Individual and ethnic differences in habitual usage of reappraisal and suppression in relationship with their associated psychological and physiological consequences 3

In line with this, the present project attempted to investigate the mechanism and effect of two emotion regulation strategies – reappraisal and suppression, on subjective well-being SWB

CHAPTER I INTRODUCTION Successful regulation of emotion and appropriate expression of emotion are vital for everyday social life (Gross & John, 2003) Publications discussing this important ability such as the mechanisms underlying emotion regulation, as well as the effects of different emotion

regulation strategies have increased exponentially over the past years (Joormann & D’Avanzato, 2010) With the large number of studies on the effect of emotion regulation on health outcome, however, very few studies with Southeast Asian populations have been reported In line with this, the present project attempted to investigate the mechanism and effect of two emotion regulation strategies – reappraisal and suppression, on subjective well-being (SWB) and cardiovascular reactivity (CVR) among Singaporeans, focusing on two ethnic groups: Chinese and Indians

Process Model of Emotion Regulation

It has been widely acknowledged that people do not just passively experience their emotions but rather are actively involved in all stages of emotion generation and have the ability

to modify them (Eid & Diener, 2001) Emotions are believed to have evolved to serve survival functions, which can be seen in many species (Levenson, 1999) As conscious social creatures, human beings constantly engage in the process of modulating emotions to fit societal rules (Manstead & Fischer, 2000) As important and interesting the topic is emotion regulation is a very difficult topic to study empirically because emotion generation and emotion regulation processes are intertwined with each other and difficult if not impossible to tease apart in real experiences (Joormann & D’Avanzato, 2010) Researchers can still define the two processes theoretically Emotion can be seen as a response to an external or internal situation, or in other words, a person—situation transaction (Lazarus, 1991) On the other hand, emotion regulation is often defined as the “processes by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions” (Gross, 1998, p 275)

Emotional experience is dynamic, and constantly changes in response to the changing situations and the coping attempts of the person (Lazarus, 1991) Appraisal theories emphasize that it is the interpretation of the situation, but not the situation itself, that determines the nature of the emotional responses In many studies researchers have documented that people exposed to the same event displayed a variety of emotional responses depending on their different appraisals of the situation For example, Siemer, Mauss, and Gross (2007) reported that participants responded

to a standardized laboratory task with a variety of different emotions, and different aspects of appraisal could predict the intensity of individual emotions across participants These theories and evidence all support the position that emotion generation (response to a situation) and emotion regulation (modulating the emotional responses) processes are recursive in nature, and in theory people can intervene to modify their emotions at any point along the chain of emotional response (Joormann & D’Avanzato, 2010) Gross (1998) has proposed the process model, which depicts emotion regulation as interventions which may take place at numerous points in the emotion generation sequence The advantage of this model is that it takes into account the aspects of both emotion generation and regulation, thereby providing a platform to empirical studies on emotion experience

The process model of emotion regulation (Figure 1) differentiates two broad categories

of emotion regulation strategies: antecedent-focused and response-focused strategies (Gross, Richards, & John, 2006) Antecedent-focused strategies focus on the modulation of future

emotional responses, which refer to things people do before the emotion tendencies become fully activated For example, situation selection and situation modification are antecedent-focused emotion regulation strategies that affect the emotion-eliciting situation directly, and thereby alter the entire emotion trajectory By comparison, the purpose of response-focused strategies is to manage the existing emotions, which refer to things people do after the emotion is fully activated and the response tendencies have been generated Expression suppression is one example of the

response-focused strategies For example, one might bite one’s lip to appear unfazed by an unfair comment from a manager despite underlying feelings of anger

Figure 1 Process Model of Emotion Regulation (from Gross & John, 2003)

Two Specific Strategies: Cognitive Reappraisal and Expression Suppression

As discussed in the above section, there are numerous entry points at which one can

actively intervene in the process of emotional experience It is not practical to study all of the

possible regulation strategies at once Rather, two strategies are of particular interest in the

current project: cognitive reappraisal, an antecedent-focused emotion regulation strategy, and

expressive suppression, a response-focused strategy Despite the fact that each strategy belongs to one broad category, three factors were considered when selecting which strategies to study: (1)

the two strategies are commonly used in everyday life and there are a number of available studies documenting the effects of the two strategies; (2) the two strategies are potentially influenced by culture; and (3) there are no studies reporting the comparison between Chinese and Indians in

their usage of these two strategies

Cognitive reappraisal refers to a type of early regulatory effort to modulate one’s potential emotional responses, including changing the subject one sees in the situation, or

changing the way one interprets the situation (Gross & John, 2003) It was first studied by

Lazarus and colleagues, who reported that instructing participants to view an upsetting surgical procedure in a more analytical and detached way could decrease their subjective and

physiological responses (Lazarus & Alfert, 1964)

Expressive suppression, on the other hand, refers to efforts that come in late in the emotional generation and regulation process In particular, it is a form of response modification that inhibits ongoing emotion expressive behaviors (Gross & John, 2003) This strategy has been studied in different theoretical frameworks, such as from the perspective of self-control in

Lazarus’ coping inventory (Lazarus, 1993) In addition, it has been repeatedly observed that suppression of emotion expressions is often associated with heightened physiological responses (e.g Gross & Levenson, 1993)

Gross and John (2003) conceptualized the individual differences in emotion regulation based on the process model, and developed the Emotion Regulation Questionnaire (ERQ), a ten-item scale, to measure one’s general usage of reappraisal and expression suppression in everyday life The ERQ has been tested in different populations and in different languages (e.g Balzarotte, John, & Gross, 2010; Zhu, Auerbach, Yao, Abela, Xiao, & Tong, 2008), and its reliability and validity have been established The current study focused on reappraisal and suppression and used the ERQ as one of the main measurements, and also extended the investigation on the mechanism and consequences of the two strategies

Two Specific Health Indicators: Subjective Well-being and Cardiovascular Reactivity

Much past work in health psychology has emphasized that negative emotions, such as anger, depression, and anxiety, are associated with higher morbidity and mortality from a range

of chronic illnesses, including cardiovascular disease (Barefoot, Brummett, Helms, Mark, Siegler,

& Williams, 2000) and diabetes (Lustman, Frank, & McGill, 1991) More recent evidence has suggested that negative emotions may also play an important role in the development of such diseases (e.g Carnethon, Kinder, Fair, Stafford, & Fortmann, 2003) Positive emotions, on the other hand, may play a protective role with respect to the development of hypertension, diabetes, and respiratory tract infections (Richman, Kubzansky, Maselko, Kawachi, Choo, & Bauer, 2005) Therefore, the success of emotion regulation, that is, successful enhancement of positive

emotions and decrease of negative ones, is closely related to one’s health condition in a positive manner

The mechanisms that have been proposed to link emotional experiences and health outcomes include direct pathways involving physiological activation, and indirect pathways that involve behavioral and cognitive elements, as well as coping resources For physiological

pathways, many studies have focused on the acute alterations of sympathetic nervous system (SNS) and/or parasympathetic nervous system (PNS) in response to an emotional episode In particular, the regulation of negative mood in relationship with cardiovascular health has attracted the interest of researchers for decades For example, it is believed that the inhibition of anger expression, as well as exaggerated anger expression, can both lead to an increased risk on the development and progression of cardiovascular diseases (Everson, Goldberg, Kaplan, Julkunen,

& Salonen, 1998) The link between the acute effect of regulation in isolated anger episodes and the long-term influence of emotion regulation on the development of cardiovascular disease (CVD) is specified in the reactivity hypothesis, which proposes that frequent experience of exaggerated cardiovascular reactivity (CVR) is related to enhanced atherosclerosis progression, and therefore facilitates CVD development (e.g Jennings, Kamarck, Everson-Rose, Kaplan, Manuck, & Salonen, 2004) For indirect pathways, much work has emphasized the cognitive elements such as attitudes and beliefs, as well as health behaviors In particular, the stress and

coping model attempts to account for both deleterious and protective health correlates of emotion regulation According to the stress and coping model (Lazarus & Folkman, 1984), cognitive appraisal, which refers to “the process of categorizing an encounter, and its various facets, with respect to its significance for well-being” (p 31), influences individual reaction to the

environment The cognitive appraisal process mediates one’s health beliefs, behaviors, as well as physiology, therefore has an important impact on one’s health outcome

In this research, subjective well-being (SWB) and CVR are selected as two indicators of the consequences of emotion regulation for the following reasons Firstly, SWB and CVR are widely used as health indicators in studying the relationship between psychosocial factors and health And they are closely related to the hypothesized mechanisms linking emotional

experiences and health, as discussed in the above paragraphs: CVR reflects the short-term

physiological changes in one emotional episode, and SWB indicates long-term affective

consequence Secondly, there is a rich literature on SWB and CVR studies, so the current

research can build on a solid theoretical framework Thirdly, SWB and CVR patterns are both potentially influenced by cultural background

SWB is an overall assessment on one’s self-reported well-being level and has been widely used in cross-cultural studies In short, SWB is the operationalized concept of happiness

by self-report The greatest challenge has been how to define happiness in a way that enables it to

be studied A very comprehensive review of the subjective well-being literature found that high levels of life satisfaction and positive affect, and low levels of negative affect are very commonly used to indicate one’s happiness level (Tov & Diener, 2007) Therefore these indicators are used

in the current research as surrogates of SWB Moreover, the CES-D (Center for Epidemiological Studies Depression Scale) was also used as an additional indicator for one’s SWB with

depression indicating lower SWB The CES-D is commonly used to measure depressive

symptomatology and the items focus on behavioral responses Firstly, depression itself is a

manifestation of impaired emotion regulation (Joormann & Gotlib, 2010) Secondly, items in the Life Satisfaction Scale are primarily focused on cognitive evaluation, whereas the Positive and Negative Affect Schedule (PANAS) focuses on affects that reflect personal feelings, it is believed that the CES-D provides necessary behavioral-oriented information on one’s self-reported well-being state In addition, many studies have included self-reported depression scales as one of the indicators of well-being (e.g Gross & John, 2003) Combining all these dimensions, an overall SWB index can be obtained, with higher scores indicating higher self-reported well-being level

In laboratory experiments, researchers commonly use CVR patterns as indicators of SNS or PNS activation, which are closely linked to cardiovascular health It is believed that individuals who exhibit exaggerated cardiovascular responses to simulated challenges have an increased risk for subsequent CVD, compared to their low-reactive counterparts (Lavallo, 2005) CVD is the world’s No 1 killer, claiming 17.1 million lives a year (WHO, 2011); therefore CVR pattern provides useful information as an indicator of health risk

Affective and Physiological Consequences of Reappraisal and Suppression

As social animals with societal rules, human beings constantly regulate experienced emotions through a number of effortful strategies, which is a type of adaptive behavior that is not seen in other species (Kalisch, 2009) However, emotion regulation is not cost-free, as energy and cognitive resources must be summoned in order to modulate our emotions (Muraven, Tice, & Baumeister, 1998) As proposed by the process model of emotion regulation, different regulatory strategies are associated with differentiated affective and physiological consequences

By definition, reappraisal occurs early in the emotion generative and regulative process Therefore, it is hypothesized that reappraisal, when used to down regulate negative emotions, will effectively lead to reduced negative experience and expressions, with relatively little

physiological cost (Mauss, Cook, Cheng, & Gross, 2007) By contrast, suppression starts late in

the chain of emotion generation and regulation process, and targets primarily the behavioral aspect of the emotional responses As such, suppression is hypothesized to require more effortful management, without effectively reducing negative emotional experience, and is associated with stronger physiological responses (Gross, Richards, & John, 2006) A series of survey and

experimental studies have tested these hypotheses in regard to the affective and physiological consequences of reappraisal and suppression The differentiated effects between reappraisal and suppression were generally supported in the literature

For example, it has been shown that people with high reappraisal scores in comparison

to those with low reappraisal scores, report greater positive and less negative emotional

experience, are more effective in interpersonal functions (Gross & John, 2003), and show a more adaptive profile of cardiovascular responding to a laboratory provocation procedure (Mauss et al, 2007) On the contrary, suppression has been linked to lower positive but higher negative

emotional experience, impaired interpersonal functions (Gross & John, 2003), as well as

magnified blood pressure responses (Roberts, Levenson, & Gross, 2008)

Moderation Effect of Culture between Emotion Regulation and Consequences

One obvious limitation of the previous research on emotion regulation is the failure to take into account the influence of culture on the consequences of emotion regulation Most studies were conducted in Western countries with predominantly North American participants Cultural psychologists have documented notable differences between Eastern and Western cultures As a result, the relationship between emotion regulation and health in the Southeast Asian context might be different from the existing reports that were obtained from participants in the Western countries In addition, Chinese and Indians may differ in important ways concerning emotion regulation In this section, the relationships between culture and emotion regulation and

its consequences will be discussed, and the specific design and implications relative to this research will be emphasized

Individual and Group Differences in Usage of Reappraisal and Suppression

Individuals differ in their use of various cognitive processes during any given event in order to regulate their feelings (John & Gross, 2004) For example, one individual may shift attention away from disturbing elements of a situation in order to down-regulate negative

emotions, while another individual may focus on those very elements and reinterpret their

meaning to feel better This illustrates that individuals may differ in their usage of cognitive reappraisal The existence of individual differences is actually the theoretical basis to further investigate the relationship between the usage of reappraisal and its consequences in different domains The mechanism underlying differences in the usage of reappraisal, however, has not been investigated systematically Research on reappraisal differences has mainly focused at the individual level (e.g see Mauss et al, 2007)

Nevertheless, differences in perception and cognition have been documented between traditional Eastern and Western ways of thinking There is evidence that differences exist

between different cultural groups in how people perceive and analyze specific situations For example, there are traditions of holistic versus analytic reasoning in East Asian and Western cultures respectively (Nisbett, 2003) Holistic thought involves a tendency to perceive the field as

a whole by explaining the events on the basis of the relationship between objects and the field By contrast, analytic thought involves the inclination to detach the objects from their context and focus on attributes and category of the objects, and then use rules about the categories to explain the behaviors of the objects (Norenzayan, Choi, & Peng, 2007) Although both systems are available to all people in all cultures, cultural context may encourage the usage of one system over the other, giving rise to systematic cultural differences

If Westerners and Easterners focus and analyze the event differently, it is very likely that they interpret and perhaps reinterpret the entire situation differently Therefore, it can be expected that the exact mode and frequency of reappraisal usage might be different in different groups, based on their habit of activating different cognitive systems of thinking Until now, however, there has been no report of cultural differences in reappraisal usage using the ERQ reappraisal scale (ERQ.R): Asian Americans reported statistically similar levels of reappraisal usage as European Americans, African Americans, and Latino Americans (Gross & John, 2003) One possible reason to explain the discrepancy between the expected difference and the non-significant results may be that the ERQ.R is a general assessment on individual’s usage of the particular strategy to do one type of particular regulation — down regulation of negative feelings

— and may not be sensitive enough to detect cultural differences Possible group differences may emerge when other aspects of seeing the situation are tested, such as whether one keeps positive feelings down Another reason may be that the different groups in the Gross & John (2003) study were assessed within the same cultural context, American culture, which may have reduced differences A recent study conducted at NUS has documented that Chinese Singaporeans

reported significantly different reappraisal usage than Caucasians to anger-provoking situations (Zhou, 2010) This research aims to take a closer look into the potential difference of reappraisal usage among Chinese and Indians

On the other hand, group level differences in emotion suppression have been reported repeatedly with respect to cultural display rules for emotion As early as the 1970’s, Friesen (1972) reported a study that became very famous in the field of cultural difference of emotional display rules, even though it was an unpublished dissertation In that study, American and

Japanese participants viewed a highly stressful film under two conditions Similar levels of facial signs of negative expression, such as disgust, fear, or distress, were observed for Americans and Japanese when they viewed the film alone However, in the presence of an experimenter,

Japanese used smiles to mask their negative feelings while Americans showed the same level of negative expressions as the alone condition Those findings can be accounted by a neurocultural theory of emotion, proposed by Ekman This theory proposed that there exists an innate facial affect program, which stores the prototypical emotional expressions explaining the universality of certain emotions, and cultural display rules, which explain cultural specificity in their expression (Ekman & Friesen, 1971) Since then, the proposition that culture influences the way people express their emotions has been largely accepted Gross and John (2003) has reported that

European Americans score lower on ERQ suppression scale (ERQ.S) than Asian Americans However, there has been no report on the comparison of emotion expression patterns between Chinese and Indians This research aims to fill this gap in the literature

Culture Shapes the Relationship between Emotion Regulation Strategies and Consequences

There is a lack of consensus on a definition of culture In the field of cultural

psychology, culture is generally referred to as a socially shared transmitted pattern of ideas (values, norms, and beliefs, etc.) that are instantiated in everyday practices (Tsai, Levenson, & McCoy, 2006) Many psychological processes which were considered as universal prove to be shaped by culture One such example is that individual emotional responses are influenced by cultural background A specific culture provides dominant themes for some emotions rather than others; for example, Western culture encourages disengaging emotions, whereas Eastern culture promotes engaging emotions (Kitayama, Mesquita, & Karasawa, 2006) In addition, culture modifies emotional responses by valuing or devaluing particular expression styles and regulatory strategies These are the social affordance and social reinforcement mechanisms, proposed in the

“dual process model” (Kitayama et al., 2006) Emotion regulation, especially reappraisal and suppression, is embedded in the meanings and practices that constitute the sociocultural

environment Therefore, culture directly influences an individual’s usage of emotion regulation,

as well as the consequences of those emotion regulation strategies

For example, as discussed in the study by Butler, Lee, and Gross (2009), cultural differences between West and East could influence the outcome of emotion suppression

European American traditions value and encourage self-expression Children are taught to speak

up, and emotional expression is not only condoned with in-group members, but also with group members By contrast, Asian traditions do not assume the expression of one’s personal feelings to be equally appropriate as do European Americans In line with this, Asian Americans indeed reported significantly higher scores than European Americans on the ERQ.S (Gross & John, 2003) Moreover, Asian Americans who suppressed their emotions could remain

out-interpersonally responsive, probably due to repeated engagement in this behavior European Americans, on the other hand, showed a more negative consequence when engaging in

suppression than Asian Americans (Butler et al., 2009) Considering that Asians use suppression more frequently and under more conditions, the usage of suppression can be expected to require fewer cognitive resources to execute with the process being more automated This is the

“automaticity hypothesis” that can explain the moderation effect of culture between emotion regulation strategy and its consequences

Chinese and Indians

Traditional Chinese and Indian societies are hierarchically organized and patriarchal in nature, and both cultures are predominantly interdependent and collectivistic Direct comparison

of emotion-related concepts between these two cultures is rare, but a general appreciation of Confucian and Hindu worldviews will be helpful to understanding the similarities and differences

in socialization goals and practices in Chinese and Indian cultures

Firstly, there is a different emphasis of “nature versus nurture” in these two traditions If people in a society believe that “nature” or predetermined tendencies limit and shape the

development of an individual, this society is more likely to accept individual differences Indian

culture is one example: nature is given more importance than nurture in Hindu conceptualization (Rao, McHale, & Pearson, 2003) On the other hand, in Confucian thought, the environment is considered the most important influence on the developmental course of an incident or even an individual (Wu, 1996) From this perspective, Chinese are expected to view a given incident as more associated with the context compared to Indians In other words, Chinese would be more likely to see the situation, whereas Indians see the person as the center of an event As a result,

Chinese are encouraged to think of the situation repeatedly before taking an action from (san si er

hou xing) Since this is in line with the meaning of reappraisal, it is expected that Chinese will

score higher on reappraisal scale than Indians

Secondly, traditional Chinese thought emphasizes control of emotions and behavioral conformity (Ho, 1996) Therefore Chinese are expected to report a high level of expressive suppression of their emotions In the literature there are very few studies on emotion suppression among Asian Indians In one study, Mandal and colleagues have shown that compared to

Caucasian Canadians, Indians rated negative emotions significantly more unpleasant (Mandal, Bryden, & Bulman-Fleming, 1996) The authors argue this result indicated that expressions of negative emotions are perceived with greater degree of unpleasantness, thereby the expressive behavior is less preferred among Indians Therefore, the currently available literatures suggest that both cultures may have the elements to encourage expressive suppression The direct

comparison between Chinese and Indians on their usage of different emotion expression styles has not been conducted, though Based on daily observation we would expect Indians to report higher emotion expression (or lower emotion suppression) than Chinese under certain

circumstances

As discussed in the above section, culture is a poorly defined concept with multiple facets In the existing cross-cultural research on emotion regulation, cultural differences are often represented by ethnicity differences To the best of the author’s knowledge, there are no good

scales in comparing Chinese and Indian cultures in their beliefs, values, or attitudes Therefore, this research uses ethnic background as an indicator of cultural influence, which has been widely used in intergroup comparison in Singapore context (e.g Eong, Tay, & Lim, 1993, Chua, 1998; Ounpuu & Yusuf, 2003)

Overview of the Present Research

The current research is based on the process model of emotion regulation, taking a socio-cultural functioning perspective Three studies are conducted with the main objectives of the thesis as follows:

Firstly, the main objective of the study is to compare the habitual usage of emotion regulation strategies between Chinese and Indians In the cultural psychology literature, the main focus has been put on the systematic comparison between Western and Eastern cultures on self-construct, perception and cognition, as well as emotion and motivation (e.g see Kitayama & Cohen, 2007) The comparison at a finer level, that is, comparison among groups within Western

or Eastern systems, has not been widely conducted It is likely that groups within the Western or Eastern system will show variation in the psychological domains mentioned above For example, Chinese and Indians together take up more than one third of the total population in the world, and both groups have very long history It is unlikely that they think and behave similarly The direct comparison between the two groups, however, has been rare in empirical studies Casual

observations suggest that the two groups have different emotion expression patterns For example,

it is sometimes noted that Chinese are very reluctant to say “you bother me” in public, whereas Indians will speak up with less restriction

Existing scales on emotion expression/suppression style have not shown differences between Chinese and Indians In the terminology of cultural psychology, Chinese and Indians are both considered interdependent groups, and they both express emotions to a moderate degree and

both groups reported relatively high suppression scores in general However, there is reason to believe that Indians are more emotionally expressive than Chinese under some particular

situations For instance, Rao and colleagues (2003) have documented that Indian mothers assign greater importance to emotional expression than Chinese mothers in child raring, which may reflect different socialization goals The exploration of the possibly differences in emotion

expression patterns among Chinese and Indians is one major objective of this thesis

In regard to the habitual usage of reappraisal, there have been no reports of group differences However, Asians did have a higher score of ERQ.R (though not statistically

significant) than other groups (Gross & John, 2003) When Asian groups are broken down by ethnicity, we believe the difference should be more obvious In Particular, we believe that

Chinese should report a higher ERQ.R than Indians, as discussed in the previous section

The second main objective is to explore the relationship between emotion regulation strategies and self reported well-being, and whether these relationships are moderated by cultural background Research to date (Gross & Thompson, 2007) indicates that reappraisal is associated with a higher level of happiness, while suppression is associated with a lower level of happiness

As discussed in the above sections, Chinese are expected to use both reappraisal and suppression very frequently At the same time, it has been repeatedly documented that Chinese report low SWB level compared to Western groups (Diener, Shu, Smith, & Shao, 1995) How can we explain the seemingly complicated relationship between reappraisal, suppression, and SWB for Chinese? On the other hand, Indians also report relatively low SWB (Tov & Diener, 2007) The few available studies on Asian Indians’ emotional experience suggested that Indians also inhibit their expression on negative feelings, at least compared with North Americans (Mandal et al., 1996) In this situation, how can we understand the relationship between the emotion regulation and SWB in Indians? The potential moderation effect of cultural background between emotion regulation strategies and consequences will be addressed in this research

Lastly, this study also aims to investigate the relationship between the habitual usage of different emotion regulation strategies and the cardiovascular parameters The tradition of the investigation on this relationship is long standing However, the results have remained

inconsistent and the mechanism unclear It has been reported that Chinese and Indians showed different CVR patterns in response to anger provocation (Bishop & Robinson, 2000) Is it

possible that the different CVR patterns among Chinese and Indians are due to differences in emotion regulation effectiveness? Reappraisal has been associated with more adaptive CVR profile in response to an anger provocation (Mauss et al., 2007), and suppression associated with heightened CVR (Roberts et al., 2008) Recent study has also shown that Chinese participants, compared to Caucasian participants, exhibited significantly lower reactivity of heart rate and cardiac output in response to an anger provocation when instructed to do emotion suppression (Zhou, 2010) How will the habitual usage of reappraisal and suppression be associated with CVR among Chinese and Indians? These questions will be addressed in the laboratory-based

experiment

All three studies are designed with a coherent theme on how people engage in

reappraisal and suppression, and their in relationship with health outcomes under different

cultural influences Study 1 and Study 3 are survey studies and Study 2 is a laboratory-based study

suppression, start relatively late and have been associated with less desirable consequences, such

as impaired interpersonal functioning and lower level of wellbeing Gross and John (2003)

developed a short inventory, the Emotion Regulation Questionnaire (ERQ), assessing one’s

habitual usage of reappraisal and suppression, which has been validated in different populations with different languages (e.g Zhu et al, 2008) The short inventory has also been used in some clinical populations Its reliability and validity have been established (Joormann & Gotlib, 2010)

However, there has not been any published report on ERQ among Southeast Asian

populations This study is among the first to administer the ERQ scale using a Singaporean

sample Singapore is a country that is resident to people of different ethnic backgrounds, with 74% Chinese, 13% Malay, and 9% Indian as the three main ethnic groups for its resident population (Singapore Statistics Department, 2010) Matsumoto (1993) has emphasized the importance of studying the possible cultural differences in emotion within a country In such studies, not only can researchers test the psychometric properties in different groups, inter-group comparisons can also be made In the current study, special focus will be put on the comparison between Chinese and Indians because there is evidence that these two groups may show different patterns of

emotion regulation, as well as different response to negative emotions (Bishop & Robinson, 2000; Why et al., 2003) Surprisingly few studies have compared Chinese and Indians in their emotion regulation directly, so the hypotheses are mainly based on information that was gathered through different sources, such as anthropological documents As discussed in the general introduction in Chapter 1, under the influence of Confucian traditions, it is expected that Chinese will score higher on reappraisal than Indians

Emotion regulation style can be considered as a relatively stable trait As mentioned above, the ERQ is able to measure an individual’s habitual usage of reappraisal and suppression However, it is worth noting that the expression and suppression of subjective emotional feelings are highly related to the context that elicits the emotional reaction People from interdependent cultures are believed to have distinctively different expression styles in private versus in public For example, as documented in Friesen’s study (1972), Japanese masked their expression of negative feelings with smiles when they watched a disgusting film with an authority figure, but they showed more facial expressions when they believed that they were watching the film alone Basically, display rules govern what type of emotion and how much emotion is appropriate to be shown to whom and under what circumstances, and everyone will show different emotion

expression according to the different situations People from interdependent cultural backgrounds,

in particular, may show a larger variation across different situations than do people from

independent cultures (Oishi, Diener, Scollon, & Biswas-Diener, 2004) Along this line, it is expected that Chinese and Indians will show different expression styles in front of different audiences, depending on the relationship between the subject and the audience Oishi and

colleagues (2004) also reported that situations have considerable impact on one’s affective

experiences, such that people report more positive and less negative experiences when with a romantic partner, friends, or family members than when with a stranger or a classmate/coworker Similarly, there are logical reasons to believe that people would show higher emotion expression

in front of close others than not-close others Therefore, items were developed to test the emotion expression patterns in front of different audiences among Chinese and Indians

Moreover, the relationship between the usage of a given emotion regulation strategy and SWB is tested SWB is an indicator on one’s self-reported happiness level The purpose of emotion regulation is to make one feel more positive and less negative emotions Therefore, the successful regulation of one’s emotions should be closely linked to high SWB Reappraisal helps people see the silver lining of a dark cloud, thereby it should help people be more satisfied with their current life situation, increasing one’s positive feelings and decreasing negative feelings, and thereby report fewer depressive symptoms On the other hand, suppression may have a negative influence on one’s evaluation of well-being in general When a person is unhappy and not

expressing the negative feeling, he or she is unlikely to receive any support that may change the situation This lack of natural flow of interpersonal exchange may decrease one’s life satisfaction and positive affect, and increase negative affect and ratings of depressive symptoms

There is a growing literature that tests the relationship between emotion regulation strategies and SWB directly Results are largely consistent with the prediction from the process model That is, reappraisal is positively linked to SWB and suppression shows the opposite trend However, results from different studies are not completely consistent For example, Martini and Busseri (2010) have grouped emotion regulation as passive and active strategies, according to the distinction between problem-solving and emotion-focused approaches Both reappraisal and suppression belong to passive regulation category in this study They have shown that passive emotion regulation strategies all together predict greater negative affect, lower level of

satisfaction, and less positive affect for both selves and partners in mother and adult daughter dyads This way of grouping the regulatory strategies makes the positive link between reappraisal and health consequences questionable In another study, Schutte, Manes, and Malouff (2009) used an Australia sample, and found that antecedent focused regulation strategies (including

reappraisal) were associated with greater well-being as expected, but response modulation

(including suppression of emotion expression) was not necessarily linked with negative being outcomes These different conclusions in regard to the relationship between emotion regulation strategies and health outcomes are likely due to the different measuring instruments, as well as the different participant samples used in the different studies In line with this, the current study attempts to explore the relationship between reappraisal, suppression and SWB among Chinese and Indians, using well-established instruments It is believed that reappraisal should be associated with decreased negative emotion and greater positive feelings However, the

well-relationship between suppression and SWB might be largely attenuated in our Asian population compared to Western population, based on the automaticity hypothesis

Objectives and Hypotheses

Firstly, the habitual usage of reappraisal and suppression will be compared across groups As reported in previous studies, American females scored lower on suppression level than males (Gross & John, 2003) due to the socio-cultural influence on gender role: males are

expected to be cool and express few emotions, while females are allowed to express their feelings more freely This gender difference is expected in Asian sample as well In addition, with the influence of Confucian tradition, Chinese are believed to think about a situation more, as well as exert more expression control than Indians do Moreover, a few items on one’s emotion

expression style in front of different audience are tested Hypothesis 1.1 is about ERQ scores and Hypothesis 1.2 is about the expression levels under specific situations

Hypothesis 1.1a: There will be a main effect for ethnicity such that Chinese will score

higher on ERQ.R than Indians

Hypothesis 1.1b: There will be a main effect for ethnicity such that Chinese will score

higher on ERQ.S than Indians

Hypothesis 1.1c: There will be a main effect for gender such that Asian males will

score higher on ERQ.S than Asian females

Hypothesis1.2: There will be a significant interaction effect for ethnicity by situation,

such that Indians will report greater emotion expression than Chinese with others to whom they are not close

Next, this study tests the relationship between different emotion regulation strategies and SWB, and whether gender and ethnicity moderate this relationship There is good reason to believe that reappraisal is an adaptive strategy and therefore positively associated with SWB among both males and females, and Chinese and Indians On the other hand, emotion suppression

is negatively associated with SWB (e.g Gross & John, 2003) However, the negative relationship between suppression and SWB may be attenuated in people who are good at it and use it

frequently As proposed by Butler et al (2009), individuals from cultures that promote emotion suppression may make this behavior normative and automated by daily practice thereby the associated negative consequences may become ameliorated If Chinese and Indians report

different frequencies of suppression usage, the degree of the negative consequences associated with suppression may also be different between Chinese and Indians There is evidence as well as general impression that Indians are more expressive than Chinese Therefore it is hypothesized that:

Hypothesis 1.3a: ERQ.R will show a positive association with SWB, and this

relationship will hold for males and females, as well as Chinese and Indians

Hypothesis 1.3b: On the other hand, there will be a significant moderation effect of

ethnicity on the relationship between ERQ.S and SWB, such that suppression will show

a more pronounced negative association with SWB in Indians than in Chinese

Method

Participants

Prior to recruiting participants, the procedures were approved by Institutional Review Board (IRB) of the National University of Singapore (NUS) Participants of this study included Chinese and Indian male and female university students aged between 18 and 37 years, with an average age of 21.99 years Volunteers were approached on the NUS campus The purpose of the study was explained, and after obtaining the consent from the volunteer, a booklet of

questionnaires was passed to them It took about 15 minutes to finish the survey 130 participants finished the survey Of these, data from 24 participants were not included because they belong to ethnic groups other than Chinese or Indian Thus the final data analysis included data from 106 participants (Chinese, 32 females and 28 males; Indian, 22 females and 24 males)

Measures

Emotion Regulation Strategies

Reappraisal: The ERQ reappraisal subscale was used There were six items, and one

example was “I control my emotions by changing the way I think about the situation I am in” Responses were on a 7-point scale ranging from “strongly disagree” to “strongly agree” The complete scale was listed in the Appendix A Alpha for the six items was 83 in our sample

Suppression: The ERQ suppression subscale was used to measure the general usage of

suppression One example of suppression items is “I keep my emotions to myself” (Appendix A lists the complete scale) Responses were on a 7-point scale ranging from “strongly disagree” to

“strongly agree” Alpha for the four-item scale was 72

Emotion Expression Pattern: Eight statements on one’s positive and negative emotion

expression pattern in front of different audiences were tested Responses were on a 7-point scale

from “strongly disagree” to “strongly agree” Items on the expression in front of close others were: 1) “When something good has happened to me, I share my feelings with family and close friends”; 2) “If a family member or close friend makes me happy, I tell him or her how I feel”; 3) “When something bad has happened to me, I share my feelings with family and close friends”; 4) “If a family member or close friend annoys me, I tell him or her how I feel” Items on the expression if front of not very close others were: 5) “When something good has happened to me, I share my feelings with colleagues or acquaintances”; 6) “If a colleague or acquaintance makes me happy, I tell him or her how I feel” 7) “When something good has happened to me, I share my feelings with colleagues or acquaintances”; 8) “If a colleague or acquaintance makes me happy, I tell him

or her how I feel”

Subjective Well-being

Life satisfaction: The 5-item Life Satisfaction Scale was used (Diener, Emmons, Larsen,

& Griffin, 1985) Responses were on a 7-point scale ranging from “strongly disagree” to

“strongly agree” One example was “I am satisfied with my life” The complete scale is included

in the Appendix A This scale showed good internal consistency (Cronbach’s α = 80)

Positive and negative affect: Participants indicated their general affective states on a

cluster of ten positive and ten negative emotions Responses were on a 5-point scale ranging from

“very slightly or not at all”, “a little”, “moderate”, “quite a bit”, to “very much” The emotion terms were taken from the Watson, Clark, & Tellegen (1988) PANAS study The positive cluster was enthusiastic / interested / determined / excited / inspired / alert / active / strong / proud / attentive (α = 88), while the negative cluster was afraid / ashamed / distressed / guilty / hostile / irritable / jittery / nervous / scared / upset (α = 80)

Depression: The twenty-item CES-D scale developed by Radloff (1977) was used

Participants were asked how often they felt the way stated in the questionnaire during the past

week Responses were on a 4-point scale which included rarely or none of the time (less than 1 day), some or a little of the time (1-2 days), occasionally or moderate amount of time (3-4 days),

or most of the time (5-7 days) An example of the statement was “I was bothered by things that usually don’t bother me” The complete scale is included in the Appendix A Alpha for the 20-item scale was 82

SWB index: SWB is often indicated by high life satisfaction, high frequency of

experiencing positive affect, low frequency of experiencing negative affect and depressive

symptoms Therefore, an SWB index was calculated by summing the life satisfaction score and positive affect score, then subtracting the negative affect score and depression score Standardized scores were used in all scales to give each scale equal weight in the SWB index Similar way of calculating SWB index score had been reported in previous studies, in which the Alpha was 71 (Schutte et al., 2009) Alpha for the SWB index in the current study was 68

Results

Preliminary Analyses

Structure of the ERQ Scale

Gross and John (2003) reported that the ERQ scale fit a two-factor model with an American sample Confirmatory factor analysis (CFA) was conducted to test the structure and psychometric properties of ERQ among our Singaporean sample AMOS 18 was used to compare three models: (a) one general-factor model of emotion regulation; (b) an orthogonal two-factor model with two factors correlation zero; (c) an oblique two-factor model with the factor inter-correlation freely estimated Across all standard fit indexes, the oblique two-factor model

provided the best fit Both the general one-factor model and orthogonal two-factor model were significantly worse fit than the oblique two-factor model (∆χ2 (1) = 3.99, p = 045; ∆χ2

(1) = 6.78,

p = 01, respectively)

This result was slightly different from the report by Gross and John (2003), who found that the orthogonal two-factor model was not statistically different from the oblique two-factor model with North American sample in their study In our sample, however, the reappraisal and suppression subscales were not completely independent but had an inter-correlation 316 in the two-factor CFA model Nevertheless, the two-factor model was still significantly better than the one-factor model In the two-factor model, the coefficient for each item was significant, loading

to the corresponding factor (all p < 01) Therefore, the ERQ scale also fit a two-factor model in

our Singaporean sample

SWB Scales

A series of 2 (ethnicity) by 2 (gender) ANOVA tests were conducted on each individual

scale Results revealed main effects of ethnicity (Chinese M = 4.28 [SD = 1.11], Indian M = 4.74 [SD = 1.04]), F (1, 102) = 5.46, p = 021, η2

p = 05, and gender (Male M = 4.22 [SD = 0.99], Female M = 4.73 [SD = 1.04]), F (1, 102) = 7.02, p = 009, η2

p = 06, on LS score Previous studies have documented similar patterns that females reported higher life satisfaction than males, and Chinese reported a low level of life satisfaction in general (Tov & Diener, 2007) No

significant gender or ethnic differences were detected on PA, NA, or CES-D scores

A model with LS, PA, NA, and CES-D scores loading on one latent construct,

subjective well-being, was tested with AMOS 18 All regression weights were significant (p < 01) and the model fit indices were good: GFI = 0.99, CFI = 0.99, RMSEA < 001, AIC =17.34 These results suggested that the dimensions of LS, PA, NA and CES-D formed one latent construct very well, χ2

(df =2) = 1.34, p = 512 It further supported the calculation of the SWB index by

combining these four composite scores No significant differences were found by ethnicity,

gender, or their interaction for SWB index Means and standard deviations of each scale and the SWB index were summarized in Table 1

Note CHN = Chinese, IND = Indians, LS = Life satisfaction, PA = Positive affect, NA =

Negative affect, CES-D = Depression

Intercorrelations among Variables

Intercorrelations among SWB and ERQ scales are summarized in Table 2 SWB

indicators that assess positive-related aspects (LS, PA, SWB) were generally positively associated with each other, and negatively associated with negative-related indicators (NA, CES-D) PA and

NA were only weakly associated and the correlation was not significant (r(104) = -.08, p = 412),

as reported by previous studies (e.g Kang, Shaver, Sue, Min, & Jing, 2003) Moreover, ERQ.R score was positively associated with PA and SWB index, and negatively associated with CES-D score ERQ.S score, on the other hand, was negatively associated with PA and SWB index

ERQ.R and ERQ.S did not have a significant correlation (r(104) = 15, p = 0.135)

Note: LS = life satisfaction, PA = positive affect, NA = negative affect, CES-D = scores on

depression symptoms, ERQ.R = ERQ reappraisal score, and ERQ.S = ERQ suppression score *p

< 05, **p < 01, ***p < 001

Hypothesis Testing

Group Differences of ERQ Subscales

Hypothesis 1.1 predicted that Chinese should report higher scores on ERQ.R and ERQ.S than Indians, and males would report higher ERQ.S than females Two by two ANOVA tests were conducted on ERQ.R and ERQ.S, with gender and ethnicity as between-subject

variables Consistent with the expectation, there was a main gender effect for ERQ.S, with men

scoring significantly higher than women (Male M = 4.18 [SD = 1.12], Female M = 3.63 [SD = 1.22]), F (1, 102) = 5.82, p = 018, η2

p = 05 However, Chinese and Indians reported similar level

of suppression in daily usage (Chinese M = 3.99 [SD = 1.09], Indian M = 3.79 [SD = 1.35]), F (1, 102) = 0.98, p = 324, η2

p = 01 On the other hand, there was a main effect of ethnicity on ERQ.R

as expected (Chinese M = 5.14 [SD = 0.91], Indian M = 4.74 [SD = 1.18]), F (1, 102) = 4.38, p

= 039, η2

p = 04 In summary, Hypothesis 1.1a and 1.1c were supported by the significant ethnic difference obtained on ERQ.R and gender difference on ERQ.S Hypothesis 1.1b was not

supported as Chinese and Indians reported similar ERQ.S

Emotion Expression in front of Different Audiences

Hypothesis 1.2 predicted that Indians would report higher expression than Chinese with those with whom they are not close In order to examine the overall moderation effect of cultural background on emotion expression style, a repeated measure MANOVA test was conducted, with all ratings of emotion expression in front of different audiences as DVs, relationship (close vs not-close) as a within-subject factor, gender and ethnicity as between-subject factors (Table 3) The means for the emotion expression levels in front of different audiences among Chinese and Indians are summarized in Table 4 As expected, there was a significant interaction for

relationship by ethnicity (F(4, 98) = 3.93, p = 005, η2

p = 14) The interaction was followed up with univariate tests within each ethnic group, with relationship as a within-subject variable, and

gender as a between-subject variable The effects of relationship for each pair of ratings among Chinese and Indians are summarized in Table 5 Figure 2 illustrated the interaction clearly: the interpersonal relationship between the subject and the audience moderated the expression level for Chinese in all positive and negative items that had been tested Chinese had the tendency to show high level of emotion expression in front of close others but much less so in front of not-close others In contrast, when Indians were pleased (P2) or annoyed (N2), they reported similar levels of appreciation or dissatisfaction to both close and not-close others

Table 3

Repeated Measure MANOVA on Expression In Front Of Different Audiences

pBetween

Note P1 (close) = “When something good has happened to me, I share my feelings with family

and close friends”, P1 (not-close) = “When something good has happened to me, I share my feelings with colleagues or acquaintances”, P2 (close) = “If a family member or close friend makes me happy, I tell him or her how I feel”, P2 (not-close) = “If a colleague or acquaintance makes me happy, I tell him or her how I feel”, N1 (close) = “When something bad has happened

to me, I share my feelings with family and close friends”, N1 (not-close) = “When something bad has happened to me, I share my feelings with colleagues or acquaintances”, and N2 (close) = “If a family member or close friend annoys me, I tell him or her how I feel”, N2 (not-close) = “If a colleague or acquaintance annoys me, I tell him or her how I feel”

Note df and F statistics are Greenhouse-Geisser adjusted values

Figure 2 Emotion Expression Ratings by Relationship and Ethnicity CHN = Chinese, IND =

Indian #p < 10, *p < 05, ***p < 001

Relationship between Ethnicity, Reappraisal, Suppression, and SWB

Hypothesis 1.3 predicted that ERQ.R should show a positive correlation with SWB and this relationship should hold for males and females, as well as Chinese and Indians That is, there should be a significant main effect of ERQ.R on SWB On the other hand, ERQ.S should show a negative correlation with SWB and this relationship should be moderated by cultural background

In other words, there should be a significant interaction of ERQ.S by ethnicity, such that the negative relationship should be more pronounced in Indians than Chinese To examine the

moderation effect of gender and ethnicity between reappraisal and SWB, moderated regression

analysis was conducted, using PROC GLM (SAS 9.2) SWB index was the DV; ethnicity, gender, centered ERQ.R score, centered ERQ.S score, and their interactions were entered as IVs As

expected, there was a main effect of reappraisal, B = 31, F (1, 93) = 7.88, p = 006, η2

= 0.08

However, neither the main effect of ERQ.S, B = -.04, F (1, 93) = 0.10, p = 754, η2

p = 0.00, nor

the interaction of ERQ.S by ethnicity, F (1, 93) = 0.72, p = 398, η2

= 0.01, was significant The estimated regression coefficients for the other interaction terms in the moderated regression are listed in the Appendix B In summary, Hypothesis 1.3a was supported by a significant positive association between reappraisal and SWB, but Hypothesis 1.3b was not supported as the

interaction of ERQ.S and ethnicity was not significant

Discussion

The current survey study documented ethnic differences between Chinese and Indians

in their habitual usage of reappraisal, as well as their different emotion expression levels as a function of interpersonal relationships In addition, the results suggested a positive relationship between reappraisal usage and SWB, but the general suppression usage failed to show significant association with SWB in our Asian samples

Ethnic Difference in Reappraisal Usage and Its Association with SWB

No group difference in reappraisal usage has been documented in previous research by administering ERQ to a general population Males and females, and people with different ethnic background (such as Asian Americans, African Americans, European Americans, and Latino Americans), all reported similar levels of ERQ reappraisal (Gross & John, 2003) A recent study has shown that patients diagnosed with major depression endorsed less use of reappraisal than did control group or remitted major depression patients, who did not differ significantly from each other (Joormann & Gotlib, 2010) The authors interpreted this result as an indicator that

reappraisal was an adaptive way to regulate emotion, and less usage of reappraisal could be associated with undesirable outcome

In our study, Indians reported lower ERQ.R scores than Chinese However, this group difference did not indicate that Indians were more depressive than Chinese Firstly, Indians and Chinese reported similar scores on CES-D scale Indians actually reported higher life satisfaction

and higher overall SWB than Chinese, although the latter did not reach statistical significance (p

= 09) Secondly, the relationship between emotion regulation and SWB was not moderated by ethnic background High reappraisal score in both Chinese and Indians were associated with higher SWB Therefore, the positive relationship between the usage of reappraisal and high level

of SWB has been supported with our result

Then there seems to be a contradiction: Chinese reported higher reappraisal score but significantly lower life satisfaction and near-to-significantly lower SWB than Indians Is this due

to report bias? For example, maybe Chinese have a tendency to report low SWB in general? In a very comprehensive cross-nation study, Diener and his colleagues compared SWB report from large samples in different countries They concluded that the group difference in SWB scores were not caused by reporting bias or other artifacts, but reflected true difference (Diener et al, 1995) We also believe that the report of lower SWB level may be caused by some other

cognitive habits inherited from Chinese traditions

For example, as mentioned in the previous chapter, Confucian traditional teaching emphasizes “thinking repeatedly before taking an action” This is exactly the meaning of using reappraisal to analyze a situation Additionally, Chinese tradition also encourages the pursuit of balance in every aspect of life, emotional experience included Therefore, Chinese may

deliberately down regulate positive feelings when something very good happens, by thinking of

potential bad things following the event There is a well-known idiom, sai weng shi ma

(translated as “an old man losing his horse”), describing a story of this kind The story said that once upon a time an old man lived in a north frontier city One day his horse disappeared

Neighbors came to comfort him but the old man was not upset at all and said that the loss might turn out to be a good thing He was right In the following days, not only did his horse find its way back home, it also brought with it a very precious wild horse, which could run very fast His friends congratulated him for getting a good fortune, but he simply said that “good luck” might bring misfortune Strangely, as if he had predicted it, his son broke his leg when riding the wild horse The man’s neighbors and friends were sad and showed condolence to the old man, but he insisted to look at the situation differently In the subsequent month, the army came to recruit soldiers for a brutal battle and took all males from 15 to 40 years old forcefully However the man’s son was exempted because of his broken leg, thereby was able to stay with the family

This idiom is taught to schoolchildren in order to let them understand that things can develop in quite unexpected ways Therefore, the event has to be viewed as part of a bigger picture and never be seen separately This “holistic” way of viewing the development of any event by Chinese has been thoroughly studied by Nisbett and his team (e.g Nisbett, 2003) The story also carries a message on how one should handle emotions: since nothing holds constant, there should not be dramatic emotional experiences associated with any given situation One should see the potential danger in a good situation (such as obtaining a treasure) and not to feel ultimately happy about it; on the other hand, a bad thing may involve some good element that will turn the situation to a positive direction This may help to explain the relationship between

reappraisal score and SWB among Chinese Chinese apparently think a lot about the situation, by analyzing all possible good or bad elements, and therefore report a high ERQ.R On the other hand, Chinese may also see the world not as positive as the other groups do, due to the always existing “potential danger” that may happen in the future In line with this, a few items on the potential “negative reappraisal” are developed and tested in the follow-up survey study The results are summarized in Chapter 4 Nevertheless, the positive relationship between ERQ.R and higher SWB preserves, which indicates that positive reappraisal is an adaptive emotion regulation strategy

Emotion Expression in Chinese and Indians

It was hypothesized that Chinese should report lower emotion expression than Indians

in public, because Confucian teachings particularly stress affective control, while Indians might have a tradition to emphasize self-expression more than Chinese do The result suggested a significant interaction effect for ethnicity by situation such that Chinese and Indians reported different emotion expression levels in front of different audiences Chinese reported a tendency to show distinctively different emotion expression levels in front of close others compared to not-close others, while Indians expressed similar level of direct expression of appreciation or

dissatisfaction to close and not-close others (Figure 2) The important message is that emotion expression behaviors are highly situational dependent As such, the general emotion expression or suppression score may be useful in describing the group differences in the mean expression level, but it is insufficient in detecting the differences in specific behavioral tendencies For example, Chinese and Indians reported similar level of ERQ.S, but casual observation and data from this exploratory study suggested that the two groups behave differently according to the situations

The cultural theoretical framework proposed by Triandis may help to explain this different pattern between Chinese and Indians In order to argue that cultures differ in the

tendency to sample different information from the environment, Triandis proposed a cultural model with a horizontal-vertical dimension imposing on collectivism-individualism dimension (Triandis, 2001) This results in categories of HC (horizontal collectivist), HI (horizontal

individualist), VC (vertical collectivist), and VI (vertical collectivist) All individuals have access

to all the four types of cognitions and are likely to use some combination of them, but people from different cultures may have the tendency to activate specific ones under specific situations Individualist cultures ample mostly personal, internal attributes of a person, while collectivist cultures sample mostly relations, roles, duties, and obligations On the other hand, horizontal

cultures accept equality as a given, whereas vertical cultures take social hierarchy as a given It is possible to calculate the percentage of time people from one particular culture use one of the four cognitive patterns, and this collective pattern can be seen as the cultural-cognitive signature of this society Verma and Triandis (1998) found that Chinese from Hong Kong sampled HI 25% of the time, HC 36%, VI 20%, and VC 19%, whereas Indians reported a more vertical profile (HI 24%, HC 28%, VI 23%, and VC 25%) Actually Indians showed the most vertical profile

compared to other samples in the study, such as Illinois sample with HI 38%, HC 26%, VI 23%,

VC 13%; and German sample with HI 43%, HC 27%, VI 20%, VC 10% It is also important to note that the hierarchical Cast system is still largely practiced in Indians’ everyday life

The difference between Chinese and Indians in this horizontal-vertical dimension may help to explain the similarities and differences that have been detected in the current study Both traditional Chinese and Indian cultures are collectivist, and promote affective control in pursuit of group harmony Therefore both groups reported frequent usage of emotion suppression and moderate expression under different situations that have been tested However, the socialization goal of “affect control” may be realized in different modes For Chinese, since the platform is mainly horizontal, to whom emotion should be expressed largely depends on interpersonal relationship In particular, the direct expression of dissatisfaction to not very close others is considered inappropriate because it puts the group harmony at risk To people who are close to oneself, however, the expression is actually appropriate, for obtaining support from close others is the essence of collectivist culture For Indians though, there is the tendency of being

individualistic within the collectivist cultural environment, such that self-expression is

encouraged to certain degree in Indian society (Sinha, Sinha, Verma, & Sinha, 2001) Therefore, the expression level in front of not close others was also high among Indians, compared to among Chinese

There are as many varieties of collectivism as there are collectivist cultures The

discovery of the different trend of dissatisfaction expression between Chinese and Indians is a

good example of why the examination of micro cultural systems is needed in capturing group differences

Consequence of Suppression among Asian Populations

It has been widely believed that culture shapes the impact of expressive suppression (Butler & Gross, 2009) For example, Butler, Lee, and Gross (2009) reported that instructed emotion-expressive behavior was inversely related to blood pressure in European Americans, but the direction of effect was reversed in Asian American participants In another study by the same group of researchers, results showed that Americans holding Western values reported more negative emotions when engaging in habitual suppression, compared to the counterparts who reported less usage of suppression However the deleterious effects of suppression were reduced for individuals holding more Asian values (Butler, Lee, & Gross, 2007) Zhou (2010) also

documented significantly different CVR patterns in response to anger provocation between Chinese Singaporeans and Caucasian participants when they were instructed to suppress their emotions All these results supported the notion that many of suppression’s negative impacts may

be moderated by cultural background In some other studies however, the suppression effect did not show different patterns in different ethnic groups Roberts, Levenson, and Gross (2008) reported that experimentally instructed suppression usage was associated with decreased facial expression, equal amount of negative emotional experience, increased cardiovascular activation

in all participants they tested, including African Americans, Chinese Americans, European Americans, and Mexican Americans One possible reason for this may be that all subjects were recruited from the United States, and as such, there may have been acculturation effect

Nevertheless, suppression was consistently linked with poorer health outcome, and some

inconsistency existed about the moderation role of culture on the relationship between

suppression and its negative consequences

The current study used subjects from an Asian context and the result suggested that suppression showed no significant association with SWB One possible explanation is that

emotion suppression is a context dependent behavior, and especially so for Asians ERQ.S

measures the general trend to inhibit one’s emotion expression Therefore, the relationship

between ERQ.S and SWB may be less indicative for people who regulate emotions in drastically different ways across situations As a result, a question as general as “I control my emotions by not expression them” may be less predictive on the health outcome to Asian people who change their emotion expression patterns according to the social context

Another way to understand this non-significant relationship between suppression and SWB is the “automaticity hypothesis” This hypothesis suggested that some emotion regulation strategies, which are required by sociocultural norms and learnt early in childhood, may become automated and require fewer cognitive resources to engage (Mauss, Bunge, & Gross, 2007) And therefore, the rather maladaptive consequences associated with this emotion regulation strategy

are ameliorated Moreover, the objective for emotion suppression is hiding the behavioral

responses As discussed in the process model of emotion regulation, emotional responses mainly include experiential, physiological and behavioral domains When suppression is practiced in everyday life and becomes “automatic”, according to the automaticity hypothesis, the goal should

be achieved with minimal cognitive resource, and with little impact in the experiential and

physiological domains Our result is consistent with the automaticity hypothesis, such that ERQ.S has no significant association with long-term experiential consequences, indicated by self-

reported SWB, among Asian participants who reported high suppression scores in general

In summary, different emotion regulation strategies have different relationship with SWB, and external factors, such as cultural background and situation, seem to play a role in moderating this relationship In the next study, therefore, attempts are made to examine the relationship between habitual usage of these two different emotion regulation strategies and the short-term experiential and physiological consequences within one emotional episode in Asian samples

CHAPTER III STUDY 2 Individual and Ethnic Differences in Habitual Usage of Reappraisal and Suppression

in Relationship to Cardiovascular Profile

Overview

Habitual Usage of Emotion Regulation In Relationship With Cardiac Health

It is rather common that people in everyday conversation link emotions with

cardiovascular health, such as the expression of “a broken heart” after experiencing a major disappointment The impact of emotion regulation on physical health has caught researchers’ attention for many years However, the role of emotion regulation has not been drawn as very positive in this history For example, it has been hypothesized that chronic hostility and anger inhibition may be associated with higher incident of coronary heart disease (Vogele & Steptoe, 1993) The results from different researchers form a rather united theme and suggest that tight control of negative emotions will adversely affect physical health (Gross, 1998) The exact mechanism is not very well understood, but the underlying premise usually is that inhibiting the natural generation of emotional responses leads to acute increase in sympathetic activation that may damage the cardiovascular health over the long term (Krantz & Manuck, 1984; Suls & Bunde, 2005) This reactivity hypothesis has been widely accepted as the link between acute effects of emotion regulation in single emotional episode and the long-term impact associated with CVD development

Many recent studies have adopted the process model of emotion regulation, and

differentiated antecedent- versus response-focused emotion regulation strategies These studies have documented that different emotion regulation strategies have different effectiveness, which

further leads to differentiated patterns of CVR Therefore, the habit of using different emotion regulation strategies is expected to be associated with different cardiac health outcome The results have been rather consistent that frequent usage of reappraisal is linked to more adaptive profile of cardiovascular responding to laboratory simulation tasks (Mauss et al, 2007), and that deliberate suppression of emotion leads to increased cardiovascular activation (Roberts et al, 2008)

In most studies, however, participants were instructed to use a particular emotion regulation strategy, and effects were obtained by comparing the results from different strategy groups (e.g Goldin, McRae, Ramel, & Gross, 2008) This approach is effective in documenting the differentiated trend of physiological responding through different emotion regulation

strategies, when used deliberately This set-up, however, will not be able to depict the natural association between emotion regulation and cardiovascular responses In line with this

consideration, a laboratory-based experiment was designed There was no instruction on emotion regulation, and participants were allowed to respond naturally during the entire experiment procedure The habitual usage of reappraisal and suppression was assessed by ERQ and

participants’ affective and cardiovascular responses were recorded while they performed an emotional-eliciting task In doing so, we are able to investigate the relationship between natural emotion regulation tendency and cardiovascular responding

Mauss and colleagues (2007) conducted a study that examined individual differences in their natural usage of reappraisal in association with experiential and physiological responses to a negative mood induction procedure in the laboratory Their results showed that high reappraisers reported less anger and more positive emotion, showed greater cardiac output and less total peripheral resistance, during both baseline and provocation periods These findings suggested that the tendency to use reappraisal in everyday life is associated with better emotional experience, as well as more adaptive physiological responding across different situations, including baseline and

in response to a cognitive task In Mauss’s study, though, participants were pre-screened by their reappraisal score, and only data from the top and bottom tertile, grouped as high and low

reappraisers respectively, were analyzed The current study was designed to include subjects that were not pre-screened by reappraisal score, and therefore a more representative sampling group that fully covers the distribution of reappraisal scale As shown by Mauss et al (2007), habitual usage of reappraisal is expected to be positively associated with healthy cardiovascular

hemodynamic profile in the current study

On the other hand, suppression, as a response-focused strategy, has been associated with fewer emotion revealing facial expressions, increased cardiovascular activation, and no impact on subjective emotional experiences, a pattern that holds for African Americans, Chinese Americans, European Americans, as well as Mexican Americans (Roberts et al., 2008) In the Roberts et al study, participants were instructed to suppress their expressions when they were watching a disgust-eliciting film Therefore, the response difference between suppression and control groups may be due to deliberate efforts in following the instruction, but not necessarily the deleterious consequence of emotion suppression per se In short, it is unclear whether high level of emotion suppression in daily life will have an impact on one’s CVR in response to a negative mood-eliciting task In line with the “automaticity hypothesis” it should not Because someone who uses suppression often will become good at it; this behavior can be carried out with fewer cognitive resources being consumed, and with little affective or physiological cost This hypothesis under emotion regulation context, however, has not been tested in laboratory-based experiment Therefore, the correlation between habitual suppression score and the CVR in

response to stress will contribute to our understanding of whether the usage of suppression has a direct influence on one’s cardiac health

Moderators Between Emotion Regulation and Its Consequences

With an increasing number of studies on the effect of reappraisal on experiential and cardiovascular reactivity, very few studies with Southeast Asian population have been reported Chinese and Indian both belong to interdependent cultures and emphasize affect control Study 1 showed that these two Asian groups had different reappraisal scores and expression patterns However, ethnic background did not moderate the relationship between ERQ.R or ERQ.S and SWB in the survey study On the other hand, there was evidence for ethnic differences in the expression of emotions in different interpersonal contexts In addition, previous studies have reported that Chinese and Indians showed different cardiovascular responses to anger

provocations (e.g Bishop & Robinson, 2000; Why et al., 2003) It poses an interesting question

as to whether cultural background and interpersonal relationship moderate the relationship

between emotion regulation and cardiovascular parameters

In the present study, interpersonal relationship was manipulated using a feedback paradigm Participants were asked to perform a cognitive arithmetic counting task (adopted from Mauss et al, 2007) and made to believe that another subject, either a stranger (Stranger condition) or a friend (Friend condition), was delivering timely feedback on their

performance-performance There were two considerations when the experiment was designed

Firstly, the task is designed in this way to make sure participants are irritated to some degree Suls and Wan (1993) conducted a quantitative review on the relationship between trait hostility and cardiovascular reactivity They concluded that the relevant psychosocial factor is only related to CVR under provocation during interpersonal tasks Therefore, the performance-feedback paradigm has been adopted in the current study Evaluations of one’s performance occurring in daily life generally elicit subjective appraisal of one’s specific ability besides

objective right or wrong information, thereby inducing emotional responses (Leng & Zhou, 2010)

In particular, participants in this study received pre-programmed negative feedback regardless of their actual performance It was believed that the rather difficult arithmetic counting task in front