott, & Good, 1978). Prior to watching the film clip,
participants were instructed to suppress their emotional expres-
sions as they watched. More specifically, participants were told to
“keep your face perfectly expressionless as you watch; that is, try
to inhibit any outward expressions of emotion during the film
Forty-five undergraduate students (22 men) par-
1Two participants did not recall any word sets in full and therefore
earned a score of 0 on the working memory measure. Excluding these
participants did not alter the statistical significance of any of the results.
1528
SCHMEICHEL, VOLOKHOV, AND DEMAREE

Page 4

clip.” In this study and all subsequent studies, participants were
aware that their faces were being recorded by a small video camera
as they viewed film clips (for “record keeping purposes”), and
participants consented to having the recordings of their faces
coded for emotionally expressive behavior.
Immediately after the clip, participants reported their mood state
using the UWIST Mood Adjective Checklist (Matthews, Jones, &
Chamberlain, 1990). We focused on participants’ responses to the
negative mood items (sad, depressed, sorry, dissatisfied; ? ? .78).
Expressive suppression typically does not alter subjective
responses to emotional stimuli in general (e.g., Gross & Levenson,
1993; Richards & Gross, 2000) or the Faces of Death clip in
particular (e.g., Demaree, Schmeichel, et al., 2006), so we did not
expect working memory capacity to influence negative mood in
this study. Last, participants were debriefed regarding the purpose
of the experiment and thanked for their participation.
Three naı ¨ve judges, working independently, subsequently
viewed the videotapes of participants’ faces and rated how emo-
tionally expressive each face was on a continuous scale from 1 (not
at all) to 100 (extremely). The judges did not use any particular
coding protocol. Rather, they simply rated facial expressions of
emotion in an ecologically valid manner. The ratings of the judges
were highly interrelated (rs ? .70), so they were averaged together
to form a single measure of emotional expression (? ? .90).
Results
Expression of emotion.
tive to participants lower in working memory capacity, participants
higher in capacity would more successfully suppress the expres-
sion of emotion in response to a negative emotional stimulus. This
hypothesis was supported by a regression that predicted judges’
ratings of emotional expression from working memory capacity
(centered), gender of participant (coded 0 ? female and 1 ? male),
and their interaction. Working memory capacity was the only
significant predictor of emotional expression (B ? ?0.40, p ?
.05). Neither the main effect of gender (B ? 1.79, p ? .77) nor the
Gender ? Working Memory interaction (B ? ?0.58, p ? .08) was
statistically significant. Thus, when instructed to suppress emo-
tional expressions in response to a gruesome film clip, participants
higher in working memory capacity expressed less emotion rela-
tive to participants lower in working memory capacity. Figure 1
displays the scatter plot of the correlation between working mem-
ory capacity and emotional expression, r(45) ? ?.35, p ? .05.
We also assessed whether the relationship between working
memory and expressive suppression remained significant even
when controlling for individual differences in the tendency to
express negative emotion. Trait negative expressivity correlated
with the judges’ ratings of emotional expression, r(45) ? .26, p ?
.08, such that participants who reported a dispositional tendency to
express negative emotions expressed more emotion during the film
clip. Trait negative expressivity and working memory capacity
correlated at r(45) ? ?.22, p ? .14. As expected, the partial
correlation between working memory capacity and emotional ex-
pression, controlling for trait negative expressivity, was signifi-
cant, rp(42) ? ?.31, p ? .05. Hence, working memory predicted
expressive suppression above and beyond the influence of dispo-
sitional tendencies to express negative emotion.
The central hypothesis was that, rela-
Experience of emotion.
sured immediately after the film clip so that we could assess the
relationship between working memory capacity and subjective
responses to the film clip. A regression analysis in which negative
mood was the dependent variable and working memory capacity
(centered), gender of participant (coded 0 ? female and 1 ? male),
and their interaction were the predictors revealed no significant
effects: for working memory capacity, b ? 0.01, p ? .80, for
gender, b ? ?1.14, p ? .20, and for Gender ? Working Memory,
b ? ?0.02, p ? .65. Hence, negative mood states following the
film clip were similar across the range of working memory capac-
ity scores. Note, however, that our mood measure did not assess
disgust, which is a common response to the film clip used in this
study. We therefore cannot rule out the possibility that participants
higher in working memory capacity experienced less disgust rel-
ative to participants lower in capacity. Our findings indicate only
that working memory capacity did not relate to general negative
mood under instructions to suppress.
Participants’ mood states were mea-
Discussion
In Study 1, participants attempted to maintain stone-faced or
stoic expressions while watching a film clip that induces a high
degree of negative emotion. Compared to people lower in working
memory capacity, people higher in capacity expressed less emo-
tion in response to the film clip. This finding shows for the first
time that individual differences in working memory are associated
with the successful suppression of emotional expression.
Although the findings from Study 1 supported our hypothesis
that working memory makes essential contributions to the regula-
tion of emotional responses, two limitations of the study should be
noted. First, Study 1 examined only the suppression of negative
emotional expressions. If our hypothesis is correct, then people
higher in working memory capacity should be more effective at
suppressing positive emotional expressions as well.
Second, all participants in Study 1 were instructed to sup-
press their emotional expressions. Although we found that
people higher in working memory capacity expressed less emo-
0
20
40
60
010
Working Memory Capacity
2030
Emotional Expression
Figure 1.
expression under instructions to suppress emotional expressions (Study 1).
Relationship between working memory capacity and emotional
1529
WORKING MEMORY AND EMOTION REGULATION

Page 5

tion, it is possible that people higher in capacity are simply less
expressive or less emotionally reactive relative to people lower
in capacity. In this view, working memory capacity may be
linked to lower emotional expressivity rather than to higher
ability to regulate emotional expressivity. We conducted a
second study to address these issues.
Study 2: Working Memory and the Suppression of
Amusement Expressions
Study 2 had two primary goals. First, we sought to assess
whether working memory capacity predicts the voluntary
suppression of emotional expression versus emotional expres-
sivity generally. To this end, participants in Study 2 were
randomly assigned either to suppress or to express their emo-
tional expressions while watching an emotionally evocative
video. If working memory capacity contributes specifically to
the suppression of emotional expression, then we should find a
relationship between working memory and emotional expres-
sion when participants are instructed to suppress emotion but
not when they are instructed to express emotion.
Second, whereas Study 1 examined the suppression of negative
emotional expressions, Study 2 undertook to show that working
memory capacity predicts the ability to suppress positive emo-
tional expressions. Hence, participants in Study 2 watched an
amusing video clip that has been shown to elicit smiles, laughter,
and positive affect (Demaree et al., 2004).
Method
Participants.
pated in a laboratory study concerning memory and emotion. The
average age of the participants was 18.71 years (SD ? 0.99;
range ? 18–22).
Procedure.
Participants first completed a questionnaire and a
working memory test. The BEQ (Gross & John, 1995, 1997)
measures dispositional tendencies to express positive and negative
emotions. Most relevant for Study 2 was the Positive Expressivity
subscale, which includes items such as “When I’m happy, my
feelings show” and “I laugh out loud when someone tells me a joke
that I think is funny” (? ? .75). Participants indicated their
agreement with each item on the BEQ using a scale from 1
(strongly disagree) to 7 (strongly agree). The average score on this
four-item subscale in the present sample was 22.25 (SD ? 4.49).
We used these scores to examine whether working memory ca-
pacity predicts the ability to suppress positive emotional expres-
sions above and beyond individual differences in positive expres-
sivity.
After completing the BEQ, participants completed the measure
of working memory capacity. As in Study 1, we used the OSPAN
task to measure participants’ abilities to encode and maintain
target words in memory while performing mathematical calcula-
tions. Consistent with past work and Study 1, working memory
capacity was operationalized as the number of words comprising
only word sets recalled in full. Working memory capacity scores in
the present sample ranged from 6 to 40, and the average capacity
score was 16.60 (SD ? 6.92).
During the second half of the experiment, participants watched
a humorous video clip. The 2-min clip depicted a series of comical
Fifty undergraduate students (19 men) partici-
moments (including man-on-the-street interviews and amusing
newspaper headlines) originally aired on the Tonight Show With
Jay Leno. Emotion regulation was manipulated by assigning par-
ticipants at random to one of two viewing conditions. Participants
assigned to the express condition were instructed to watch “as if
you were at home watching TV. If you have any emotional
responses to the clip please express them in whatever way is
natural and comfortable for you.” In contrast, participants assigned
to the suppress condition were instructed to “watch closely, and if
you have any emotional responses to the clip please try not to let
them show.”
Participants’ faces were videotaped by a small camera as they
watched the video clip. Immediately following the video clip,
participants reported their mood state using the Positive and Neg-
ative Affect Schedule (PANAS; Watson, Clark, & Tellegen,
1988). Participants indicated the extent to which they felt each of
10 positive (e.g., interested, inspired, excited; ? ? .90) and 10
negative (e.g., upset, scared, hostile; ? ? .73) emotional states
using a scale from 1 (very slightly or not at all) to 5 (extremely).
Because expressive suppression typically has no effect on emo-
tional experience, we did not expect mood states to vary as a
function of viewing condition, working memory capacity, or their
interaction. Last, participants were debriefed about the purpose of
the study and thanked for their participation.
Two naı ¨ve judges, working independently, subsequently viewed
the videotapes of participants’ faces and rated how much emotion
was expressed on each face using a continuous scale from 0 ? no
expression of emotion to 100 ? extreme expression of emotion.
The ratings of the judges were highly correlated (r ? .65, p ? .01),
so they were averaged into a single score to represent emotional
expression.
Results
Expression of emotion.
tive to participants lower in working memory capacity, participants
higher in capacity would more successfully suppress their emo-
tional expressions. This hypothesis was confirmed in a regression
that predicted the judges’ ratings of emotional expression from
working memory capacity (centered), viewing condition (coded
0 ? express and 1 ? suppress), gender of participant (coded 0 ?
female and 1 ? male), and all possible interaction terms. The
regression analysis revealed a significant main effect of viewing
condition (B ? ?31.56, p ? .01). Not surprisingly, participants
expressed less emotion in the suppress condition than they did in
the express condition. The main effect of gender of participant
approached statistical significance (B ? ?9.36, p ? .07), such that
male participants expressed less emotion than female participants
did. The main effect of working memory capacity on emotional
expressivity was small and nonsignificant (B ? ?.86, p ? .44).
Most important, the predicted Viewing Condition ? Working
Memory interaction was significant (B ? ?4.51, p ? .05),
whereas none of the other interaction terms approached statistical
significance (ps ? .25).
The significant interaction indicated that the effectiveness of
expressive suppression was moderated by working memory capac-
ity. Please refer to Figure 2, which shows the predicted emotional
expression values at 1 standard deviation above and 1 standard
deviation below the mean on the working memory measure. Anal-
The central hypothesis was that, rela-
1530
SCHMEICHEL, VOLOKHOV, AND DEMAREE

Page 6

yses of simple slopes revealed that participants expressed less
emotion in the suppress condition than they did in the express
condition (ps ? .01) regardless of their working memory capacity.
As predicted, among participants in the suppress condition, those
higher in working memory capacity expressed less emotion than
did participants lower in working memory capacity (p ? .01). Put
another way, working memory was not related to emotional
expression when participants simply expressed their responses,
r(24) ? .19, p ? .38. In the suppress condition, however, working
memory was significantly and negatively related to expressed
emotion, r(26) ? ?.39, p ? .05. These findings support the
hypothesis that working memory capacity facilitates the regulation
of emotional expression.
We also assessed whether the relationship between working
memory and expressive suppression remained significant even
when controlling for individual differences in the tendency to
express positive emotions. It did. Among participants in the sup-
press condition, the partial correlation between working memory
and emotional expression, controlling for trait positive expressiv-
ity, was significant, rp(23) ? ?.55, p ? .01. Working memory
capacity and trait positive expressivity correlated at r(50) ? .20,
p ? .16. Hence, working memory capacity predicted the suppres-
sion of amused expressions over and above the influence of
dispositional tendencies to express positive emotion.
Experience of emotion.
Participants reported their emotional
state on the PANAS immediately after watching the amusing video
clip. PANAS Positive Affect was uniformly high in the express
(M ? 22.92, SD ? 8.91) and suppress conditions (M ? 22.15,
SD ? 6.46) and, as expected, equally so for those high versus low
in working memory capacity. A regression analysis that predicted
self-reported positive emotion from working memory capacity
(centered), viewing condition (coded 0 ? express and 1 ? sup-
press), gender of participant (coded 0 ? female and 1 ? male), and
all possible interactions revealed no significant effects (ps ? .30).
A regression that predicted scores on the Negative Affect subscale
of the PANAS from the same predictor variables also produced
null results (ps ? .20). In summary, neither working memory nor
expressive suppression influenced participants’ subjective emo-
tional responses to the humorous video clip, but participants higher
in working memory capacity were better able to suppress the
outward expression of emotion.
Note, however, the PANAS did not measure amusement, and it
is possible that not all participants found this particular video clip
amusing. We therefore cannot rule out the possibility that partic-
ipants in the suppress condition experienced less amusement than
did those in the express condition, nor can we dismiss the possi-
bility that working memory capacity helps to reduce feelings of
amusement under instructions to suppress emotional expressions.
Discussion
The results of Study 2 permit two major conclusions. First,
working memory capacity facilitates the suppression of positive
emotional expressions. This result extends the findings of Study 1,
which examined the suppression of negative expressions. Taken
together, these two studies strongly support the hypothesis that
working memory facilitates the voluntary regulation of the expres-
sion of emotion.
Second, working memory capacity is not associated with less
emotional expression in general. Under instructions to suppress,
people higher in capacity expressed less emotion relative to people
lower in capacity. Under instructions simply to express emotion,
however, working memory capacity was not associated with emo-
tional expression. Thus, it was not the case that participants higher
in working memory capacity always expressed less emotion; they
expressed less only when they had the explicit goal of hiding
emotional expressions.
Although the first two studies revealed that working memory
was associated with the successful suppression of emotional ex-
pressions, in neither study did we find that working memory was
associated with reduced emotional experience. This pattern is
consistent with previous research that has shown that expressive
suppression has little effect on subjective emotional states. In most
previous research, not only has suppression failed to reduce the felt
experience of emotion but also the effort involved in suppressing
responses has increased physiological indicators of emotion, such
as heart rate, skin conductance, and other measures of sympathetic
nervous system activation (e.g., Demaree, Schmeichel, et al., 2006;
Gross & Levenson, 1993).
Altering emotional experience (and not just suppressing emo-
tional expressions) is a common goal of emotion regulation. Ac-
cordingly, we sought to examine the contributions of working
memory capacity to an emotion regulation strategy that alters
emotional experience. One such strategy is to appraise emotional
events in neutral or unemotional terms (e.g., Gross, 1998). The
remaining two studies therefore focused on the strategy of adopt-
ing neutral appraisals of emotional events so that we could test the
hypothesis that working memory capacity contributes to success at
modulating emotional experience.
Study 3: Working Memory and Neutral Appraisals of
Negative Emotional Stimuli
Study 3 tested the hypothesis that working memory capacity
contributes to the effectiveness of adopting neutral appraisals of
emotional stimuli as an emotion-regulation strategy. Appraising
emotional stimuli in neutral or unemotional terms is a form of
0
20
40
60
ExpressSuppress
Emotional Expression
Low WM (-1 SD)
High WM (+1 SD)
Figure 2.
working memory (WM) capacity (Study 2).
Emotional expression as a function of viewing condition and
1531
WORKING MEMORY AND EMOTION REGULATION

Page 7

antecedent-focused emotion regulation that seeks to prevent an
emotional response. Research has shown that neutral appraisals
may prevent both the experience and the expression of emotion
(e.g., Gross, 1998). We reasoned that working memory capacity
could be applied to sustaining cool thoughts in the face of hot
emotional impulses. Hence, we predicted that adopting neutral or
benign appraisals of an emotional stimulus would lead to a more
neutral emotional response among individuals higher in working
memory capacity relative to those lower in capacity.
Indirect support for the proposed relationship between working
memory and neutral appraisals was provided by research on the
control of mood-related thought. That research showed that cog-
nitive resources are crucial for mood control. One study, for
example, found that participants capably controlled their moods
unless they attempted mood control under cognitive load. Partic-
ipants who attempted mood control under cognitive load experi-
enced moods opposite to the ones they were trying to achieve
(Wegner, Erber, & Zanakos, 1993). We sought to build on those
findings in Study 3. Rather than diverting resources with a cogni-
tive load, however, we examined the extent to which individual
differences in cognitive resources (i.e., working memory) contrib-
uted to the successful regulation of emotion via neutral cognitive
appraisals.
To increase the generality of our findings, we used a different
measure of working memory in Study 3. One of the most popular
measures of working memory capacity in the human neuroimaging
and cognitive neuroscience literatures is the n-back task, in which
respondents monitor a series of stimuli and indicate whether the
current stimulus matches the one presented n-trials previously
(Gray, 2001; Jonides et al., 1997; Owen, McMillan, Laird, &
Bullmore, 2005). Participants in Study 3 completed two n-back
tasks—one verbal and the other spatial—and then watched the
same gruesome film clip used in Study 1. Some participants were
instructed to appraise the contents of the film clip in a detached
and unemotional manner, whereas others were instructed simply to
view the clip. We videotaped participants’ faces during the film
clip and afterward asked them to report how disgusted the clip had
made them feel. Further, we measured and statistically controlled
individual differences in sensitivity to aversive events so that we
could assess whether working memory contributed to sustaining
neutral appraisals over and above the influence of negative emo-
tional tendencies.
Method
Participants.
participated in a laboratory study concerning emotion and cogni-
tion. The average age of the participants was 19.42 years (SD ?
4.97; range ? 18–59).
Procedure.
Participants first completed questionnaires, in-
cluding the Behavioral Inhibition Scale (BIS) and Behavioral
Activation Scale (Carver & White, 1994). Most relevant for the
present study was the BIS, a seven-item scale that measures
sensitivity to aversive events. Sample items include “I worry about
making mistakes” and “If I think something unpleasant is going to
happen I usually get pretty worked up” (? ? .74). Participants
rated how much they agreed with each item on a scale from 1
(strongly agree) to 4 (strongly disagree). The average score on the
Seventy-one undergraduate students (35 men)
BIS in the present sample was 20.85 (SD ? 3.15) (coded so that
higher scores represent greater BIS sensitivity).
After completing the self-report measures, participants performed
two working memory tasks in counter-balanced order. One was a
spatial 2-back task and the other was a verbal 2-back task. Correla-
tions between n-back tasks and other putative measures of working
memory capacity (including OSPAN) have ranged from .13 to .55 in
previous research, with a mean correlation of .27 in published work
(Kane, Conway, Miura, & Colflesh, 2007; Oberauer, 2005; Shelton,
Metzger, & Elliott, 2007). The 2-back tasks used in the present study
were identical to tasks used by Gray (2001).
Each 2-back task had 100 trials. For each trial, a 4.5 ? 4.5 cm
square with a letter inside appeared at one of six possible locations
on a computer screen. The square appeared for 500 ms and was
followed by a blank-screen intertrial interval of 2,500 ms. Partic-
ipants were instructed to press one of two keys to indicate whether
the square on the current trial matched the square displayed two
trials before (i.e., press “s” for same or “d” for different). For the
spatial 2-back task, participants had to determine whether the location
of the square was the same as or different from the square presented
two trials before, ignoring the letters. Conversely, for the verbal
2-back task, participants had to determine whether the identity of
the letter in the square was the same as or different from the square
presented two trials before, ignoring the locations. Participants
were instructed to respond as quickly and accurately as possible.
Reaction time and accuracy were recorded for each trial. Trials
with accurate responses within 3,000 ms of stimulus onset were
treated as accurate trials. Only accurate trials were included when
calculating mean reaction time.
Consistent with Gray (2001), we eliminated data from partici-
pants who did not appear to understand the working memory tasks
(i.e., those with 2 standard deviations more than the average error
rate). Two participants were excluded for high error rates on the
verbal 2-back task, and 2 more were excluded due to high error
rates on both verbal and spatial tasks. Following these exclusions,
data from 67 participants (34 male) remained for further analysis.
For these participants, the average response time on the verbal
2-back task was 973.69 ms (SD ? 248.03), with an average error
rate of 12.75% (SD ? 7.93). For the spatial 2-back task, the
average response time was 934.42 ms (SD ? 251.52), and the error
rate was 13.46% (SD ? 7.95). Response times for the 2-back tasks
were highly correlated (r ? .71, p ? .001), so we computed an
average response time to serve as our measure of working memory
capacity, consistent with past research (e.g., Gray, 2001; Gray &
Braver, 2002; Shackman et al., 2006). Response times were not
associated with error rates (ps ? .60), indicating little evidence of
a speed versus accuracy trade-off.
Following completion of the 2-back tasks, participants watched
a film clip that depicted scenes of animal mutilation and slaughter
(see Study 1 for additional details). Participants were randomly
assigned to watch the clip either in a natural manner (express
condition) or in a detached, unemotional manner (neutral-appraisal
condition). More specifically, participants in the express condition
were asked to “view the film as you normally would, as if you
were watching it at home.” Conversely, those in the neutral-
appraisal condition were instructed to “adopt a detached and
unemotional attitude. Try to think objectively or about the techni-
cal aspects of the film” (see Richards & Gross, 2000, for similar
instructions).
1532
SCHMEICHEL, VOLOKHOV, AND DEMAREE

Page 8

As participants viewed the film clip, their faces were videotaped by
a small camera. Immediately following the film clip, participants
reported their affective responses using the Discrete Emotions Ques-
tionnaire (borrowed from Gross & Levenson, 1995). The Discrete
Emotions Questionnaire asks participants to rate the extent to which
they experienced each of 16 emotions during the film clip (from 1 ?
not at all to 9 ? most I’ve ever felt). We focused our analysis on
responses to the disgust item. Although this film clip has been
shown to increase negative emotional responses generally, it elicits
higher levels of disgust than it does sadness, anger, or other
discrete negative emotions (Schmeichel et al., 2006). At the end of
the study, participants were debriefed regarding the purpose of the
study and thanked for their participation.
Two naı ¨ve judges, working independently, subsequently viewed
the videotapes of participants’ faces and rated how much emotion
was expressed on each face using a continuous scale from 0
(neutral emotional expression) to 100 (extreme emotional expres-
sion). The ratings of the judges were highly correlated (r ? .65,
p ? .01), so they were averaged together to form a single measure
of emotional expression.
Results
Experience of emotion.
to participants lower in working memory capacity, participants
higher in capacity (i.e., those with faster reaction times) would
more successfully adopt neutral appraisals of a negative emotional
stimulus. Specifically, we predicted that working memory would
be associated with the experience of disgust in the neutral-
appraisal condition but not in the express condition. This hypoth-
esis was confirmed in a regression that predicted self-reported
disgust from average working memory reaction time (centered),
viewing condition (coded 0 ? express and 1 ? neutral appraisal),
gender of participant (coded 0 ? female and 1 ? male), and all
possible interaction terms. The only significant main effect was for
the viewing condition variable (B ? ?1.03, p ? .05), such that
participants reported less disgust in the neutral-appraisal condition
than they did in the express condition. Most important, the pre-
dicted Viewing Condition ? Working Memory interaction was
significant (B ? ?0.004, p ? .05), whereas none of the other
interaction terms approached statistical significance (ps ? .16).
The significant interaction indicated that the effectiveness of
neutral cognitive appraisals was moderated by working memory
capacity. Please refer to Figure 3, which depicts the predicted
values of self-reported disgust at 1 standard deviation above and 1
standard deviation below the mean on the working memory mea-
sure. Among participants in the neutral-appraisal condition, higher
working memory capacity was associated with less disgust,
r(35) ? ?.43, p ? .01. Among participants in the express condi-
tion, however, working memory was not associated with self-
reported disgust, r(32) ? .01, p ? .98. Thus, working memory
capacity influenced emotional experience when participants at-
tempted to appraise a negative stimulus in a neutral manner but not
when participants simply viewed the stimulus.
Next we assessed whether the Working Memory ? Viewing Con-
dition interaction remained significant after controlling for individual
differences in sensitivity to negative events (i.e., BIS). It did. Among
participants in the neutral-appraisal condition, the partial correlation
between working memory and self-reported disgust, controlling for
Our main hypothesis was that, relative
BIS, was significant, rp(32) ? .44, p ? .01. Hence, working memory
capacity predicted less disgust over and above the influence of dis-
positional sensitivities to negative stimuli.
Expression of emotion.
We also examined the interactive ef-
fects of working memory and viewing condition on emotional
expression. We had predicted that, relative to participants lower in
working memory capacity, participants higher in capacity would
express less emotion under instructions to appraise a negative
stimulus in neutral terms. A regression analysis that predicted the
judges’ ratings of emotional expression from working memory
capacity (centered), viewing condition (coded 0 ? express and 1 ?
neutral appraisal), gender of participant (coded 0 ? female and
1 ? male), and all possible interaction terms revealed only two
significant main effects, one for viewing condition (b ? ?20.88,
p ? .05) and one for gender of participant (b ? ?17.20, p ? .05).
Participants expressed less emotion in the neutral-appraisal condi-
tion than they did in the express condition, and male participants
expressed less emotion than female participants did. Neither the
predicted interaction between working memory capacity and view-
ing condition (b ? .04, p ? .26) nor any of the other interaction
terms was statistically significant (ps ? .22).2
Discussion
Previous research has demonstrated the effectiveness of adopt-
ing neutral cognitive appraisals as an emotion-regulation strategy.
Study 3 found evidence that working memory modulates the
effectiveness of adopting neutral appraisals. Under instructions to
adopt a neutral attitude prior to a gruesome film stimulus, partic-
2We repeated the analysis including trait negative expressivity (from the
BEQ; Gross & John, 1995, 1997) as a covariate. The main effect of
viewing condition remained significant (b ? ?20.87, p ? .05), and neither
the main effect of working memory capacity nor the Working Memory ?
Viewing Condition interaction approached significance (ps ? .40). In this
sample, the average score on the Negative Expressivity subscale was 20.74
(SD ? 6.21; ? ? .78). Working memory capacity and negative expressivity
correlated at r(67) ? .10, p ? .44.
0
1
2
3
4
5
6
7
8
9
ExpressNeutral Appraisal
Feelings of Disgust
Low WM (-1 SD)
High WM (+1 SD)
Figure 3.
working memory (WM) capacity (Study 3).
Feelings of disgust as a function of viewing condition and
1533
WORKING MEMORY AND EMOTION REGULATION

Page 9

ipants with higher working memory capacity (i.e., faster reaction
times on verbal and spatial n-back tasks) experienced less disgust
relative to those lower in working memory capacity.
The findings from Study 3 are notable for several reasons. First,
they suggest that working memory facilitates the ability to adopt
neutral cognitive appraisals, a relatively low-stress emotion-
regulation strategy that is associated with psychological well-being
(Gross & John, 2003). Second, the finding that working memory
capacity did not relate to emotional expression or experience under
the nonregulatory (i.e., express) condition replicated Study 2. In
both Study 2 and Study 3, only under instructions to regulate
emotional responses did the contributions of working memory
emerge. Third, the working memory measure used in Study 3 was
different from the measure used in the previous studies, suggesting
that the relationship between working memory and emotion reg-
ulation generalizes across assessment techniques.
One limitation of Study 3 was its focus on the neutral appraisal
of a negative emotional stimulus. Although this focus was in
keeping with the majority of research on emotion regulation, our
hypothesis holds that working memory capacity facilitates emotion
regulation regardless of whether the target emotion is positive or
negative. Hence, a test of working memory’s contribution to the
neutral appraisal of a positive stimulus is needed. Further, although
working memory clearly contributed to lower feelings of disgust
under instructions to adopt a neutral appraisal, the relationship
between working memory and emotional expression was not sig-
nificant. We conducted a fourth and final study to address these
concerns.
Study 4: Working Memory and Neutral Appraisals of
Positive and Negative Emotional Stimuli
In Study 4 we examined the neutral appraisal of both positive
and negative emotional stimuli. Previous research has focused
almost exclusively on negative stimuli, so including positive stim-
uli extended previous research and allowed us to assess the gen-
eralizability of the relationship between working memory and
neutral appraisals across distinct emotional states.
Unlike the previous studies, Study 4 sampled only students who
scored high or low on a measure of working memory capacity (i.e.,
OSPAN task). Several theorists have suggested that low working
memory capacity is underrepresented in college student samples
like the samples tested in Studies 1–3 (e.g., Kane et al., 2004; Shah
& Miyake, 1996). Partly for this reason, many prominent investi-
gations of working memory have sampled students from the ex-
tremes of the working memory capacity distribution (e.g., Conway
& Engle, 1994; Heitz, Schrock, Payne, & Engle, 2007; Rosen &
Engle, 1998). In the current study, we sampled from the extremes
to ensure that the particular attributes (i.e., low capacity and high
capacity) crucial to our hypothesis were adequately represented.
Hence, the extreme-groups approach strengthened the internal
validity of our test of the hypothesis that working memory con-
tributes to the effectiveness of adopting neutral appraisals of
emotional stimuli. We considered internal validity to be particu-
larly desirable in light of the equivocal findings for emotional
expression in Study 3. Although extreme groups may be useful for
establishing the existence of a relationship between working mem-
ory and the effectiveness of neutral appraisals, note that the
extreme-groups approach may overestimate the magnitude of such
a relationship (Conway et al., 2005; Preacher, Rucker, MacCul-
lum, & Nicewander, 2005). Further, the selection of extreme
groups renders a nonnormal sample distribution, so treating their
scores as a continuous measure is not appropriate. We therefore
analyzed the data from Study 4 using analysis of variance.
Study 4 also separated the measurement of working memory
capacity from the measurement of emotion regulation by a period
of 1 week. Previous research has found that performing two
demanding self-regulatory tasks in succession temporarily depletes
a limited resource for self-regulation (see Baumeister, Schmeichel,
& Vohs, 2007; Schmeichel, 2007). Because all participants in
Studies 1–3 completed the working memory assessment before the
emotion regulation task, one might argue that participants lower in
working memory capacity were more depleted by the working
memory assessment relative to those higher in capacity, and this
difference may have contributed to the observed relationships
between working memory and emotion regulation. Study 4 sought
to eliminate the possibility of a resource-depletion confound by
assessing working memory and emotion regulation on different
days.
Method
Participants.
ported to a laboratory for a study concerning memory and emotion.
The average age of the participants was 18.67 years (SD ? 0.87;
range ? 18–21).
Procedure.
At an initial laboratory session, participants com-
pleted questionnaires and a working memory test. As in Studies 1
and 2, we used the OSPAN task to measure individual differences
in working memory capacity. We then invited the top 21 scorers on
the OSPAN task (high working memory capacity) and the bottom
21 scorers on the OSPAN task (low working memory capacity) to
return for a follow-up experiment. Participants in the high-capacity
group had an average working memory capacity score of 26.50
(SD ? 5.64), whereas participants in the low-capacity group had
an average capacity score of 12.32 (SD ? 5.30).
In the follow-up experiment, high-capacity and low-capacity
participants were instructed to adopt neutral appraisals prior to
viewing either a funny video clip or a sad video clip. The funny
clip was the same clip used in Study 2. The sad clip was new; it
depicted children discussing family hardships (e.g., a young boy
describes the death of his infant brother). Each clip lasted 2 min.
The video clip participants viewed was determined by random
assignment. The funny clip was viewed by 10 high-capacity and 11
low-capacity participants, whereas the sad clip was viewed by 11
high-capacity and 10 low-capacity participants. All participants
were instructed to “pay close attention, but try to adopt a detached
and unemotional attitude as you watch. Please try to think about
what you are seeing objectively, in such a way that you don’t feel
anything at all.”
After participants had indicated they understood the viewing
instructions, the video clip started. Participants’ faces were video-
taped by a small camera as they watched the video presentation.
Immediately following the video clip, participants reported their
mood state using the UWIST Mood Adjective Checklist (Mat-
thews et al., 1990). For all participants, we analyzed responses to
the Hedonic Tone subscale, which reflects the relative balance
between positive mood (i.e., happy, cheerful, contented, satisfied;
Sixty-three undergraduate students (25 men) re-
1534
SCHMEICHEL, VOLOKHOV, AND DEMAREE

Page 10

? ? .91) and negative mood (i.e., sad, depressed, sorry, dissatis-
fied; ? ? .83). Positive mood and negative mood correlated at
r(42) ? ?.74, p ? .001. At the end of the study, participants were
debriefed regarding the purpose of the study and thanked for their
participation.
Three naı ¨ve judges, working independently, subsequently
viewed the videotapes of participants’ faces and rated how much
emotion was expressed on each face using a continuous scale from
0 ? no expression of emotion to 100 ? extreme expression of
emotion. As in the previous studies, the ratings of the judges were
highly correlated (rs ? .60, ps ? .01), so they were averaged into
a single score to represent emotional expression (? ? .86).
Results
Experience of emotion.
to participants low in working memory capacity, participants high
in capacity would more capably regulate the subjective experience
of emotion by adopting neutral appraisals of emotional stimuli.
More precisely, we predicted that participants high in capacity
would report a less positive hedonic tone after the funny video and
a more positive hedonic tone after the sad video compared to
participants low in capacity. This hypothesis was supported in a 2
(Working Memory: high vs. low) ? 2 (Video Clip: funny vs.
sad) ? 2 (Gender of Participant: male vs. female) analysis of
variance on hedonic tone measured immediately after the video
clips. The main effect of video clip was significant, such that
participants reported a more positive hedonic tone after the funny
clip versus the sad clip, F(1, 34) ? 30.78, p ? .01, partial ?2?
.48. Neither the main effect of working memory capacity (F ? 1,
partial ?2? .002) nor the main effect of gender of participant, F(1,
34) ? 2.37, p ? .13, partial ?2? .07, was statistically significant.
Most important, the Working Memory ? Video Clip interaction
was significant and in line with predictions, F(1, 34) ? 6.74, p ?
.05, partial ?2? .17, whereas none of the other interaction terms
approached statistical significance (Fs ? 1).3The Working Mem-
ory ? Video Clip interaction is displayed in Figure 4.
Simple effects tests revealed that, after the funny clip, partici-
pants high in working memory capacity reported less positive
hedonic tone compared to participants low in capacity, F(1, 38) ?
11.46, p ? .01. After the sad clip, high-capacity participants
reported more positive hedonic tone compared to low-capacity
participants, but this difference fell short of conventional levels of
statistical significance, F(1, 38) ? 3.67, p ? .07. Put differently,
among participants low in capacity, hedonic tone was less positive
after the sad clip than it was after the funny clip, F(1, 38) ? 41.45,
p ? .01. Among those high in working memory capacity, however,
hedonic tone was statistically equivalent after the two videos, F(1,
38) ? 2.18, p ? .15. These results suggest that adopting neutral
appraisals for emotional stimuli more effectively minimized emo-
tional experience among participants high versus low in working
memory capacity.
Expression of emotion.
We also assessed the contributions of
working memory to emotional expression under instructions to
appraise emotional stimuli in neutral or unemotional terms. Our
hypothesis was that, relative to participants low in working mem-
ory capacity, participants high in capacity would express less
emotion while adopting neutral appraisals of emotionally charged
video clips. This hypothesis was confirmed in a 2 (Working
Our main hypothesis was that, relative
Memory) ? 2 (Video Clip) ? 2 (Gender of Participant) analysis of
variance on the judges’ ratings of participants’ emotional expres-
sion. The results are displayed in Figure 5. Please note that
emotional expression was judged without regard to valence of
expression. A significant main effect of working memory capacity
indicated that participants high in capacity expressed less emotion
than did participants low in capacity, F(1, 34) ? 4.40, p ? .05,
partial ?2? .11. The main effect of video clip was also significant,
such that participants expressed less emotion during the sad clip
than they did during the funny clip, F(1, 34) ? 5.45, p ? .05,
partial ?2? .14. The main effect of gender was not significant
(F ? 1). Furthermore, none of the interaction terms approached
statistical significance (Fs ? 1).
Simple effects tests revealed that, compared to participants low
in working memory capacity, participants high in capacity ex-
pressed less emotion during both the sad clip, F(1, 38) ? 3.94, p ?
3As in the previous studies, we assessed whether the relationship
between working memory and emotion regulation would remain even after
controlling for trait affective tendencies. We used the General Dimension
Scales of the PANAS–X (Watson & Clark, 1994) to measure individual
differences in the tendencies to experience positive affect (e.g., interested,
inspired, excited; ? ? .78) and negative affect (e.g., upset, afraid, nervous;
? ? .82), respectively. The average score on the Positive Affect subscale
in the present sample was 33.02 (SD ? 5.21), and the average score on the
Negative Affect subscale was 17.05 (SD ? 4.97).
Using the Positive Affect and Negative Affect subscales of the
PANAS–X as covariates and hedonic tone as the dependent variable, we
found once again a significant Video Clip ? Working Memory interaction,
F(1, 36) ? 8.84, p ? .01. This result indicates that working memory
facilitated the successful adoption of neutral appraisals and therefore
produced more neutral emotional experience above and beyond the influ-
ence of trait affective tendencies.
0
10
20
30
40
p i lC daSp i lC ynnuF
Video Clip
Hedonic Tone
Low WM
High WM
†
*
Figure 4.
(WM) capacity (Study 4). Higher values reflect more positive hedonic
tone.†p ? .10.?p ? .05.
Hedonic tone as a function of video clip and working memory
1535
WORKING MEMORY AND EMOTION REGULATION

Page 11

.05, and the funny clip, F(1, 38) ? 3.97, p ? .05.4Taken together,
these results reveal that attempting to adopt neutral appraisals
more effectively minimized emotional expression among partici-
pants high versus low in working memory capacity.
Discussion
Compared to participants low in working memory capacity,
participants high in capacity expressed and experienced less emo-
tion when they attempted to appraise otherwise emotional stimuli
in unemotional terms. Moreover, high working memory capacity
facilitated the effectiveness of neutral appraisals regardless of
whether the target stimulus was positive or negative in emotional
tone. Thus, in addition to verifying a relationship between working
memory and the effectiveness of neutral cognitive appraisals, the
results of Study 4 extend previous research by showing that
adopting neutral appraisals may minimize or prevent responses to
both positive and negative stimuli.
Study 4 was similar to Study 3 insofar as both studies found
evidence linking working memory to the effective use of neutral
appraisals as a means to regulate emotional experience. The stud-
ies had some notable differences as well. First, whereas Study 3
measured the subjective consequences of reappraisal using a mea-
sure of discrete emotional experience (i.e., disgust), Study 4 used
a more general measure of mood (i.e., hedonic tone). Working
memory capacity was associated with more neutral emotional
experience in both studies. Second, the two studies used different
measures of working memory capacity, so the consistent findings
across the two studies suggest that the relationship between work-
ing memory and neutral appraisals is robust across different mea-
sures of working memory. Third, whereas Study 3 found a non-
significant relationship between working memory capacity and
emotional expression, Study 4 found a significant negative rela-
tionship between working memory capacity and emotional expres-
sion. Study 4 sampled only participants high or low in working
memory capacity, and this likely contributed to the stronger result
in Study 4, although we cannot rule out the possibility that differ-
ences in the working memory measures (i.e., n-back tasks in Study
3 versus OSPAN task in Study 4) contributed to the observed
results. Fourth and last, Study 4 manipulated negative versus
positive emotional states but did not manipulate emotion regula-
tion (i.e., all participants were instructed to adopt neutral apprais-
als), whereas Study 3 did manipulate emotion regulation but did
not manipulate emotional states (i.e., all participants viewed a
negative film clip).
In sum, the conservative conclusion to draw from Studies 3 and
4 is that attempting to adopt neutral cognitive appraisals effec-
tively prevents a strong inner emotional response, and especially
so among individuals higher in working memory capacity. Neutral
appraisals also effectively minimize emotional expression, but
working memory may play a less decisive role in this regard. The
strongest evidence for working memory’s contribution to emo-
tional expression under neutral appraisals emerged when high-
capacity participants were compared to low-capacity participants
(Study 4). When we analyzed a continuous range of working
memory capacities (Study 3), the relationship between working
memory and emotional expression under neutral appraisals was
less robust.
In light of previous work supporting the ecological utility of
adopting neutral appraisals of emotional stimuli, the results from
Studies 3 and 4 suggest that high working memory capacity may
benefit real-world efforts at emotion regulation. For example,
previous research found that participants who appraised a stressful
speaking task in unemotional terms experienced less negative
affect, less autonomic arousal, and better recall of the content of
their talk relative to those who suppressed their emotional re-
sponses (Egloff, Schmukle, Burns, & Schwerdtfeger, 2006). Such
findings are consistent with laboratory-based research on the sal-
utary effects of reappraisal (e.g., Gross, 1998; Richards & Gross,
2000) and, in conjunction with the present work, underscore the
potential benefits of working memory for everyday emotion reg-
ulation.
General Discussion
In this article, we proposed that a cognitive ability factor—
working memory capacity—is important to success at emotion
regulation. The results of four studies supported the hypothesis that
people higher in working memory capacity regulate their emo-
tional responses more successfully than other people do. First,
working memory capacity correlated with the successful suppres-
sion of facial expressions of negative emotion (Study 1). Second,
working memory capacity correlated with reduced expression of
positive emotion in an expressive-suppression condition but did
not correlate with emotional expression in a free-expression (i.e.,
no-regulation) condition (Study 2). Third and last, working mem-
ory capacity predicted more neutral emotional experience (Study 3
and Study 4) and less emotional facial expressions (Study 4) under
instructions to adopt neutral appraisals of emotional stimuli.
4Emotional expression was not significantly related to emotional expe-
rience (i.e., hedonic tone) in the funny clip condition, r(21) ? .21, p ? .35,
or in the sad clip condition, r(21) ? .17, p ? .46.
0
10
20
30
40
p i lC daS p i lC ynnuF
Video Clip
Emotional Expression
Low WM
High WM
Figure 5.
memory (WM) capacity (Study 4). Within each video clip type, the value
of the statistical significance test comparing high- versus low-capacity
participants is p ? .05.
Emotional expression as a function of video clip and working
1536
SCHMEICHEL, VOLOKHOV, AND DEMAREE

Page 12

The four studies reported here used diverse methods and exper-
imental designs to ascertain the relationship between working
memory and emotion regulation. For example, Studies 1 and 2
focused on expressive suppression, a response-focused form of
emotion regulation that intends to alter the outer expression of an
emotional response, whereas Studies 3 and 4 focused on adopting
neutral cognitive appraisals, an antecedent-focused form of emo-
tion regulation that is intended to prevent an emotional response.
We found that working memory capacity facilitated success at
both forms of emotion regulation. Further, whereas Studies 1 and
4 required all participants to engage in emotion regulation, Studies
2 and 3 included free-expression conditions in which some partic-
ipants did not engage in emotion regulation. The results from the
free-expression conditions led us to conclude that working mem-
ory contributes to emotion-regulation ability specifically rather
than emotional responding more generally (i.e., in the absence of
regulatory efforts).
Previous research regarding individual differences moderators
of emotion regulation has tended to rely on self-reported measures,
for both the relevant individual difference variables and the
emotion-regulation outcomes (e.g., Gross & John, 2003; Kashdan
et al., 2008; Mennin et al., 2005). By contrast, all four studies in
the present investigation used behavioral measures of working
memory capacity and both self-reported and behavioral (expres-
sive) measures of emotion regulation. This approach reduced the
likelihood that shared method variance or self-report biases in-
flated the relationship between the key individual difference vari-
able (i.e., working memory capacity) and success at emotion
regulation.
In the present studies, we asked participants to report their
emotional states so that we could examine the regulation of emo-
tional experience. We found that higher working memory capacity
predicted superior regulation of emotional experience when par-
ticipants attempted to appraise emotional stimuli in a neutral or
dispassionate manner but not when they attempted to suppress
emotional expressions. This dissociation between the subjective
consequences of neutral appraisals versus suppression replicates
previous work (e.g., Gross, 1998; Richards & Gross, 2000) and
argues against a simple social desirability explanation for the
results. If the successful regulation of experience among high-
capacity participants in the neutral-appraisal conditions reflected a
simple tendency for those participants to report desirable emotions,
then we would have observed differences in emotional experience
as a function of working memory capacity in the suppression
conditions as well. We found no such differences.
Alternative Explanations
We have argued that working memory capacity facilitates the
regulation of emotional expressions (Studies 1, 2, and 4) and
experiences (Studies 3 and 4), but the correlational nature of our
findings precludes us from making a strong causal claim. It is
possible that an unmeasured third variable explains the relation-
ship between working memory and emotion regulation, though we
ruled out two plausible alternatives—trait affective tendencies and
trait emotional expressivity. It is also possible that, rather than
working memory facilitating emotion regulation, good emotion
regulation facilitates working memory. This intriguing possibility
does not contradict the significant relationships we observed be-
tween working memory capacity and emotion regulation. Rather, it
simply suggests we are not able to say definitively in which
direction the causal arrow flows, whether from high capacity to
good emotion regulation or from good emotion regulation to high
capacity.
The current studies focused exclusively on efforts to down-
regulate or to minimize emotional responses. The extent to which
working memory capacity contributes to the up-regulation of an
emotional response remains to be seen. Moreover, the relationship
between working memory and emotion regulation may be due in
part to differences in effort expenditure. For example, participants
higher in working memory capacity may have expended more
effort on the emotion-regulation tasks relative to participants lower
in capacity. Some previous evidence argues against this view
insofar as individuals lower in working memory capacity ex-
pended more effort than individuals higher in capacity did, at least
in some circumstances (Heitz et al., 2007). Future research that
assesses the relationship between working memory capacity and
activation of the sympathetic nervous system (e.g., skin conduc-
tance, preejection period) during emotion regulation would help to
clarify any potential role of effort expenditure.
Implications
The current results have several practical and theoretical impli-
cations. One practical implication concerns the treatment of mood
and anxiety disorders. On the basis of the current findings, treat-
ment providers may expect people higher in working memory
capacity to benefit from cognitive-oriented treatment strategies
more readily than people lower in capacity would. For example,
some forms of cognitive behavioral therapy encourage depressed
individuals to stop their automatic negative thoughts and to gen-
erate alternative thoughts. This approach to emotion regulation is
highly similar to cognitive reappraisal, and indeed some treatment
protocols teach reappraisal skills to combat anxiety and mood
disorders (e.g., Campbell-Sills & Barlow, 2007). Insofar as people
high in working memory capacity are more successful at adopting
neutral appraisals of emotional events, they may benefit more
readily from treatments that encourage neutral appraisals to min-
imize negative emotions. People low in capacity, by contrast, may
require more extensive training and treatment to benefit from
neutral appraisals. Note, however, that depression and anxiety are
associated with reduced capacity for controlled processing (e.g.,
Dalgleish et al., 2007), and it remains to be seen whether individ-
ual differences in working memory contribute to treatment out-
comes, or whether mood and anxiety disorders undermine working
memory capacity and render moot any potential benefit of high
capacity.
The present findings also suggest that individual differences in
working memory capacity may represent a risk factor for the
development of mood and anxiety disorders. We observed that
people lower in capacity were less successful at regulating their
emotional responses. It may be the case, then, that people lower in
capacity are more susceptible to depression or other mood disor-
ders in response to traumatic experiences, such as the death of a
loved one or debilitating personal injury. Some evidence suggests
that people who score lower on intelligence tests are more likely to
develop posttraumatic stress in response to highly traumatic events
relative to people who score higher on such measures (Kremen et
1537
WORKING MEMORY AND EMOTION REGULATION

Page 13

al., 2007; Macklin et al., 1998). Given that working memory
capacity and intelligence are related (Conway, Kane, & Engle,
2003; Engle, Tuholski, Laughlin, & Conway, 1999), it is plausible
that working memory capacity and the susceptibility to mood
disorders are related. At this point, however, the association be-
tween working memory and susceptibility to mood and anxiety
disorders is unknown.
A related question is whether working memory capacity is
associated with long-term emotional outcomes among populations
with no mood disorders. In the current research, high-capacity
participants used neutral appraisals to minimize the subjective
experience of emotion more successfully than low-capacity par-
ticipants did. The use of neutral appraisals to regulate unwanted
emotions has been associated with higher subjective well-being
(Gross & John, 2003). Insofar as people higher in capacity use
neutral appraisals to regulate unwanted emotions more effectively
than other people do, working memory capacity may predict
subjective well-being. Research investigating the relationship be-
tween working memory and long-term emotional outcomes ap-
pears warranted (see Robinson, 2007).
The current findings also have implications for research on
cognitive control and working memory. The results revealed that
working memory capacity is associated with the regulation of
emotional responses. This finding builds on previous research on
individual differences in working memory capacity, which has
focused almost exclusively on the extent to which working mem-
ory relates to other cognitive processes. The fact that working
memory capacity is associated with the regulation of emotional
and cognitive responses alike speaks to the issue of whether
cognitive control is best construed as a unitary capacity, or whether
cognitive control is best construed as a collection of distinct
abilities. The current findings lend support to the unitary view,
insofar as two apparently distinct abilities—maintaining informa-
tion in working memory and regulating emotional responses—
were correlated. This finding is consistent with factor analytic
evidence that various measures of cognitive control, including
performance on the OSPAN task, load on a common factor (e.g.,
Kane et al., 2004). On the basis of the current results, we would
predict that measures of emotion regulation load on that same
factor as well.
Conclusion
The present research provides the first evidence that individual
differences in working memory are associated with the self-
regulation of emotion. Working memory’s contribution to emotion
regulation was robust across two distinct forms of emotion regu-
lation (expressive suppression and neutral appraisals) and across
diverse emotional states (positive and negative). These findings
indicate that people higher in working memory capacity are more
adept than others are at controlling their emotional expressions and
experiences.
References
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Ashcraft, M. H., & Kirk, E. P. (2001). The relationships among working
memory, math anxiety, and performance. Journal of Experimental Psy-
chology: General, 130, 224–237.
Baumeister, R. F., Schmeichel, B. J., & Vohs, K. D. (2007). Self-regulation
and the executive function: The self as controlling agent. In A. W.
Kruglanski & E. T. Higgins (Eds.), Social psychology: Handbook of
basic principles. (2nd ed., pp. 516–539). New York: Guilford.
Beilock, S. L., & Carr, T. H. (2005). When high-powered people fail:
Working memory and “choking under pressure” in math. Psychological
Science, 16, 101–105.
Campbell-Sills, L., & Barlow, D. H. (2007). Incorporating emotion regu-
lation into conceptualizations and treatments of anxiety and mood dis-
orders. In J. J. Gross (Ed.), Handbook of emotion regulation (pp.
542–559). New York: Guilford.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral
activation, and affective responses to impending reward and punishment:
The BIS/BAS Scales. Journal of Personality and Social Psychology, 67,
319–333.
Conway, A. R. A., Cowan, N., & Bunting, M. F. (2001). The cocktail party
phenomenon revisited: The importance of working memory capacity.
Psychonomic Bulletin and Review, 8, 331–335.
Conway, A. R. A., & Engle, R. W. (1994). Working memory and retrieval:
A resource-dependent inhibition model. Journal of Experimental Psy-
chology: General, 123, 354–373.
Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm,
O., & Engle, R. W. (2005). Working memory span tasks: A method-
ological review and user’s guide. Psychonomic Bulletin and Review, 12,
769–786.
Conway, A. R. A., Kane, M. J., & Engle, R. W. (2003). Working memory
capacity and its relation to general intelligence. Trends in Cognitive
Sciences, 7, 547–552.
Dalgleish, T., Williams, J. M. G., Golden, A. J., Perkins, N., Feldman-
Barrett, L., Barnard, P. J., et al. (2007). Reduced specificity of autobio-
graphical memory and depression: The role of executive control. Journal
of Experimental Psychology: General, 136, 23–42.
Darke, S. (1988). Effects of anxiety on inferential reasoning task perfor-
mance. Journal of Personality and Social Psychology, 55, 499–505.
de Fockert, J. W., Rees, G., Frith, C. D., & Lavie, N. (2001, March 2). The
role of working memory in visual selective attention. Science, 291,
1803–1806.
Demaree, H. A., Robinson, J. L., Pu, J., & Allen, J. J. B. (2006). Strategies
actually employed during response-focused emotion regulation research:
Affective and physiological consequences. Cognition and Emotion, 20,
1248–1260.
Demaree, H. A., Schmeichel, B. J., Robinson, J. L., & Everhart, D. E. (2004).
Behavioral, affective, and physiological effects of negative and positive
emotional exaggeration. Cognition and Emotion, 18, 1079–1097.
Demaree, H. A., Schmeichel, B. J., Robinson, J. L., Pu, J., Everhart, D. E.,
& Berntson, G. G. (2006). Up- and down-regulating facial disgust:
Affective, vagal, sympathetic, and respiratory consequences. Biological
Psychology, 71, 90–99.
Dimberg, U., Thunberg, M., & Grunedal, S. (2002). Facial reactions to
emotional stimuli: Automatically controlled emotional responses. Cog-
nition and Emotion, 16, 449–471.
Edelstein, R. S. (2006). Attachment and emotional memory: Investigating
the source and extent of avoidant memory impairments. Emotion, 6,
340–345.
Egloff, B., Schmukle, S. C., Burns, L. R., & Schwerdtfeger, A. (2006).
Spontaneous emotion regulation during evaluated speaking tasks: Asso-
ciations with negative affect, anxiety expression, memory, and physio-
logical responding. Emotion, 6, 356–366.
Engle, R. W. (2001). What is working-memory capacity? In H. L. Roediger
III & J. S. Nairne (Eds.), The nature of remembering: Essays in honor
of Robert G. Crowder (pp. 297–314). Washington, DC: American Psy-
chological Association.
Engle, R. W., Kane, M. J., & Tuholski, S. W. (1999). Individual differ-
ences in working memory capacity and what they tell us about controlled
1538
SCHMEICHEL, VOLOKHOV, AND DEMAREE

Page 14

attention, general fluid intelligence and functions of the prefrontal cor-
tex. In A. Miyake & P. Shah (Eds.), Models of working memory:
Mechanisms of active maintenance and executive control (pp. 102–134).
New York: Cambridge University Press.
Engle, R. W., Tuholski, S. W., Laughlin, J. E., & Conway, A. R. A. (1999).
Working memory, short term memory and general fluid intelligence: A
latent variable approach. Journal of Experimental Psychology: General,
128, 309–331.
Gray, J. R. (2001). Emotional modulation of cognitive control: Approach-
withdrawal states double-dissociate spatial from verbal two-back task per-
formance. Journal of Experimental Psychology: General, 130, 436–452.
Gray, J. R., & Braver, T. S. (2002). Personality predicts working memory
related activation in caudal anterior cingulate cortex. Cognitive, Affec-
tive, and Behavioral Neuroscience, 2, 64–75.
Gross, J. J. (1998). 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. (2002). Emotion regulation: Affective, cognitive, and social
consequences. Psychophysiology, 39, 281–291.
Gross, J. J., & John, O. P. (1995). Facets of emotional expressivity: Three
self-report factors and their correlates. Personality and Individual Dif-
ferences, 19, 555–568.
Gross, J. J., & John, O. P. (1997). Revealing feelings: Facets of emotional
expressivity in self-reports, peer ratings, and behavior. Journal of Per-
sonality and Social Psychology, 72, 435–448.
Gross, J. J., & John, O. P. (2003). Individual differences in two emotion
regulation processes: Implications for affect, relationships, and well-
being. Journal of Personality and Social Psychology, 85, 348–362.
Gross, J. J., & Levenson, R. W. (1993). Emotional suppression: Physiol-
ogy, self-report, and expressive behavior. Journal of Personality and
Social Psychology, 64, 970–986.
Gross, J. J., & Levenson, R. W. (1995). Emotion elicitation using films.
Cognition and Emotion, 9, 87–108.
Heitz, R. P., Schrock, J. C., Payne, T. W., & Engle, R. W. (2007). Effects
of incentive on working memory capacity: Behavioral and pupillometric
data. Psychophysiology, 44, 1–11.
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.
Jackson, D. C., Mueller, C. J., Dolski, I., Dalton, K. M., Nitschke, J. B.,
Urry, H. L., et al. (2003). Now you feel it, now you don’t: Frontal EEG
asymmetry and individual differences in emotion regulation. Psycholog-
ical Science, 14, 612–617.
James, W. B., Scott, R. T., & Good, P. B. (Producers). (1978). Faces of
death [Motion picture]. (Available from Gorgon Video, 16101 South
108th Avenue, Orland Park, IL 60467).
Jensen-Campbell, L. A., Knack, J. M., Waldrip, A. M., & Campbell, S. D.
(2007). Do Big Five personality traits associated with self-control influ-
ence the regulation of anger and aggression? Journal of Research in
Personality, 41, 403–424.
Jonides, J., Schumacher, E. H., Smith, E. E., Lauber, E. J., Awh, E.,
Minoshima, S., & Koeppe, R. A. (1997). Verbal working memory load
affects regional brain activation as measured by PET. Journal of Cog-
nitive Neuroscience, 9, 462–475.
Kane, M. J., Bleckley, M. K., Conway, A. R. A., & Engle, R. W. (2001).
A controlled-attention view of working-memory capacity. Journal of
Experimental Psychology: General, 130, 169–183.
Kane, M. J., Conway, A. R. A., Miura, T. K., & Colflesh, G. J. H. (2007).
Working memory, attention control, and the N-back task: A question of
construct validity. Journal of Experimental Psychology: Learning, Mem-
ory, and Cognition, 33, 615–622.
Kane, M. J., & Engle, R. W. (2003). Working-memory capacity and the
control of attention: The contributions of goal neglect, response com-
petition, and task set to Stroop interference. Journal of Experimental
Psychology: General, 132, 47–70.
Kane, M. J., Hambrick, D. Z., Tuholski, S. W., Wilhelm, O., Payne, T. W., &
Engle, R. W. (2004). The generality of working memory capacity: A
latent-variable approach to verbal and visuospatial memory span and rea-
soning. Journal of Experimental Psychology: General, 133, 189–217.
Kashdan, T. B., Zvolensky, M. J., & McLeish, A. C. (2008). Anxiety
sensitivity and affect regulatory strategies: Individual and interactive
risk factors for anxiety-related symptoms. Journal of Anxiety Disorders,
22, 429–440.
Kensinger, E. A., & Corkin, S. (2003). Effects of negative emotional content
on working memory and long-term memory. Emotion, 3, 378–393.
Klein, K., & Boals, A. (2001). The relationship of life event stress and
working memory capacity. Applied Cognitive Psychology, 15, 565–579.
Kremen, W. S., Koenen, K. C., Boake, C., Purcell, S., Eisen, S. A., Franz,
C. E., et al. (2007). Pretrauma cognitive ability and risk for posttraumatic
stress disorder: A twin study. Archives of General Psychiatry, 64,
361–368.
Macklin, M. L., Metzger, L. J., Litz, B. T., McNally, R. J., Lasko, N. B.,
Orr, S. P., & Pitman, R. K. (1998). Lower precombat intelligence is a
risk factor for posttraumatic stress disorder. Journal of Consulting and
Clinical Psychology, 66, 323–326.
Matthews, G., Jones, D. M., & Chamberlain, A. G. (1990). Refining the
measurement of mood: The UWIST Mood Adjective Checklist. British
Journal of Psychology, 81, 17–42.
Mennin, D. S., Heimberg, R. G., Turk, C. L., & Fresco, D. M. (2005).
Preliminary evidence for an emotion dysregulation model of generalized
anxiety disorder. Behaviour Research and Therapy, 43, 1281–1310.
Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal
cortex function. Annual Review of Neuroscience, 24, 167–202.
Oberauer, K. (2005). Binding and inhibition in working memory: Individ-
ual and age differences in short-term recognition. Journal of Experimen-
tal Psychology: General, 134, 368–387.
Ochsner, K. N., Bunge, S. A., Gross, J. J., & Gabrieli, J. D. (2002).
Rethinking feelings: An fMRI study of the cognitive regulation of
emotion. Journal of Cognitive Neuroscience, 14, 1215–1229.
Ochsner, K. N., Ray, R. D., Cooper, J. C., Robertson, E. R., Chopra, S.,
Gabrieli, J. D. E., & Gross, J. J. (2004). For better or for worse: Neural
systems supporting the cognitive down- and up-regulation of negative
emotion. NeuroImage, 23, 483–499.
Owen, A. M., McMillan, K. M., Laird, A. R., & Bullmore, E. (2005).
N-back working memory paradigm: A meta-analysis of normative func-
tional neuroimaging studies. Human Brain Mapping, 25, 46–59.
Preacher, K. J., Rucker, D. D., MacCallum, R. C., & Nicewander, W. A.
(2005). Use of the extreme groups approach: A critical reexamination
and new recommendations. Psychological Methods, 10, 178–192.
Richards, J. M., & Gross, J. J. (2000). Emotion regulation and memory:
The cognitive costs of keeping one’s cool. Journal of Personality and
Social Psychology, 79, 410–424.
Robinson, M. D. (2007). Gassing, braking, and self-regulating: Error
self-regulation, well-being, and goal-related processes. Journal of Ex-
perimental Social Psychology, 43, 1–16.
Rosen, V. M., & Engle, R. W. (1998). Working memory capacity and
suppression. Journal of Memory and Language, 39, 418–436.
Schmader, T., & Johns, M. (2003). Converging evidence that stereotype
threat reduces working memory capacity. Journal of Personality and
Social Psychology, 85, 440–452.
Schmeichel, B. J. (2007). Attention control, memory updating, and emo-
tion regulation temporarily reduce the capacity for executive control.
Journal of Experimental Psychology: General, 136, 241–255.
Schmeichel, B. J., Demaree, H. A., Robinson, J. L., & Pu, J. (2006). Ego
depletion by response exaggeration. Journal of Experimental Social
Psychology, 42, 95–102.
Shackman, A. J., Sarinopolulos, I., Maxwell, J. S., Pizzagalli, D., Lavric,
1539
WORKING MEMORY AND EMOTION REGULATION

Page 15

A., & Davidson, R. J. (2006). Anxiety selectively disrupts visuospatial
working memory. Emotion, 6, 40–61.
Shah, P., & Miyake, A. (1996). The separability of working memory
resources for spatial thinking and language processing: An individual
differences approach. Journal of Experimental Psychology: General,
125, 4–27.
Shelton, J. T., Metzger, R. L., & Elliott, E. M. (2007). A group-
administered lab task as a measure of working memory. Behavior
Research Methods, 39, 482–493.
Smith, E. E., & Jonides, J. (1999, March 12). Storage and executive
processes in the frontal lobes. Science, 283, 1657–1661.
Turner, M. L., & Engle, R. W. (1989). Is working memory capacity task
dependent? Journal of Memory and Language, 28, 127–154.
Unsworth, N., Heitz, R. P., & Engle, R. W. (2005). Working memory
capacity in hot and cold cognition. In R. W. Engle, G. Sedek, U. Hecker,
& D. N. McIntosh (Eds.), Cognitive limitations in aging and psychopa-
thology (pp. 19–43). New York: Cambridge University Press.
Urry, H. L., Nitschke, J. B., Dolski, I., Jackson, D. C., Dalton, K. M.,
Mueller, C. J., et al. (2004). Making a life worth living: Neural correlates
of well-being. Psychological Science, 15, 367–372.
Vohs, K. D., & Schmeichel, B. J. (2003). Self-regulation and the extended
now: Controlling the self alters the subjective experience of time. Jour-
nal of Personality and Social Psychology, 85, 217–230.
Watson, D., & Clark, L. A. (1994). Manual for the Positive and Negative
Affect Schedule (expanded form). Iowa City, IA: University of Iowa.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
dation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Wegner, D. M., Erber, R., & Zanakos, S. (1993). Ironic processes in the
mental control of mood and mood related thought. Journal of Person-
ality and Social Psychology, 65, 1093–1104.
Wood, J. V., Heimpel, S. A., & Michela, J. L. (2003). Savoring versus
dampening: Self-esteem differences in regulating positive affect. Jour-
nal of Personality and Social Psychology, 85, 566–580.
Received June 5, 2007
Revision received June 18, 2008
Accepted June 27, 2008 ?
1540
SCHMEICHEL, VOLOKHOV, AND DEMAREE