International Journal of Psychophysiology 86 (2012) 291–298

Contents lists available at SciVerse ScienceDirect

International Journal of Psychophysiology

journal homepage: www.elsevier.com/locate/ijpsycho

Culture moderates the cardiovascular consequences of anger regulation strategy☆

Ting Zhou, George D. Bishop ⁎
National University of Singapore, Singapore

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: This research examined cultural differences in experiential and cardiovascular outcomes of three
Received 4 June 2012 anger regulation strategies (expression, suppression and reappraisal).
Received in revised form 16 October 2012 Methods: Forty-five Chinese and 45 Caucasian females participated in a laboratory experiment in which role
Accepted 18 October 2012 play was used to induce anger. During this role play participants were instructed to either express or
Available online 26 October 2012
suppress their feelings or engage in cognitive reappraisal. Emotional experience was measured before and
after the role play. Cardiovascular indices were measured continuously during the experiment.
Keywords:
Anger regulation
Results: Significant interactions were obtained such that Caucasians showed stronger cardiovascular
Cultural background responses to suppression than expression of anger whereas the opposite was true for Chinese.
Emotional responses Conclusions: These results demonstrate that physiological consequences of emotion regulation strategies vary
Cardiovascular reactivity by cultural background. Possible reasons as well as implications of these findings are discussed.
© 2012 Elsevier B.V. All rights reserved.

1. Introduction focused strategies, such as cognitive reappraisal, are concerned with

Anger is a negative emotion which people experience from time to
time in daily life. The successful regulation of anger is essential to the
maintenance of good social relationships (Tavris, 1989) and there is
also reason to believe that anger and its regulation are related to cardio-
vascular disease (CVD) risk (Kawachi et al., 1996; Williams et al., 2000).
In particular, a high propensity to anger is believed to result in un-
healthy levels of cardiovascular reactivity (CVR) which in turn acceler-
ates the rate of atherosclerosis (Kamarck et al., 1997; Siegman and
Smith, 1994). As such it is important to understand the processes in-
volved in the experience and regulation of anger and to explore their
cardiovascular consequences.
The Process Model of emotion regulation, proposed by Gross and his
colleagues (Gross, 1998; Gross and Levenson, 1993, 1997; John and
Gross, 2004) traces the experience and regulation of emotion through
a series of stages beginning with the selection and perception of situa-
tions, moving through the meanings given to those situations, and the
resulting emotions. The regulation of emotion is an integral part of
this process with emotion regulation strategies divided into those that
focus on antecedents to emotions (antecedent-focused strategies) and
those in which the person attempts to regulate the emotional response
(response-focused strategies). Response-focused strategies, such as
suppression, are concerned with modifying the expression of an emo-
tion that is already experienced by the person whereas antecedent-
☆ This research was supported by grant no. R-581-000-090-101 from the Faculty of
Arts and Social Sciences, National University of Singapore.
⁎ Corresponding author at: Department of Psychology, National University of Singapore, 9
Arts Link, Singapore 117570, Singapore. Tel.: +65 6518 8748; fax: +65 6773 1843.
E-mail address: psygb@nus.edu.sg (G.D. Bishop).
modifying one's interpretation of the situation so as to change emotion-
al experience.
Research to date has suggested different consequences for different
emotion regulation strategies. In particular, suppression, while it may
be successful in reducing the outward expression of negative emotions,
has been found to have a number of negative side effects. Habitual use
of suppression has been found to be related to negative affective, cogni-
tive, social and physiological effects (John and Gross, 2004). Individuals
who more frequently engage in suppression have been found to contin-
ue to experience negative emotions, have heightened physiological
reactivity to emotion evoking events, reduced memory for social infor-
mation and experience disruption in social relationships. In contrast,
individuals that are more prone to use reappraisal for dealing with neg-
ative emotions have shown reduced experience of negative emotion,
increased positive emotion, and also reduced physiological reactivity
(Gross, 1998; John and Gross, 2004; Mauss et al., 2007).
Although the results to date have been generally consistent in dem-
onstrating negative effects for suppression and positive effects for
reappraisal, cultural differences in the effects of emotion regulation
strategies have been relatively unexplored. A large body of literature
provides evidence for the universality of many emotions (Ekman,
1992, 1994; Scherer and Wallbott, 1994) but also points to the fact
that those emotions show substantial cultural variation with respect
to their antecedents and expression (Kitayama and Markus, 1994;
Scherer, 1997; Scherer and Wallbott, 1994; Scherer et al., 1988). One
key factor in this is the existence of culturally determined display
rules that dictate when it is or is not appropriate to exhibit a particular
emotion (Ekman, 1993; Friesen, 1972). In this respect there are notable
differences between North Americans and East Asians in emotion dis-
play rules with East Asians being much more likely to suppress negative

0167-8760/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijpsycho.2012.10.010
292 T. Zhou, G.D. Bishop / International Journal of Psychophysiology 86 (2012) 291–298
emotions, particularly when interacting with strangers, than is the case
for North Americans (Kitayama and Markus, 1994). Similarly, Gross and
John (2003) report that Asian Americans score higher on a question-
naire measure of emotional suppression than do European Americans.
These cultural differences raise the question of whether the effects of
different emotion regulation strategies might differ between cultures.
In cultures where emotional suppression is favored it can be expected
that suppression would become relatively automatic for the person
and thus require less effort. Under such circumstances it would seem
to follow that the negative effects of suppression described above
might be attenuated or even eliminated. Consistent with this Butler et
al. (2007) found that when European-Americans engaged in suppres-
sion during a laboratory task they showed reduced interpersonal re-
sponsiveness whereas this was not true for Asian-Americans. Further,
the same authors (Butler et al., 2009) found that when European-
American women discussed a disturbing film their expression of nega-
tive emotion was negatively related to blood pressure whereas this was
not true for Asian-American women.
In line with these considerations the research reported here exam-
ined differences in cardiovascular responses to different emotion regu-
lation strategies in response to anger among Chinese and Caucasians in
Singapore. Kitayama et al. (2006), note that anger can be considered to
be a socially disengaging emotion that is grounded in independence and
autonomy of self. In the Chinese cultural context, social harmony is
highly valued with an emphasis on maintaining interpersonal harmony
by avoiding conflicts. Open expression of anger is considered inappro-
priate in most social situations. Because of this “other orientation” in-
cluding high conformity and strong concern about social norms (Yang,
1995), Chinese regard controlling the impulse to express anger as a
good quality and pursue it as an achievement. Even when they feel
angry, there is a tendency to hide the emotion and avoid open expres-
sion. Further, since Chinese receive positive reinforcement when they
suppress anger, the usage of suppression can be expected to be more
frequent than would be the case in individualistic cultures. In addition,
with their emphasis on social harmony Chinese can be expected to be
more attuned to what others are feeling and to use reappraisal more
in situations of conflict. Evidence for this pattern is provided by
Matsumoto et al. (2008) who found that cultures emphasizing social
order had higher scores on measures of both suppression and re-
appraisal as strategies for regulating emotions. In line with this, partic-
ipants from China in their research had higher suppression and
reappraisal scores than did participants from the USA. As such we ex-
pect that Chinese participants in this study will report more use of
both suppression and reappraisal as anger regulation strategies than
will Caucasian participants.
More importantly, we expect that these differences in habitual use
of emotion regulation strategies will have an impact on the cardiovas-
cular consequences of engaging in those strategies. Evidence from stud-
ies done in North America shows higher physiological reactivity to
negative emotions when participants are instructed to suppress their
emotions (Gross, 1998; Gross and Levenson, 1997). This makes sense
in the context of social norms encouraging the expression of emotion.
In a society that emphasizes the suppression of negative feeling, howev-
er, this pattern may well not be found. Among Singapore Chinese where
norms favor the public suppression of negative emotions such as anger
and where expression of such emotions is frowned on it may well be the
case that the physiological consequences of emotion regulation strate-
gies will be different. One way of approaching this difference comes
from work showing differential cardiovascular responses as a function
of task engagement. As proposed by Obrist (1981) and elaborated by
Wright and Kirby (2001) tasks requiring greater engagement tend to
produce greater activation of the sympathetic nervous system (SNS),
particularly for beta-andrenergically mediated responses such as cardi-
ac output (CO) and pre-ejection period (PEP). In line with this since
suppression tends to be habitual in Chinese societies it can be expected
that Singapore Chinese participants will require less engagement to
suppress their emotions and will show lower levels of SNS response
to the suppression of emotion than has been found in studies in North
America. With expression, however, the opposite effect can be expected
in that expression of emotion for Chinese participants can be expected
to require more engagement than is the case for cultures, such as in
North America, where expression of emotion is encouraged. As such
the purpose of present study is to examine the experiential and cardiovas-
cular outcomes of three specific anger regulation strategies (expression,
suppression and reappraisal) in a laboratory environment and to com-
pare these outcomes between Chinese and Caucasians.
Three hypotheses are tested. First, because of an emphasis on har-
mony in relationships and restraint of emotion expression, Chinese par-
ticipants should be more likely to suppress anger across situations
compared with their Caucasian counterparts. This will be indicated by
higher scores on a measure of emotional suppression. Second, Chinese
participants should be more likely to engage in reappraisal in anger sit-
uations as indicated by higher scores on a measure of reappraisal. Third,
and most importantly, culture will have a moderating effect in the rela-
tionship between emotion regulation strategies and CVR. Overall,
Caucasians are expected to have greatest CVR, particular for beta-
andrenergically mediated responses such as CO and PEP when assigned
to anger suppression, followed by the expression and reappraisal condi-
tions. In contrast, because of an emphasis on harmony in relationships
and restraint of social expression, Chinese participants are expected to
show the strongest CVR when assigned to anger expression, followed
by the suppression and reappraisal conditions.
2. Method
2.1. Participants
Forty-five Chinese and 45 Caucasian female undergraduates attend-
ing the National University of Singapore were recruited for the experi-
ment. Chinese participants were required to have two Chinese parents
and to have been born and raised in Asia. Caucasian participants, all of
whom were exchange students, were required to have two European
or European-American parents and to have been born and raised in
either North America or Europe. None of the participants had been
diagnosed with heart disease or hypertension and none were taking
medications which might affect cardiovascular indices. Examination
of possible differences in age by ethnicity and condition found no
differences in age among the six ethnicity by condition groups, all
ps >0.50. Mean age across all groups was 20.91. Further, no differences
were obtained for self-reported family history of CVD by either ethnicity
or condition, both ps b .50. Examination of body mass index (BMI) by
ethnicity and condition found no differences for condition or the inter-
action of ethnicity by condition, both ps>.10. However, there was a
significant difference by ethnicity (20.41 vs. 21.75), F (1, 81)= 4.04,
p = .0481, η2 = .047.
2.2. Materials
2.2.1. Emotion rating scale
Before and after the role play participants were asked to rate on a
7-point Likert scale the degree to which they felt a series of 13 emo-
tions. These emotions were divided into three components: anger
(angry, frustrated, and annoyed), positive emotions (happy, joyful,
and pleased) and negative emotions (nervous, sad, anxious, guilty,
ashamed, worried, and afraid) as defined by Mauss et al. (2007). In-
ternal consistency for these indices for both baseline and task ratings
were as follows: anger Cronbach's α = .68 (baseline) and .83 (task);
positive emotions, α = .85, .86; negative emotions, α = 84, .83.
2.2.2. Emotion regulation questionnaire (ERQ)
Items of the ERQ (Gross and John, 2003) were adapted to measure
habitual usage of suppression and reappraisal in the context of anger
 T. Zhou, G.D. Bishop / International Journal of Psychophysiology 86 (2012) 291–298
provoking situations. The original version of the ERQ is concerned with
the regulation of emotions generally without reference to specific emo-
tions such as anger. As the focus of this study is anger, items in the ERQ
were modified to specifically ask about anger. For example, the item “I
control my emotions by changing the way I think about the situation
I'm in.” was changed to “I control anger by changing the way I think
about the situation I'm in.” One item in the suppression scale asking
about the suppression of positive emotions was dropped. All responses
were made on a seven point scale from 1 = strongly disagree to 7 =
strongly agree. In this sample the six items of the adapted reappraisal
scale had a Cronbach's α of 0.85 whereas the three items of the adapted
suppression scale had α of 0.87. Scores reported here are the average
item scores for each scale.
2.2.3. Cultural values measure
The Schwartz et al. (2001) portrait values questionnaire was admin-
istrated to all the participants in the present study to test the cultural
representativeness of two ethnicity groups. This instrument contains
40 items tapping seven components of cultural values including confor-
mity, tradition, benevolence, universalism, self-direction, stimulation,
hedonism, achievement, power, and security. Participants were asked
to rate how much the description in each item was like them on a
6-point scale ranging from 1 = “not at all like me” to 6 = “very much
like me” (Schwartz et al., 2001). The internal consistency of the dimen-
sions ranged from α = .63 to 82.
2.3. Physiological measures
Six measures of cardiovascular functioning were measured continu-
ously during the two baseline and task periods. For purposes of analysis
only measures for the last half of the first baseline period were included
in the calculation of baseline values whereas the full task period was in-
cluded for task values. Systolic blood pressure (SBP), diastolic blood
pressure (DBP) and heart rate (HR) were examined as overall measures
of CVR. Blood pressure reactivity, in particular has been associated with
the increased rate of atherosclerosis (Kamarck et al., 1997). Measures of
CO, total peripheral resistance (TPR) and PEP were included to further
examine the underlying hemodynamics of the blood pressure re-
sponses. For ease of interpretation PEP were multiplied by −1 so as to
have all predictions showing the same direction. The Finometer Pro
(Finapres Medical Systems Company, Amsterdam), a non-invasive
beat-to beat blood pressure monitor, was used to obtain measures of
SBP, DBP, HR, CO and TPR. The Finometer records cardiovascular param-
eters continuously using a finger cuff and can provide accurate estima-
tion of intra-arterial blood pressure. The Finometer is calibrated with an
upper arm cuff measurement using Return to Flow (RTF) technology.
RFT technology allows blood pressure to be reconstructed on set inter-
vals against brachial measurements using an upper arm cuff (Finopress
Medical Systems, 2005). The accuracy of the Finometer findings have
been supported in empirical studies comparing the Finometer results
with intra-arterial findings (Jansen et al., 2001). Internal consistency
of both the baseline and task measures were obtained by taking 1 min
averages for each parameter for the final 5 min of the baseline and all
3 min of the task period. Cronbach's α computed on these values
showed very high internal consistency with values for baseline mea-
sures ranging from .91 to .99 for baseline values and between .84 and
.98 for values during the task period.
The HIC-2000 (Bio-Impedance Technology, Inc., Chapel Hill, NC), a
non-invasive instrument for detecting and monitoring bioelectric im-
pedance cardiograph signals from human subjects, was used to obtain
PEP. Four spot electrodes, one applied to the front of the neck above
the collar bone, one applied to the nape of the neck and the other
two applied to the xiphysternal junction and the lower back, were
used to obtain the ZCG signal (i.e., the first derivative of the imped-
ance waveform). An external ECG cable using a modified Lead II con-
figuration was used to obtain ECG waves. Signals were ensemble
293
averaged over all cardiac cycles within 1 min using the CopWin soft-
ware (Bio-Impedance Technology, Inc., Chapel Hill, NC). PEP was
identified as the time elapsed between the Q point on the ECG wave
and the B inflection on the ZCG wave. The Q and B points were iden-
tified manually using the criteria described by Sherwood et al. (1990).
Internal consistency was high with α = .99 for baseline and .95 for
task.
2.4. Anger provocation
Role play was chosen to induce anger for two reasons. First, in ev-
eryday life anger is often provoked in interpersonal interactions and
role play attempts to simulate such interactions, hence improving
lab to field generalization. Second, results of a meta-analysis suggest
that interpersonal stressors tend to elicit more pronounced cardio-
vascular reactivity than traditional laboratory stressors such as cold
pressor and public speech (Suls and Wan, 2007). The procedure
used in this study was similar to that used by Suchday and Larkin
(2001).
In the present study participants were presented with the follow-
ing scenario and asked to imagine they were engaged in the specific
situation. They were then told to try to convince the confederate
(presented as another participant) to change her behavior.
For one of your modules you are working with a partner on a project
that counts 50% of your module grade. You have been doing your
part of the project but your partner has done very little. The deadline
is two days away and you are worried about how the project will
turn out and the grade you will get for the project.Goal: convince
your partner to do her portion of the project so that you can both
get a good grade.
The confederate delivered standard responses to the participant's
requests so as to provoke participants in the interaction. Among
suggested responses were “Don't bug me about that and I'll get to it
sooner” and “You complain too much. It's so difficult to work with
you”. The anger provocation scenario was tested in a small pilot
study and found to be effective in arousing anger.
2.5. Emotion regulation instructions
After reading the scenario description participants were given in-
structions on anger regulation. In the expression condition, participants
received the instruction to “please freely express your feelings. Let your
partner know how you feel”. In the suppression condition participants
were told to “hide your emotions. Do not let your partner know what
your feeling is”. In the reappraisal condition participants received the
instruction to “think about the situation more positively and from
your partner's point of view”.
2.6. Procedure
This research was approved by the National University of Singapore
Institutional Review Board. Participants came for the experiment indi-
vidually. All participants were told to refrain from alcoholic and caffein-
ated beverages as well as heavy exercise at least 2 h prior to the study.
On arrival at the lab, each participant was told that she would be engag-
ing in a negotiation with another participant. Then the participant was
introduced to a female confederate of the same race who was described
as “another participant”. The confederate then left the room. After the
participant signed a consent form, she was attached with the equipment.
The demographic information and the first emotion ratings were
obtained prior to the first 10 min resting baseline. The role play inter-
action began after the baseline. In the role play interaction, the partic-
ipant was presented with the scenario given above followed by the
emotion regulation instructions. During the role play the participant
294 T. Zhou, G.D. Bishop / International Journal of Psychophysiology 86 (2012) 291–298
and confederate were seated in separate rooms with the confederate
able to observe the participant through a one-way mirror. The one
way mirror was visible to the participant who was aware that she
could be observed by the confederate. Communication was done
through an intercom. This procedure was used so as to avoid the po-
tential confounding effects generated by the confederate's possibly
different nonverbal cues toward different participants. The role play
was 3 min in length. After completion of the role play, the participant
completed the post-task emotion rating scale which was followed by
a second 10 min baseline period. Following this, all recording appara-
tuses were detached. Finally, the participant was asked to complete
the anger-in subscale of STAXI-II, the ERQ and PVQ. These question-
naires were obtained at the end of the experiment for convenience
and also to avoid sensitizing participants to the study's focus on
anger. After the completion of these questionnaires, each participant
was thoroughly debriefed and thanked.
3. Results
3.1. Cultural representativeness of groups
To assess the cultural representativeness of the ethnic groups, cul-
tural values were compared between two groups using the Schwartz
et al. (2001) portrait values questionnaire. Centered scores were used
for each dimension with lower score indicating greater endorsement.
Results indicated that Chinese (M = 0.55) endorsed the value of tradi-
tion more than the Caucasian group (M = 1.10), t (86)= −3.03, p =
.003, d = .62. However, Caucasians (M = −0.88) endorsed the values
of self direction more than the Chinese group (M = −0.22), t (87) =
4.26, p = .001, d = .83. There was also a significant difference between
Chinese and Caucasians in the endorsement of security with Chinese
(M = −0.04) endorsing security more than Caucasians (M= 0.39),
t (87) = −2.99, p = .004, d = .54. These findings are consistent with
considerations of collectivism–individualism as well as results of a pre-
vious cross-cultural values survey with more than 35,000 respondents
from 49 countries (Schwartz, 1999). These findings confirm the repre-
sentativeness of cultural groups in the present study.
3.2. Baseline comparison
Baseline comparisons on emotional experience and cardiovascular
parameters between two cultural groups were done to assess the
comparability of the groups.
3.2.1. Emotional experience at baseline
Three 2 (cultural background) ∗ 3 (regulation condition) ANOVAs
with the three emotion rating components at baseline as DVs were
performed to examine the baseline differences between the two cul-
tural groups and across different conditions. No significant ANOVA
main effects or interactions were obtained for anger or negative emo-
tions, all ps > .10.
3.2.2. Cardiovascular parameters at baseline
In order to obtain a true resting baseline, only cardiovascular read-
ings taken during the second half of the first baseline period were in-
cluded in the analyses. For each index measures taken during the last
half of the baseline period were averaged to obtain a mean value. 2
(cultural background) ∗ 3 (regulation condition) ANOVAs done with
baseline values of CO, PEP, TPR, HR, SBP and DBP as DVs showed no
significant main effects or interaction, all ps > .07. Baseline values for
all parameters are given in Table 1.
3.3. Anger provocation check
As a manipulation check a 2 (cultural background) ∗ 3 (regulation
condition)∗2 (period: baseline vs. role play) repeated measures ANOVA
was done to examine the effect of the role play on feelings of anger. Suc-
cess of the role play in anger provocation was shown with the signifi-
cant main effect for period, F (1, 84) = 186.46, p b .001, η 2 = .69.
Participants experienced greater intensity of anger following the role
play (M = 3.90) than they did at the baseline (M =1.47). In addition,
there was a significant decrease in positive emotion as well as a signif-
icant increase in negative emotion from baseline to the role play: posi-
tive emotion decreased from M = 3.82 to M = 2.27, F (1, 84) = 109.69,
p b .001, η 2 = .57; negative emotion increased from M = 1.84 to M=
2.50, F (1, 84) =34.98, p b .001, η2 = .30. None of the interactions from
these analyses reach statistical significance (all ps > .13).
3.4. Hypothesis testing
The first hypothesis predicted that Chinese participants would use
suppression on a habitual basis more often to regulate anger than
would Caucasians. This hypothesis was tested by comparing Chinese
and Caucasian participants on the modified ERQ suppression scale.
The results indicated that there was no difference in habitual usage
of anger suppression between Chinese (M = 4.10) and Caucasians
(M = 3.83), t (87) 1 = 0.92, p = .36, d = 0.20, although the means
were in the expected direction.
Differences were obtained, however, when comparing the habitual
usage of reappraisal. Chinese participants (M=5.17) scored significantly
higher than Caucasians (M=4.72), t (87)=2.13, p= .036, d=0.44, on
the modified reappraisal scale of the ERQ. Thus, the second hypothesis
was supported.
The third hypothesis predicted that culture would moderate the re-
lationship between anger regulatory strategies and CVR responses with
Caucasians showing highest CVR in the suppression condition followed
by the expression and reappraisal conditions but Chinese showing the
highest CVR in the expression condition followed by suppression and
reappraisal. This hypothesis was tested by first computing reactivity
as the mean task value minus baseline and then using 2 (cultural
background) ∗ 3 (regulation condition) ANOVAs with reactivity for all
cardiovascular parameters as dependent variables. These analyses re-
vealed that the interactions for SBP, F (2, 84) = 4.66, p = .012, η2 =
.10, HR, F (2, 84) = 4.95, p = .009, η2 = .11, CO, F (2, 84) = 3.40, p =
.038, η2 =.08, and PEP, F (2, 82) = 9.05, p b .001, η 2 = .18, were statisti-
cally significant. The interaction for DBP was marginally significant, F (2,
84) = 2.98, p = .056, η 2 = .066, but the interaction for TPR was
non-significant, F (2, 84)= 0.52, p = .60, η 2 = .012. Significant results
were then following up with simple main effect analyses comparing
emotion regulation strategies within ethnic group and, where appropri-
ate, paired comparisons using Tukey's HSD. Means for all CVR measures
by cultural group and regulation condition are given in Table 2.
Analysis of SBP reactivity revealed that for Chinese, F (2, 42) =
4.23, p = .021, η 2 = .17, those assigned to express anger had higher
SBP reactivity (M = 30.57) than those assigned to do reappraisal
(M = 18.76), p = .019, whereas those assigned to suppress anger
(M = 22.14), did not differ from the other two conditions, ps > .12.
There were no significant differences in SBP reactivity among regula-
tion conditions among Caucasians, F (2, 42) = 1.12, p = .34, η 2 = .051,
(Fig. 1a). With respect to DBP reactivity, a similar pattern of means
emerged although the omnibus F-ratio was not significant.
For HR reactivity, analyses for simple main effects revealed that
within the Chinese group, F (2, 42) = 2.51, p = .094, η2 = .11, there
was a marginally significant difference in HR reactivity between partic-
ipants in the expression condition (M = 13.40) and those in the sup-
pression condition (M= 6.36), p = .089, with neither of these means
different than the reappraisal condition (M = 8.28), ps > .26. In the
Caucasian group, F (2, 42) = 2.80, p = .072, η 2 = .12, participants
assigned to the suppression condition (M = 17.97) had marginally
1
Due to missing data degrees of freedom are sometimes less than expected from a
full dataset. No participants were deliberately excluded from analyses.
 T. Zhou, G.D. Bishop / International Journal of Psychophysiology 86 (2012) 291–298 295

Table 1
Baseline values for physiological measures by cultural background and regulation condition.

Chinese Caucasians

Expression Suppression Reappraisal Expression Suppression Reappraisal

(n = 15) (n = 15) (n = 15) (n = 15) (n = 15) (n = 15)

SBP 115.18 ± 9.34 113.10 ± 17.80 115.05 ± 9.92 114.84 ± 11.30 120.63 ± 10.89 117.70 ± 8.54
DBP 70.05 ± 8.11 67.99 ± 6.89 68.62 ± 4.58 67.73 ± 7.29 71.97 ± 5.25 71.64 ± 5.54
HR 82.95 ± 16.63 79.20 ± 10.71 78.29 ± 12.83 71.20 ± 13.63 80.00 ± 10.47 77.48 ± 6.74
CO 5.53 ± 1.37 5.16 ± 0.81 5.06 ± 1.14 5.02 ± 1.28 5.86 ± 1.24 5.50 ± 1.02
TPR 1029.40 ± 222.47 1043.79 ± 164.71 1132.39 ± 400.10 1107.60 ± 209.88 970.80 ± 142.17 1017.23 ± 159.16
PEP 111.18 ± 16.70 109.64 ± 10.11 108.27 ± 19.37a 115.37 ± 12.93 112.45 ± 15.37 114.60 ± 12.90a

Notes. Numbers are means ± standard deviation. SBP = systolic blood pressure, DBP = diastolic blood pressure HR = heart rate, CO = cardiac output, PED = pre-ejection period,
TPR = total peripheral resistance.
a
N = 14 due to missing data.

greater HR reactivity compared with participants assigned to the ex-
pression condition (M = 8.90), p = .068, with neither condition differ-
ent than the reappraisal condition (M = 11.43) (Fig. 1b).
For CO simple main effects analyses for the Chinese group indicated
no significant differences in CO reactivity across different conditions,
F (2, 42) =0.55, p = .58, η 2 = .025. Significant differences, F (2, 42) =
3.34, p = .045, η 2 = .14, were, however, obtained within the Caucasian
group. Participants assigned to the expression condition (M = 0.70)
had lower CO reactivity than those assigned to the suppression condi-
tion (M= 1.89), p = .037, whereas CO reactivity in the reappraisal con-
dition (M = 1.15) did not differ from either of the other two conditions,
ps > .25. This is shown in Fig. 1c.
PEP reactivity differed by regulation condition for both Chinese and
Caucasians. Within the Chinese group, F (2, 41) = 5.99, p = .005,
η2 =.23, participants assigned to the expression condition (M=18.71)
had greater PEP reactivity than those in the suppression group
(M = 11.17), p = .024, as well as those in the reappraisal group
(M = 9.82), p = .008. Differences between reappraisal and suppression
were not significant, p > .87. In the Caucasian group, F (2, 41)= 5.60,
p = .007, η 2 = .22, participants assigned to the suppression condition
(M=25.26) had greater PEP reactivity than those assigned to the ex-
pression condition (M=12.27), p= .007 and marginally greater reactiv-
ity than those in the reappraisal condition (M=15.77), p= .075, with
differences between expression and reappraisal being non-significant,
p >.66 (Fig. 1d).
Taken together, the interaction of cultural background and regula-
tion condition was significant in the reactivity of SBP, HR, CO and PEP
and marginally significant for DBP reactivity. Generally, Chinese
showed the highest reactivity when asked to express anger whereas
Caucasians exhibited the highest reactivity when asked to suppress
anger. Reappraisal generally did not differ from suppression for the
Chinese or from expression for the Caucasians.
To determine whether this cardiovascular reactivity was paralleled
by changes in emotional experience we computed change in emotions
by subtracting the baseline value for each emotion index from the
task value and then performed three 2 (cultural background) by 3
(regulation condition) ANOVAs with changes in anger, positive emotion
and negative emotion as the DVs. This analysis found that no significant
effect for cultural background, all p > .31, regulation condition, all
ps >.50, or their interaction, all ps >.13, indicating that neither cultural
background nor emotion regulation instruction had reliable effects on
emotional experience.
4. Discussion
To briefly summarize, the results of this study indicated that:
(1) Chinese reported using reappraisal more frequently in anger situa-
tions than did Caucasians although no difference was obtained for
suppression and (2) cultural background moderated the effects of reg-
ulation strategy on CVR following anger provocation. As found in stud-
ies in North America, Caucasians showed stronger CVR when instructed
to suppress their emotions. However, Chinese participants showed
stronger CVR when instructed to express their anger.
The results showing greater use of reappraisal by Chinese as com-
pared with Caucasians are in line with similar findings by Matsumoto
et al. (2008) and consistent with the “other orientation” among
Chinese. As a very important feature of social relationships in Chinese
society, other orientation suggests that Chinese people are particularly
attuned to others on both psychological and behavioral levels (Yang,
1995). Other orientation is manifested as concern about being different
from others, tendency to conform to others and attempting to create a
better impression on others through self-monitoring. One way to
accomplish this is to think of situations from another's point of view.

Table 2
Means and SDs of CVR by cultural background and regulation condition.

Chinese Caucasians

Expression Suppression Reappraisal Expression Suppression Reappraisal

(n = 15) (n = 15) (n = 15) (n = 15) (n = 15) (n = 15)

ΔSBP 30.57 ± 9.06⁎⁎ 22.14 ± 12.03⁎⁎ 18.76 ± 12.92⁎⁎ 20.40 ± 11.12⁎⁎ 26.25 ± 13.96⁎⁎ 25.41 ± 9.03⁎⁎
ΔDBP 18.41 ± 5.68⁎⁎ 13.71 ± 9.63⁎⁎ 10.91 ± 6.00⁎⁎ 13.81 ± 6.43⁎⁎ 13.97 ± 5.28⁎⁎ 14.37 ± 4.30⁎⁎
ΔHR 13.40 ± 9.26⁎⁎ 6.36 ± 9.25⁎ 8.28 ± 8.14⁎⁎ 8.90 ± 8.17⁎⁎ 17.97 ± 13.81⁎⁎ 11.43 ± 9.72⁎⁎
ΔCO 1.03 ± 0.80⁎⁎ 0.79 ± 0.78⁎⁎ 0.75 ± 0.80⁎⁎ 0.70 ± 1.12⁎ 1.89 ± 1.65⁎⁎ 1.15 ± 0.97⁎⁎
ΔTPR 155.33 ± 350.97 57.03 ± 197.06 41.13 ± 158.43 127.83 ± 450.10 −69.15 ± 206.32 74.13 ± 360.14
ΔPEP 18.71 ± 8.09⁎⁎ 11.17 ± 7.06⁎⁎ 9.82 ± 7.35⁎⁎,a 12.27 ± 10.35⁎⁎ 25.26 ± 13.48⁎⁎ 15.77 ± 8.34⁎⁎,a

Notes. Numbers are mean changes from baseline to task ± standard deviation. ΔSBP = change in systolic blood pressure, ΔDBP = change in diastolic blood pressure, Δ HR = change
in heart rate, ΔCO = change in cardiac output, ΔPEP = change in reversed PEP, ΔTPR = tps10552change in total peripheral resistance. All means were tested against zero to de-
termine significant increases or decreases from baseline.
⁎ p b .05.
⁎⁎ p b .01.
a
N = 14 due to missing data.
296 T. Zhou, G.D. Bishop / International Journal of Psychophysiology 86 (2012) 291–298
Fig. 1. Means (SEM) by ethnic group and condition for changes in (a) systolic blood pressure, (b) heart rate, (c) cardiac output, and (d) pre-ejection period.
Although previous studies have indicated Asians used emotion sup-
pression more than Caucasians (Gross and John, 2003; Matsumoto et al.,
2008; Suchday and Larking, 2004), the difference between Chinese and
Caucasians on suppression was not statistically significant in this study
although the means were in the expected direction. The reason for this
discrepancy in results is not immediately clear although with a larger
sample size the differences in suppression may have been significant.
Cultural differences in habitual usage of expression and suppression in
anger situations needs further investigation.
Across conditions all participants showed a pattern of increased BP,
HR, CO and reversed PEP but unchanged TPR in response to anger prov-
ocation, which is in line with results found in previous research on
emotion-specific physiological responses. The anger response has
been characterized by alpha- and beta-adrenergically mediated cardio-
vascular effects (Kreibig, 2010). A meta-analysis by Stemmler (2004)
found that in the comparison of autonomic effects between anger and
neutral conditions, large effect sizes were noted for SBP, DBP and HR,
small effect size was found for CO, and the effect of TPR was only mar-
ginally significant.
It can also be noted that the size of the PEP reactivity obtained in this
study is somewhat larger than that obtained by Mauss et al. (2007) and
Ray et al. (2008) who obtained PEP reactivity in the range of 5 ms
whereas the reactivity obtained in the present study went as high as
25 ms. One possible reason for the difference in PEP reactivity between
those two studies and the present one may be the difference in anger in-
duction. The present study used a role play methodology whereas Ray
et al. asked participants to remember an anger provoking incident and
Mauss et al. used unfriendly feedback from an experimenter. The
anger generated by the role play task may have been greater than in
the other two studies thus accounting for the reactivity differences. It
should also be noted that PEP reactivity of the size found in the present
study has been obtained by Mendes et al. (2003) and Tomaka et al.
(1993) indicating that the PEP reactivity obtained in the present study
is not without precedent.
The key finding from this study is the reversal between Caucasians
and Chinese with respect to cardiovascular responses to suppression
and expression. A clear pattern was obtained in which CVR was larger
for Caucasians when they were instructed to suppress their anger
whereas CVR was stronger for Chinese when instructed to express
their feelings. These findings are in line with the differential engage-
ment expected for Chinese and Caucasians when utilizing emotion reg-
ulation strategies that are either in line with or contrary to cultural
expectations. Since expression of negative emotion is contrary to
Chinese cultural norms which favor suppression whereas the opposite
is true for Caucasians the differences obtained in CVR patterns are
fully in line with the task engagement model of CVR proposed by
Obrist (1981) and elaborated by Wright and Kirby (2001).
In relating these results to those obtained by Gross and his col-
leagues (Gross, 1998; Gross and Levenson, 1993, 1997) it is important
to note that the results obtained for Caucasians are fully in line with
findings showing the negative physiological implications of emotional
suppression. The findings for Chinese, however, are directly contrary
to these findings and inconsistent with findings generally showing
no cultural differences in cardiovascular responses to suppression
(Roberts et al., 2008). However, the results of a recent study comparing
the responses of Asian Americans and Caucasian Americans to a
distressing film suggested that emotion suppression was positively
related to blood pressure as indicated by mean arterial pressure
(MAP) for Caucasian Americans but the reverse pattern was found for
Asian Americans (Butler et al., 2009), a pattern consistent with our find-
ings. One possible reason for the inconsistency in previous studies
examining cultural moderation of responses to different emotion
 T. Zhou, G.D. Bishop / International Journal of Psychophysiology 86 (2012) 291–298
regulation strategies is that most of the studies to date have been done
in the US in which the prevailing norms tend to favor expression of
emotion whereas our findings were obtained in an Asian society
where public expression of negative emotion is contrary to societal
norms. Asian Americans participating in studies done in the US may
have acculturated to Western culture and norms and no longer be
fully representative of their culture of origin. Along these lines it is im-
portant to note that in the present study the cultural representativeness
of the samples was examined by measuring cultural values verifying
that the ethnic groups were culturally representative. The importance
of cultural values in determining the effects of different emotion regula-
tion strategies is demonstrated by Butler et al. (2007) who found that
the negative effects of the habitual use of emotional suppression on
partner responsiveness and negative perceptions during a dyadic task
were attenuated for participants showing greater endorsement of
Asian values.
Although we had expected that reappraisal would lead to the lowest
CVR for both Caucasians and Chinese, this was not the case. CVR in the
reappraisal condition was not different from the suppression condition
for Chinese and not different from expression for Caucasians, although
it engendered less CVR than expression for Chinese and less CVR than
suppression for Caucasians. Since suppression of feelings is counter-
normative for Caucasians whereas expression is counter-normative
for Chinese it may be that reappraisal evokes cardiovascular responses
similar to the normative emotion regulation strategy but different
from the strategy that is counter-normative. Reappraisal is strongly en-
dorsed as an emotion regulation strategy in both American and Asian
societies as indicated in results by Matsumoto et al. (2008) and indicat-
ed in the relatively high means for reappraisal for both Caucasians and
Chinese in the present study. Chinese participants did score higher on
reappraisal than Caucasians but both groups had scores well above
the midpoint of the 7 point scale.
The findings of this research have both theoretical and practical im-
plications. From a theoretical perspective these results emphasize the
importance of considering culture when making predictions about the
benefits and liabilities of different emotion regulation strategies. The
norms favoring emotional expression in American society and favoring
emotional suppression in Asian societies lead to different consequences
when these strategies are used and particularly when the counter-
normative strategy is used. As such it is definitely not the case that sup-
pression generally leads to negative physiological consequences as has
been suggested in previous studies. Rather, the physiological conse-
quences of a particular emotion regulation strategy depend on the cul-
ture of the individual involved and whether that strategy is one that can
be considered normative or counter-normative in that culture.
This has practical implications in that advising people to engage in
the counter-normative strategy is likely to incur costs for the person.
As such, although it may be appropriate to advise expression rather
than suppression of feelings to members of societies where expression
of emotion is valued and generally normative, such advice would not
be appropriate in societies emphasizing public suppression of feelings.
In particular, it would not be appropriate to advise members of Asian
societies, where public suppression of emotion is normative, to express
their feelings openly as such expression is likely to result in greater car-
diovascular reactivity which has the potential to be health damaging.
It is also important to acknowledge the limitations in this study.
First, the anger provocation process was a brief, artificial scenario,
which is different from anger situations in daily life and can be
expected to engender only a small to moderate amount of anger,
something noted by some participants at the end of the experiment.
However, it can also be noted that participants expressed significantly
increased anger after the role play than before which indicates that
the role play was successful in provoking anger in participants. Sec-
ond, being instructed to use a particular emotion regulation strategy
is somewhat unusual for participants and the artificiality of this situ-
ation may limit generalizability. This is particularly true in light of the
297
lack of a no instruction condition which should be included in future
studies. Third, this study used only young female participants. The
purpose of including only females and having them interact with a
confederate of the same sex and race was to eliminate potential con-
founds due to gender and race. However, the limitation of our sample
to young females does potentially limit the generalizability of the
findings. These findings should be replicated with males and also
with a wider age range of participants.
In conclusion, the findings reported here demonstrate that the im-
pact of different emotion regulation strategies on cardiovascular param-
eters differs as a function of culture with Caucasians showing the
strongest response to anger suppression whereas Chinese showed the
strongest response to anger expression. For both groups responses to
anger followed a challenge pattern with the pattern being stronger for
Caucasians when instructed to engage in emotional suppression but
marginally stronger for Chinese when instructed to express their
feelings.
References
Butler, E.A., Lee, T.L., Gross, J.J., 2007. Emotion regulation and culture: are the social
consequences of emotion suppression culture-specific? Emotion 7, 30–48.
Butler, E.A., Lee, T.L., Gross, J.J., 2009. Does expressing your emotions raise or lower
your blood pressure? The answer depends on cultural context. Journal of Cross-
Cultural Psychology 40, 510–517.
Ekman, P., 1992. Are there basic emotions? Psychological Review 99, 550–553.
Ekman, P., 1993. Facial expression and emotion. American Psychologist 48, 384–392.
Ekman, P., 1994. Strong evidence for universals in facial expressions: a reply to
Russell's mistaken critique. Psychological Bulletin 115, 268–287.
Finopress Medical Systems, 2005. Finger Pressure Reference Guide: Guide for the Non-
invasive Continuous Finger Blood Pressure Measurement and Recording Systems:
Portapres and Finometer. Finopress Medical Systems, Amsterdam.
Friesen, W.V., 1972. Cultural Differences in Facial Expressions in a Social Situation: An
Experimental Test of the Concept of Display Rules. University of California, San
Francisco, pp. 1–100.
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., 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. Emotion suppression: physiology, self-report and
expressive behavior. Journal of Personality and Social Psychology 64, 970–986.
Gross, J.J., Levenson, R.W., 1997. Hiding feelings: the acute efects of inhibiting negative
and positive emotion. Journal of Abnormal Psychology 106, 95–103.
Jansen, J.R., Schreuder, J.J., Mulier, J.P., Smith, N.T., Settels, J.J., Wesseling, K.H., 2001. A
comparison of cardiac output derived from the arterial pressure wave against
thermodilution in cardiac surgery patients. British Journal of Anaesthesia 87,
212–222.
John, O.P., Gross, J.J., 2004. Healthy and unhealthy emotion regulation: personality
processes, individual differences, and life span development. Journal of Personality
72, 1301–1333.
Kamarck, T.W., Everson, S.A., Kaplan, G.A., Manuck, S.B., Jennings, J.R., Salonen, R.,
Salonen, J.T., Jukka, T., 1997. Exaggerated blood pressure responses during mental
stress are associated with enhanced carotid atherosclerosis in middle-aged Finnish
man: findings from the Kuopio Ischemic Heart Disease Study. Circulation 96,
3842–3848.
Kawachi, I., Sparrow, D., Spiro, A., Vokonas, P., Weiss, S.T., 1996. A prospective study of
anger and coronary heart disease: The Normative Aging Study. Circulation 94,
2090–2095.
Kitayama, S., Markus, H.R., 1994. Emotion and Culture: Empirical Studies of Mutual
Influence. American Psychological Association, Washington, DC.
Kitayama, S., Mesquita, B., Karasawa, M., 2006. Cultural affordances and emotional
experience: socially engaging and disengaging emotions in Japan and the United
States. Journal of Personality and Social Psychology 91, 890–903.
Kreibig, S.D., 2010. Autonomic nervous system activity in emotion: a review. Biological
Psychology 84, 394–421.
Matsumoto, D., Yoo, H.S., Nakagawa, S., Members of the Multinational Study of Cultural
Display Rules, 2008. Culture, emotion regulation, and adjustment. Journal of
Personality and Social Psychology 94, 925–937.
Mauss, I.B., Cook, C.L., Cheng, J.Y.J., Gross, J.J., 2007. Individual differences in cognitive
reappraisal: experiential and physiological responses to an anger provocation.
International Journal of Psychophysiology 66, 116–124.
Mendes, W.B., Reis, H.T., Seery, M.D., Blascovich, J., 2003. Cardiovascular correlates of
emotional expression and suppression: do content and gender context matter?
Journal of Personality and Social Psychology 84, 771–792.
Obrist, P.A., 1981. Cardiovascular Psychophysiology: A Perspective. Plenum, New York.
Ray, R.D., Wilhelm, F.H., Gross, J.J., 2008. All in the mind's eye? Anger rumination and
reappraisal. Journal of Personality and Social Psychology 91, 133–145.
298 T. Zhou, G.D. Bishop / International Journal of Psychophysiology 86 (2012) 291–298
Roberts, N.A., Levenson, R.W., Gross, J.J., 2008. Cardiovascular costs of emotion suppression
cross ethnic lines. International Journal of Psychophysiology 70, 82–87.
Scherer, K.R., 1997. The role of culture in emotion-antecedent appraisal. Journal of
Personality and Social Psychology 73, 902–922.
Scherer, K.R., Wallbott, H.G., 1994. Evidence for universality and cultural variation of
differential emotion response patterning. Journal of Personality and Social Psychology
66, 310–328.
Scherer, K.R., Wallbott, H.G., Matsumoto, D., Kudoh, T., 1988. Emotional Experience in
Cultural Context: A Comparison Between Europe, Japan, and the United States.
In: Scherer, K.R. (Ed.), Facets of emotion: Recent research. Erlbaum, Hillsdale, NJ,
pp. 5–30.
Schwartz, S.H., 1999. A theory of cultural values and some implications for work.
Applied Psychology: An International Review 48, 23–47.
Schwartz, S.H., Melech, G., Lehmann, A., Burgess, S., Harris, M., Owens, V., 2001. Extending
the cross-cultural validity of the theory of basic human values with a different meth-
od of measurement. Journal of Cross-Cultural Psychology 32, 519–542.
Sherwood, A., Allen, M.T., Fahrenberg, J., Kelsey, R.M., Lovallo, W.R., van Doornen, L.J.P., 1990.
Methodological guidelines for impedance cardiography. Psychophysiology 27, 1–23.
Siegman, A.W., Smith, T.W., 1994. Anger, Hostility, and the Heart. Lawrence Erlbaum,
Hillsdale, NJ.
Stemmler, G., 2004. Physiolgical Processes During Emotion. In: Phillipot, R.S., Feldman,
R.S. (Eds.), The Regulation of Emotion. Lawrence Erlbaum Associates, Mahwah, NJ,
pp. 33–70.
Suchday, S., Larkin, K.T., 2001. Biobehavioral responses to interpersonal conflict during
anger expression among anger-in and anger-out men. Annals of Behavioral Medicine
23, 282–290.
Suchday, S., Larking, K.T., 2004. Psychophysiological responses to anger provocation
among Asian Indian and White men. International Journal of Behavioral Medicine
11, 71–80.
Suls, J., Wan, C.K., 2007. The relationship between trait hostility and cardiovascular
reactivity: a quantitative review and analysis. Psychophysiology 30, 615–626.
Tavris, C., 1989. Anger: The Misunderstood Emotion. Touchstone Books/Simon &
Schuster, Inc., New York, NY, US.
Tomaka, J., Blascovich, J., Kelsey, R.M., Leitten, C.L., 1993. Subjective, physiological, and
behavioral effects of threat and challenge appraisal. Journal of Personality and
Social Psychology 65, 248–260.
Williams, J.E., Paton, C.C., Siegler, I.C., Eigenbrodt, M.I., Nieto, F.J., Tyroler, H.A., 2000.
Anger proneness predicts coronary heart disease risk: prospective analysis from the
Atherosclerosis Risk in Communities (ARIC) Study. Circulation 101, 2034–2039.
Wright, R.A., Kirby, L.D., 2001. Effort Determination of Cardiovascular Response: An
Integrative Analysis with Applications in Social Psychology. In: Zanna, M.P.
(Ed.), Advances in experimental social psychology. Academic Press, San Diego, CA,
pp. 255–307.
Yang, K.S., 1995. Chinese Social Orientation: an Integrated Analysis. In: Lin, T.Y., Tseng,
W.S., Yeh, E.K. (Eds.), Chinese societies and mental health. Oxford University Press,
New York, pp. 19–39.