Anger regulation and culture role of culture in consequences of anger expression, suppression and reappraisal

... Albert, 2007) In Eastern culture encouragement of harmonious interpersonal relations and discouragement of conflicts is one of the principles of problem solving Culture and anger regulation Individuals’... that reappraisal is more beneficial than suppression in both experiential and physiological domains Suppression and reappraisal in the context of anger Next consider the consequences of suppression. .. consequences of suppression and reappraisal in the context of anger As indicated, effects of anger suppression (anger- in) on cardiovascular responses have been intensively examined in previous research

of studies focusing on relationships between anger expression styles and

cardiovascular disease (CVD) To date, the literature suggests that both anger

exhibition and inhibition are related with higher risk of development of CVD

Literature on anger expression style and its relation with CVD will be reviewed in this section

The hypothesized underlying mechanism linking anger regulation and CVD is through exaggerated cardiovascular reactivity (CVR) triggered by acute anger

episodes CVR is defined as change of cardiovascular parameters (e.g., blood

pressure, heart rate, cardiac output and total peripheral resistance) from baseline Repeated experience of exaggerated CVR is considered as the key linking acute effects of anger regulation and long term influence on cardiovascular disease CVR has been found to be an independent risk factor of CVD

Given adverse outcomes of both anger exhibition and inhibition, healthy regulatory strategies are of interest Cognitive reappraisal is proposed as a healthy anger regulatory strategy By comparing with suppression, reappraisal was related with beneficial outcomes in various domains including more adaptive cardiovascular

Literature Review Anger Regulation and Consequences

Anger regulation and its relation with cardiovascular health

It’s believed that anger is related with many domains of functioning including subjective well-being, social relationships, and physical health Especially given its possible relation with the development of cardiovascular diseases, research on anger regulation has attracted interest of researchers for decades There have been two main streams of research on anger regulation and cardiovascular health One is a huge body

of epidemiological investigations with cross-sectional, case-control or cohort designs The other is a series of laboratory experiments in which acute effects of situational anger expression methods are examined

A correlation of anger exhibition and negative consequences on

cardiovascular responses has been obtained in cross sectional studies There seems to

be a tendency that people who express anger more openly have greater cardiovascular

3

reactivity For example, an investigation with a Mexican American sample suggested that people scoring high in Anger-out scale of STAXI-II had greater systolic blood pressure (SBP) and diastolic blood pressure (DBP) reactivity than people low in Anger-out (Gleiberman, Greenwood, Luke, Delgado, & Weder, 2008) Another study found that people high in Anger-out displayed high blood pressure during work compared with people low in Anger-out (Bongard & al'Absi, 2005)

A large body of epidemiological research with more controlled design has been done to clarify the role of anger expression in the development of CVD In a case-control study, a higher likelihood of myocardial infarction was found associated with Anger-out in a sample of low SES middle-aged men ( Mendes, 1992) Results

of another case-control study suggested that quicker experience of anger with greater verbal expression was the best predictor of CHD compared with other types of anger experience and expression (Atchison & Condon, 1993) Stronger evidence has been obtained in prospective research Everson and colleagues (1998) studied 537 healthy volunteers for 4 years for the development of hypertension They found that 1 point increase out of 4 points in Anger-out scale increased by 12% the risk of incident hypertension after controlling all other risk factors of hypertension Anger expression was also found to influence patients’ prognosis after a coronary event A sample of female patients with CHD was followed up for an average of 6.4 years for total mortality and the combination of cardiovascular death and non-fatal acute myocardial infarction (AMI) The results indicated that a 1 point increase in the four-point Anger-out subscale of STAXI was associated with a 42% increase in the risk of cardiac death or a new AMI (László, Janszky, & Ahnve, 2008)

4

In the population research above, anger expression was examined as a

relatively stable disposition and there has been evidence that it is related to the

development of CVD in healthy people and the prognosis of patients with CVD In laboratory research, the manipulation of anger expression also impacts CVD and cardiovascular reactivity (CVR) Siegman and colleagues (1990) found that both physiological and experiential arousal were affected by how anger was vocally

expressed Anger expression in a fast-loud manner was associated with greater blood pressure and heart rate reactivity than expression in a normal manner (Siegman, Anderson, & Berger, 1990) Moreover, participants were found to have high levels of CVR in an anger-out condition in which they were asked to express anger fast and loudly whereas the heighten CVR was not observed when they were asked to keep anger-arousing events inward (Siegman & Snow, 1997)

Although evidence of negative consequences of anger expression has been obtained in many studies, there are also conflicting findings The deleterious

outcomes of anger expression were not found in some epidemiological research Anger expression was even found as a protective factor of CHD in some studies For example, habitual usage of anger-out was associated with lower heart rate and

norepinephrine reactivity in a laboratory induced stressor which was a 5 min mental arithmetic task (Mills, Schneider, & Dimsdale, 1989) And a cohort study, in which 23,522 male professionals were followed up for 2 years, suggested that moderate anger expression was a protective factor for CHD among high socioeconomic status professionals (Eng, Fitzmaurice, Kubzansky, Rimm, & Kawachi, 2003)

The inconsistency of results may come from the limited external validity of previous studies The differences on demographic characteristics of samples,

5

selection of anger expression measures and the selection of research method can influence conclusions For example, anger expression can be a protective factor for the male professionals with high socioeconomic status (Eng, et al., 2003), but a risk factor for the sample of low SES men (Mendes, 1992) It may imply that career or SES moderates the relation between anger expression and its health consequences Selection of research method plays a role in the inconsistency of results too For epidemiological research, prospective methods provide stronger evidence for causal relationship than case-control studies and cross-sectional investigations Further, Anger-out was used as a personality disposition in epidemiological studies whereas it was manipulated by vocal behaviour in some laboratory studies, e.g., Seigman’s work The extent to which the manipulation for anger expression is successful impacts the internal validity of experimental studies

Evidence of a relationship between anger suppression and cardiovascular diseases has been obtained as well High Anger-in was found related with increased arterial stiffness in adults which is part of the pathogenesis of CHD (Anderson,

Metter, Hougaku, & Najjar, 2006) MacDougall and colleagues (1985) reported that Anger-in was significantly associated with the severity of CVD in patients

undergoing angiography In the Framingham Heart Study between 1965 and 1967,

1674 coronary disease free individuals were followed up for the development of CHD over an eight-year period The results indicated that not discussing anger was an independent predictor of CHD incidence when controlling for all other coronary risk factors and other psychosocial scales (Haynes, Feinleib, & Kannel, 1980) Although some studies did not find relationships between suppressed anger and CHD

(Dembroski, MacDougall, Costa, & Grandits, 1989), most findings from

6

epidemiological studies indicated that anger suppression was an independent risk factor for development of CHD (Haynes, et al., 1980; Kneip et al., 1993; MacDougall, Dembroski, Dimsdale, & Hackett, 1985) Laboratory research reaches a similar conclusion Anderson & Lawler’s (1995) study revealed that women who suppressed anger experienced grater BP change in anger recall task Another experiment found high inhibition coupled with family history of CVD was related with the highest SBP

in a sample of 60 boys aged 12-16 (Vögele & Steptoe, 1993)

Taken together, the previous results seem to suggest that extremes of expression in both directions are related to adverse cardiovascular outcomes This is also found in some studies interested in cardiovascular consequences of both anger exhibition and inhibition (Everson, Goldberg, Kaplan, Julkunen, & Salonen, 1998; László, et al., 2008) The adverse consequences of anger expression towards both sides also indicate that healthy anger regulation is far more than simply expressing or suppressing

Mechanism of the link between anger regulation and cardiovascular diseases

To link the acute effects of regulation in single anger episodes and the long term influence of anger regulation on the development of CVD, the reactivity

hypothesis was proposed and has been widely accepted According to the reactivity hypothesis, frequent experience of heighten cardiovascular responses may have deleterious impact on cardiovascular functioning and facilitate the development of CVD In laboratory experiments, researchers are interested in comparing

cardiovascular reactivity (CVR) which serves as an indicator related to cardiovascular health Individuals who exhibit exaggerated cardiovascular responses to mental

7

challenges have a higher risk for subsequent cardiovascular diseases in contrast to their low-reactive counterparts (Kaplan, Manuck, Williams, & Strawn, 1993; Manuck, Kamarck, Kasprowicz, & Waldstein, 1993) Cardiovascular reactivity in a specific situation is also influenced by environmental variables For example, anger regulatory strategy can be a variable that affects cardiovascular reactivity in anger situations If a certain anger regulatory strategy is related with higher levels of CVR, people using it frequently may have higher risk of cardiovascular disease compared with other

regulatory strategies Results of some laboratory studies suggested that both anger expression and anger suppression were related with exaggerated CVR (Engebretson, Matthews, & Scheier, 1989; Houston, 1994) In line with this logic, suppression and exaggerated expression may share the same pathway as related to adverse

cardiovascular outcomes

However, findings of experimental research on relations between anger manipulation and consequent CVR are not consistent For example, an association between exaggerated CVR and anger suppression after provocation was observed in some studies (Engebretson, et al., 1989; Houston, 1994), but not in others (Powch & Houston, 1996; Smith & Houston, 1987) Anger expression was not consistently found related with greater CVR as well (Smith & Houston, 1987) The inconsistency may be due to differences in the effectiveness of angry provocation tasks, different timing of reactivity measurement and ignoring the influence of habitual anger

expression style (Engebretson, et al., 1989)

Can induced CVR in laboratory studies predict CHD? This question is crucial

to bridge the acute effect obtained in laboratory studies and long term health

outcomes There are several considerations in addressing this question First, repeated

8

exaggerated CVR may contribute to an increase of peripheral resistance which

ultimately contributes to the development of CVD There has been evidence that frequent and chronic exaggerated CVR may promote hypertrophy of the smooth muscle in arterioles which can lead to vasoconstriction (Mauss & Gross, 2004) The cardiac output which is excessive relative to demand can also trigger peripheral vasoconstriction Both factors may lead to chronically increased peripheral resistance Second, the trigged release of catecholamines during repeated CVR episodes is

related to the development of atherosclerotic plague Catecholamines, particularly epinephrine, may cause injury of the intimal endothelium of the coronary arteries in hemodynamic and/or biochemical ways Once the coronary endothelium is damaged, catecholamine-induced release of free fatty acids in excess of levels needed for

metabolic requirements can cause platelet aggregation, smooth muscle proliferation, and the deposition of core lipids within the lesion area CVR has been found as an independent risk factor for development of ischemic heart disease (Kamarck et al., 1997; Treiber et al., 2003) and hypertension (Carroll, Smith, Sheffield, Shipley, & Marmot, 1995; Carroll et al., 2001; Markovitz, Raczynski, Wallace, Chettur, &

Chesney, 1998; Menkes et al., 1989)

Aside from the reactivity hypothesis, Brosschot & Thayer (1998) proposed an alternative mechanism linking anger suppression and cardiovascular disease which emphasized low parasympathetic activity and slow cardiovascular recovery rather than high sympathetic tone and elevated reactivity According to their theory,

parasympathetic activity (vagal tone) is suppressed during stressful circumstances which require organized behavioral and autonomic responses (Porges, 1991) As such, low vagal tone may be involved in the process of anger suppression From a

9

physiological perspective, low vagal tone is associated with decreased control of heart rate and thus slower recovery of both heart rate (HR) and blood pressure (BP) This reasoning has been supported by experimental studies in which slower

cardiovascular recovery when suppressing anger rather than expressing anger has been observed (Engebretson, et al., 1989) Slow recovery is considered as a critical factor for the development of CVD (Brosschot & Thayer, 1998) The relationship between slow cardiovascular recovery and CVD has been found in empirical studies Such findings can serve as evidence of this model

Cardiovascular responses to anger episodes may vary in magnitude,

frequency, and duration Reactivity captures response magnitude whereas low vagal tone captures duration (Schwartz et al., 2003) Combining two mechanisms may give better explanation of the link of anger regulation and CVD

Regarding the specific cardiovascular parameters, indices utilized in the present study includes cardiac output (CO), pre-ejection period (PEP) total peripheral resistance (TPR), heart rate (HR), systolic blood pressure (SBP) and diastolic blood pressure (DBP)

HR is one of the most common indicators of autonomic activation, which reflects both sympathetic and parasympathetic influences on the sino-atrial node and the atrioventricular node CO is the volume of blood pumped by the heart per minute

It serves as a general indicator of myocardial contractility PEP is another cardiac indicator which refers to the time interval between onset of contraction to the ejection

of blood PEP has been widely used in the cardiovascular psychophysiology research

as an indicator of cardiac performance PEP was considered related to contractility of the ventricles of the heart (Brownley, Hurwitz, & Schneiderman, 2000; Wilson,

10

Lovallo, & Pincomb, 1989) TPR is an index of total peripheral resistance and was used to indicate vascular responses SBP and DBP were also measured Both SBP and DBP were determined by myocardial contractility and vascular resistance

However, SBP is more determined by the force with which the heart contracts

whereas DBP is more related to systemic vascular resistance (Wright & Kirby, 2001)

It has been argued that the cardiovascular system responds with a limited number of organized patterns of response (Allen, 2000) There is growing literature looking at the cardiovascular responding in patterns rather than looking at each index independently (e.g., Blascovich & Katkin, 1993; Blascovich, Mendes, Hunter, & Salomon, 1999; Tomaka, Blascovich, Kibler, & Ernst, 1997) CO and PEP are usually used as cardiac indicators and TPR is used as a vascular indicator

Comparison of cardiac and vascular activities differentiated two typical patterns of cardiovascular reactivity which are labeled as the challenge pattern and the threat pattern (Tomaka, Blascovich, Kelsey, & Leitten, 1993) Challenge vs threat patterns were originally determined by cognitive appraisal of demand of task and available resource for coping The challenge pattern occurs when personal abilities are

appraised as congruent with situational demands Challenge pattern was found related

to an increase in cardiac activities and little or no change in systemic vascular

resistance In contrast, the threat pattern occurs when psychological resources are perceived as less than task demands The physiological responses of threat pattern is moderate increase in cardiac activities along with an increase in vascular resistance (Mendes, Reis, Seery, & Blascovich, 2003; Tomaka, et al., 1993) CVR responses in the current study will be viewed in the context of these patterns

11

Anger regulation and CVR

Based on the existing research, anger is related to cardiovascular health and the regulation of anger is more than simply expression or suppression It is, therefore, important to explore effective methods to regulate anger To address the topic of anger regulation, the bigger picture of general emotion regulation will be briefly reviewed

Gross (1998) proposed a process model to specify each step of emotion regulation and classify two broad forms of emotion regulatory strategies: strategies to manipulate the input for emotion (antecedent-focused strategies) and strategies to manipulate the output of emotion (response-focused strategies) Antecedent-focused regulation means to regulate the potential emotion before it is triggered Take regulating negative emotion as an example One can choose to avoid some people and situations to avoid the likely negative emotion response; turn attention away from the emotional cue or reevaluate the situation in order to alter the emotion Response-focused emotion regulation deals with emotional responses when the emotion is already generated There are several types of response-focused regulation, such as to intensify, diminish, prolong or curtail ongoing experience, expression of the emotion

or physiological responding (Gross, 1998)

As representatives of antecedent-focused regulatory strategies and focused regulatory strategies respectively, reappraisal and suppression were compared on many domains of functioning including cognition, social relationship and physical health These two strategies were also chosen to be examined in this present study for two reasons First, the concept of suppression corresponds to anger inhibition in the context of anger regulation Reappraisal appears to be a beneficial

response-12

strategy judging from the process model theory and the available empirical studies The effect of reappraisal in anger situation is worth being examined Second, there has been a large body of literature comparing the effects of these two strategies, which provides rich background information to facilitate the present research

Comparison of suppression and reappraisal

Suppression is defined as inhibition of emotional expressive behavior when emotionally aroused It is a type of response-focused regulatory strategy Reappraisal

is a form of cognitive change by which a potential emotion-eliciting situation would

be reconstrued and the emotional impact would be changed (Gross & John, 2003) It

is a form of antecedent-focused regulatory strategy The acute effects of suppression and reappraisal on experiential, cognitive and physiological domains have been examined in laboratory studies In some studies, researchers manipulated the usage of suppression and reappraisal experiments and compared their effects In other studies, habitual usage of suppression and expression were used as a trait to predict the

experiential and physiological responses in the situation created in laboratory Results

of the experimental studies revealed that instructed usage of suppression did not reduce, and might even increase, negative emotional experiences such as disgust, anger, sadness and embarrassment (Gross, 1998; Gross & Levenson, 1993, 1997; Harris, 2001) As a result of increasing cognitive load, suppression impaired memory for socially relevant information (Richards & Gross, 2000) Suppression was found accompanied by increased sympathetic and cardiovascular responding in most studies (Gross, 1998; Gross & Levenson, 1993, 1997) In contrast, instructed usage of

reappraisal was related to less negative experience and more adaptive cardiovascular responding (Gross, 1998)

13

With respect to the acute effects of habitual usage of suppression and

reappraisal, similar results were found Egloff and colleagues (2006) observed the spontaneous usage of suppression and reappraisal in the standard setting and

measured the corresponding emotional experience, expressive behavior and

physiological response Their results suggested that suppression had no impact on negative affect but was related to less expressive behavior and higher physiological activation In contrast, reappraisal was related to less expressive behavior but had no impact on negative affect and physiological responses In a study that examined anger specifically, it was found that people high in reappraisal felt less angry, showed greater cardiac output and ventricular contractility, and less total peripheral resistance than people low in reappraisal (Mauss, Cook, Cheng, & Gross, 2007)

As noted above, emotion regulation style can be considered as a relatively stable trait Gross and colleagues (2003) developed the Emotion Regulation

Questionnaire (ERQ) to measure individual differences in habitual usage of emotion suppression and reappraisal A series of studies has been done to compare the

consequences of habitual usage of these two regulatory strategies Instead of focusing

on the immediate effects in the experimental studies, the questionnaire studies looked

at the long term and cumulative effects of using suppression and reappraisal on a regular basis Findings revealed that suppression was related to more negative

experience, poorer well-being and less interpersonal closeness On the other hand, reappraisal was related to less negative experience, higher level of life satisfaction and greater social closeness (Gross & John, 2003; Richards & Gross, 2000)

14

Taken together, evidence from experimental studies and questionnaire

surveys demonstrates that reappraisal is more beneficial than suppression in both experiential and physiological domains

Suppression and reappraisal in the context of anger

Next consider the consequences of suppression and reappraisal in the context

of anger As indicated, effects of anger suppression (anger-in) on cardiovascular responses have been intensively examined in previous research Available evidence suggested anger suppression was related with greater CVR and higher risk of CHD

Although reappraisal has been found to be beneficial when regulating some specific emotions (e.g., disgust and sadness), it has not been much studied in the context of anger For example, only two studies on the relationship between

reappraisal of anger and CVR responding was found In one study, researchers

manipulated the regulatory strategies used Results revealed that reappraisal led to less negative experience, less intensity of CVR and faster CVR recovery compared with rumination (Ray, Wilhelm, & Gross, 2008) In the other study, effects of

habitual usage of reappraisal were examined People used reappraisal habitually in daily life showed less cardiovascular reactivity than people seldom use reappraisal to regulate anger (Mauss, et al., 2007)

Reappraisal has been widely used in clinical practice of anger management Most anger management programs are designed following the Cognitive-Behavioral Approach, in which cognitive reconstructing plays a vital role Clients are usually asked to think before they react according to the automatic wind-up thoughts

triggered by the anger event They are trained to use monitoring and

of anger situations can engender beneficial experiential and physiological outcomes Thus, together with anger expression and suppression, effects of reappraisal will be examined in the present study

Moderation of Culture in the Relationship between Anger Regulatory Strategies and Consequences

One important limitation of the previous research on emotion regulation is that it ignores the influence of culture on the consequences of emotion regulation Most studies were conducted in Western countries with predominantly European American participants Given notable differences between Asian and European American cultures, results in the Asian cultural context might be different In the following section, relation of culture and emotion regulation will be explained and cultural influence of anger regulation will be emphasized

Western culture vs Asian culture

Kroeber & Kluckhohn (as cited in Tsai, Levenson, & McCoy, 2006) defined culture as socially shared transmitted patterns of ideas (values, norms, and beliefs) that are instantiated in everyday practices, institutions, and artifacts Individualism- collectivism is a well-known dimension in the culture studies related to Western culture and Asian culture According to this theory, a fundamental difference between

16

members of individualistic cultures and members of collectivistic culture lies in their view of the relation of self, group and environment Individuals in an individualistic cultural context place self over the group concerns They are more concerned with their own views, needs, and goals rather than others In contrast, members of

collectivistic cultures place group over individual concerns They are more concerned with the views, needs and goals of the group rather than themselves (Leung & Bond, 1984) Due to the fundamental differences in value system, members of Western culture and Asian culture are different in many aspects of social life Differences in patterns of social relationship and emotion regulation will be elaborated because they are directly related with how people deal with anger

In the research on cultural differences in social relationships, the Western style was described as individual orientation which refers to a form of relationship that “has autonomy as its predominant feature It motivates the person to dominant, control and change the biological, physical and social environment to fulfil personal desires, motives and ambitions” (Yang, 1995, p 21) In Western culture, relationships are evaluated by the extent to which they meet one’s personal needs Because of the frequent expression and negotiation of personal needs, interpersonal conflicts are considered as inevitable (Mesquita & Albert, 2007)

In contrast, the Eastern style of social relationships has been described as a collectivistic orientation which “has harmony as predominant feature And it found its expression in people’s attempts to establish and maintain harmonious relationship with the environment by submitting to, cooperating with, or merging into the physical, biological, and social surroundings” (Yang, 1995, p 21) Individuals in Eastern culture are considered in relation to others and belonging to social groups

17

Relationships are based on the fulfilment of role-based obligations ( Mesquita & Albert, 2007) In Eastern culture encouragement of harmonious interpersonal

relations and discouragement of conflicts is one of the principles of problem solving

Culture and anger regulation

Individuals’ beliefs, values, attitudes, affect and behaviours are heavily

influenced by culture Emotion regulation is also always embedded in the meanings and practices that constitute the socio-cultural world Cultural models focus attention, guide perception, lend meaning and imbue emotional value

How culture shapes emotions

Kitayama et al (2006) proposed a dual process model to explain how culture shapes individuals’ emotional responses Social affordance and social reinforcement are two important mechanisms In terms of social affordance, a specific culture

provides dominant themes for some emotions rather than others; for example,

Western culture nurtures disengaging emotions which come from success or failure in personal achievements whereas Eastern culture nurtures engaging emotions which derive from success or failure in relationship goals For example, in one study

Japanese showed a pervasive tendency to report experiencing engaging emotions more strongly than they experienced disengaging emotions, but American showed a reversed tendency (Kitayama, Mesquita, & Karasawa, 2006) Additionally, culture modifies emotion responses by valuing and devaluing the corresponding experience, expression style or regulatory strategy This is the process called social reinforcement

18

Culture differences in anger regulation

Anger is a negative emotion which mostly occurs in interpersonal situations Anger is usually generated when an event is unexpected, unpleasant, obstructs goals,

is seen as unfair and as caused by other people (Mesquita & Ellsworth, 2001) Based

on the research of Kitayama and colleagues, anger is considered as a socially

disengaging emotion, which is grounded in independence and autonomy of self

(Kitayama, et al., 2006) As a public emotion, regulation of anger is heavily

influenced by cultural context The regulation of anger is determined by anger

experience and social norms in different situations

As noted above, in the Chinese cultural context, social harmony is highly valued People are educated to maintain interpersonal harmony by avoiding conflicts Open expression of anger is considered as impropriate in most social situations Chinese tend to moderate their actions and performance to fit social norms and others’ opinions and tend to ignore their own feelings (Yang, 1995) And that’s why situation seems a better predictor of behaviour than personality for Chinese people (Wang & Cui, 2006) It seems that the socialization process of Chinese is the process of

learning to tolerate and live with unavoidable inconsistencies between the public self and private self In contrast, although open expression of anger may bring negative consequences for relationship in Western societies, norms against open expression of negative emotion are weaker with less social pressure for Westerners against

expressing negative emotions such as anger Additionally, individuals in a Western cultural context are perceived to be autonomous and self-sufficient and may be less concerned about the social restraint of anger expression Finally, expressing anger directly sometimes can be interpreted as assertiveness which may be encouraged

19

Hence, the open expression of anger can be reinforced by cultural values For these reasons, Caucasians were expected to use expression more than suppression in daily life

Therefore, it’s reasonable to hypothesize Chinese are more likely to use suppression to regulate anger in daily life rather than expressing it and the reverse is true for Caucasians Further, since Chinese receive positive reinforcement when they suppress anger and they are used to doing that, the consequences of suppression would not be as harmful as literature has suggested for Westerners These two

hypotheses will be examined in this present study

Regarding the usage of reappraisal, from Confucian teachings, Chinese

people are educated to think repeatedly before taking action (san si er hou xing) This

is in line with the meaning of reappraisal Chinese may be more likely to use

reappraisal to avoid conflicts and help them to fit others’ expectation This question will be also addressed in the current study

Empirical evidence of cultural differences in emotion regulation

Several empirical studies have been conducted to examine cultural

differences in emotion regulation Gross & John (2003) assessed emotion suppression among different ethnicities in the United States and found that minorities, including Asians, reported higher levels of habitual suppression than did Caucasians (Gross & John, 2003) Suchday & Larkin (2004) reported Asian Indian men had higher

habitual anger suppression levels than Caucasian men Butler et al (2007) categorized their sample by cultural values and found women with predominantly European values reported lower levels of habitual suppression than did women with bicultural European-Asian values

20

Does habitual use affect the outcomes of anger regulation strategies?

Engebretson, Matthews, & Scheier (1989) proposed a “matching hypothesis” to answer this question The matching hypothesis states that when an individual uses a particular method of dealing with anger (e.g., anger suppression) that is different from his or her typical method of dealing with anger (e.g., anger expression), he or she will show exaggerated cardiovascular responses to those situations Their

experiment provided evidence for the hypothesis by indicating that individuals

allowed to act in their preferred mode of anger expression showed a quicker decrease

in SBP reactivity than those not allowed to act in their preferred mode of anger expression

Considering that Asians use suppression more frequently, under more

conditions, the usage of suppression may be more automatic and require fewer

cognitive resources to execute The consequences of anger suppression should be different between Asians and Caucasians This reasoning gets some support from previous research Suh & colleagues (1998) found that emotional ambivalence, which refers to a conflicting desire to express versus inhibit emotions, was negatively

related to life satisfaction for Caucasians, but not for Chinese Butler, Lee & Gross (2007) compared Caucasian Americans and Asian Americans on anger suppression and its social consequences They found deleterious effects reduced when people with more Asian values used suppression

In terms of cardiovascular response, the existing findings are inconsistent Suchday & Larkin (2004) did a comparison of cardiovascular reactivity between Indian and Caucasian men in an anger provocation interaction They found that for Indians, the recovery of DBP was delayed when anger expression preceded anger

21

inhibition compared to the reverse order For Caucasians, DBP recovery did not differ between the two orders In a more recent research, Caucasian American females and Asian American females were compared in terms of emotion expressivity and blood pressure in a face-to-face interaction about a distressing film Increase in emotion expressivity was found to be associated with a decrease in blood pressure in

Caucasians dyads, but the reverse was true in Asian American dyads (Butler, Lee, & Gross, 2009)

In general, the consequences of anger regulation in an Asian cultural context have so far received insufficient research attention One purpose of the present study

is to clarify the effects of habitual usage of anger regulatory strategies on the

consequences of regulation for Asian and Caucasian participants

In the existing cross-cultural research on anger regulation, culture differences are often represented by ethnicity differences Although ethnicity grouping is a

reasonable way to represent different cultural models, it is still unclear the degree to which the differences are due to cultural factors In the present study, cultural values will be measured to get a clearer look at the relationship between culture and emotion regulation

Overview of the Present Study

The purpose of present study was to examine the experiential and

cardiovascular outcomes of three specific anger regulation strategies (i.e., expression, suppression and reappraisal) in a laboratory environment Two cultural groups were involved in the experiment and the moderation effect of culture in the strategy-

outcome relationship was examined

orientation which is a main feature of social relationships in Chinese culture, Chinese participants are expected to be more likely to engage in reappraisal (thinking from the partner’s point of view) in anger situations

Hypothesis 2

Culture will moderate the relationship between anger expression and

suppression and the corresponding emotional experience Experiential outcomes of reappraisal are hypothesized as most beneficial in both cultural groups

Caucasians will feel the greatest intensity of anger when they are asked to suppress anger, be less angry when expressing anger outwardly and least angry when doing reappraisal Chinese are expected to experience greatest intensity of anger when asked to express freely, be less angry when suppressing anger and least angry when doing reappraisal The same pattern of means is expected for negative emotion while the reverse pattern of means is expected for positive emotion

23

anger are in line with the challenge pattern (i.e., significant increase in CO, decrease

in PEP and decrease or no change in TPR) (Herrald & Tomaka, 2002), therefore, CVR responses of participants across all conditions are expected to show a challenge pattern

Furthermore, Caucasians assigned to the anger suppression condition will have greatest CVR challenge reactivity, followed by Caucasians in the expression condition and the reappraisal condition In contrast, Chinese assigned to the anger expression condition will have the greatest CVR challenge pattern, followed by Chinese in the suppression condition and the reappraisal condition

24

CHAPTER 2 METHOD

Participants

Forty-five Chinese and 45 Caucasian females were recruited in the

experiment Chinese participants were undergraduate students from the Research Participation Program of Psychology Department of National University of Singapore (NUS) and earned course credits for participation Caucasians were exchange

students at NUS from North American and European countries who volunteered to participate in the research activity and got S$10 in return Chinese participants were 20.82 years old on average with the standard deviation of 0.31 and the average age of Caucasian participants was 21.00 years with the standard deviation of 0.26 There

was no significant difference in age between two groups, t (87) = -.44, p ns To rule

out the possible confounding effect of gender, only female participants were involved

in the experiment

Potential participants were screened by a demographic questionnaire to ensure they had not been diagnosed with heart disease and hypertension and were not under any medication which might affect cardiovascular indices They were also screened for cultural background For Chinese participants, the requirements were that, (a) parents and grandparents were Chinese; (b) parents and grandparents were born and raised in Asian countries; (c) participants themselves were born and raised

in Asian countries and (d) participants were able to speak Chinese (e.g., Mandarin) For Caucasian participants the requirements were that (a) parents and grandparents were European or European American; (b) parents and grandparents were born and

Table 1 Cultural background Congruence

with original

culture

Ethnicity of Parents and grandparents

(N=89)

Parents and grandparents born and raised

26

Equivalencies on health condition and habits were examined between the cultural groups As shown in Table 2, two cultural groups did not differ in family history of CVD, χ2

(1, N=89) = 1.15, p ns and the habit of smoking, χ2 (1, N=89) = 1.03, p ns A marginal significant difference of BMI between two groups emerged, t (85) = -1.98, p=.051 Chinese and Caucasians also differed in alcohol consumption,

χ2

(1, N=89) = 24.96, p < 01, and frequency of exercise, t (87) = -2.09, p = 039

Due to effect of estrogen on cardiovascular functioning (Mendelsohn & Karas, 1999) and possible enhanced responsiveness to psychosocial stressor in the luteal phase relative to situation in the follicular phase (Kajantie & Phillips, 2006), menstrual cycle was also investigated No significant difference was found between two cultural groups with respect to menstrual cycle stage, χ2 (1, N=86) = 30, p ns

Table 2 Demographic Equivalence Characteristic Chinese Caucasians χ2

Note an=44 bn=43 cn=45 dn=44

*p<.05 **p<.01

27

Design

The present experiment used a 2 (cultural background) by 3 (anger regulatory strategies) between subjects design There were two levels of cultural background (Asian and European/American) and 3 levels of regulatory strategy (anger

expression, anger suppression and reappraisal) Participants in each cultural group were randomly assigned to expression, suppression or reappraisal conditions Two sets of dependent variables (changes in emotional experience and indicators of

cardiovascular reactivity including CO, PEP, TPR, HR, SBP and DBP), were

measured

Materials Demographic Measures

A demographic questionnaire (see Appendix A) was used to obtain

information about the participants’ cultural background, health and health habits To ensure participants’ representativeness of their original culture, participants were asked about the ethnicities of their grandparents and parents, the participants’ place of birth and where they were raised as well as their first language Participants were also asked about their height, weight, any history of cardiovascular disease, current health status, frequency of alcohol consumption, smoking and physical exercise and family history of cardiovascular disease

Emotion Rating Scale

To measure emotion experience before and after the role play, an emotion rating scale (see Appendix B) was used This scale was used in a previous study (Mauss, et al., 2007) Participants were asked to rate the degree to which they feel

28

with respect to a series of emotions Ratings were made on a 7-point Likert scale (1=not at all, 7=a lot) The emotions involved can be divided into three components: the anger component using the terms angry, annoyed and frustrated, the positive emotion component using the terms happy, joyful and pleased and the negative emotions using the terms anxious, jealous, sad, rejected, guilty, worried, afraid and nervous The score for each component was the mean score of included items

Internal consistency of the emotion rating scale was examined by calculating the Cronbach’s α at the baseline and during the role play in the present study As

illustrated in Table 3, the internal consistency at two phases was acceptably high

Table 3 Internal consistency of emotion rating scale (α) Components Baseline (α) after role play (α)

Positive emotion 85 86 Negative emotion 84 83

Anger-in and Anger-out Subscales of STAXI-II

The Anger-in and Anger-out subscales of STAXI-II were used to assess habitual anger expression style Reliability and validity of STAXI has been examined

in Singaporean Chinese population and results indicated an acceptable high internal consistency with Cronbach’s α of 0.69 for Anger-out subscale and α of 0.68 for Anger-in subscale The test-retest reliability for Anger-out and Anger-in subscales was 0.61 and 0.81 respectively (Bishop & Quah, 1998)

Cronbach’s α was calculated for Anger-out and Anger-in in the present study Results indicated both scales had high internal consistency with α=0.80 for Anger-out, α=0.76 for Anger-in

29

Reappraisal Subscale of Adapted Emotion Regulation Questionnaire (ERQ)

To measure habitual usage of reappraisal, the Emotion Regulation

Questionnaire (ERQ), developed by Gross and colleagues, was used (Gross & John, 2003) Since the original questionnaire was to assess the general emotion regulation, items were adapted in the context of anger (see Appendix C) For example, the item

“I control my emotions by changing the way I think about the situation I’m in.” had been changed to be “I control anger by changing the way I think about the situation I’m in.” Cronbach’s α of adapted reappraisal subscale was calculated in the current study Results indicated it had high internal consistency with α=0.85

Cultural Values Measure

Difference of ethnicity has often been used to represent difference of culture

in prior research In comparisons between Caucasians and Asians, the Asian group was often sampled from the population of American Asians who may have

acculturated to Western culture (Butler, Lee, & Gross, 2007; Butler, et al., 2009) Effects of original cultural difference may have been underestimated because of the acculturation of the Asian group This issue should be taken into account because Singaporean Chinese were used in the present study and they may acculturate to different cultures due to immersion in multicultural environment For the purpose of ensuring that participant groups were culturally distinct, Schwartz’s (1992) Portrait Values Questionnaire was administrated to all the Chinese and Caucasian participants

in the present study This instrument contains 40 items tapping seven components of cultural values including conservatism, intellectual autonomy, affective autonomy, hierarchy, egalitarianism, mastery, and harmony Participants were asked to rate how much the description in each item was like them on a 6-point scale An example of

30

items tapping Conservatism could be “it is important to her to live in secure

surroundings She avoids anything that might endanger her safety” An example for Intellectual Autonomy was “Thinking up new ideas and being creative is important to her She likes to do things in her own original way” “She thinks it is important to do lots of different things in life She always looks for new things to try” was an item to measure Affective Autonomy “She always wants to be the one who makes the decisions She likes to be the leader.” was an example tapping Hierarchy An example for Mastery was “It's very important to her to show her abilities She wants people to admire what she does” And for Harmony, participants were asked to rate the items like “It is important to her to be polite to other people all the time She tries never to disturb or irritate others” Definitions of each dimension are shown in Appendix D (Schwartz, 1999) The internal consistency of each dimension of cultural values was acceptable in the present study, with Cronbach’s α=.61 for Harmony, α=.81 for Conservatism, α=.61 for Hierarchy, α=.76 for Mastery, α=.78 for Affective

Autonomy, α=.73 for Intellectual Autonomy and α=.73 for Egalitarianism Based on previous studies, these seven cultural values dimensions can effectively distinguish Asian culture and Western culture ( Schwartz, 1992)

Equipment

The Finometer Pro (Finapres Medical Systems Company, Amsterdam, The Netherlands), a non-invasive beat-to beat blood pressure monitor, was used to obtain measures of SBP, DBP, CO, HR and TPR in the present study The Finometer

records cardiovascular parameters continuously using a finger cuff and can provide accurate estimation of intra-arterial blood pressure The absolute accuracy of the Finometer can be calibrated with an upper arm cuff measurement using Return to

31

Flow (RTF) technology The technology allows blood pressure to be reconstructed on set intervals against brachial measurements using an upper arm cuff (FMS, 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) When measuring physiological parameters using the Finometer, the finger cuff, which contains photoelectronic components and a bladder for applying pressure to the finger, was placed in the middle finger between the first and second joints and the arm cuff was placed on the upper arm of the same side as the measured finger

The HIC-2000 (Bio-Impedance Technology, Inc., Chapel Hill, NC), a invasive instrument for detecting and monitoring bioelectric impedance cardiograph signals from human subjects, was used to obtain pre-ejection period (PEP) which was used to index ventricular contractility(VC) with smaller PEP values indicating greater

non-VC A 2mA current is introduced along the thorax to measure the thoracic electrical impedance changes and infer the blood volume change A set of four spot electrodes was used to introduce current and measure voltage change from which the ZCG signal can be generated Two spot electrodes were placed at the back of the neck, at the level of 7th cervical vertebra, 6cm apart from each other The 3rd electrode was placed at the end of 9th intercostals space along the mid-clavicular line And the last electrode was placed at the 9th intercostals space, 8cm apart from mid-axillary line (Sherwood et al., 1990) An external ECG cable with three plug-in leads was used to obtain ECG waves PEP was derived using the HIC-2000 and identified as time elapse between the Q point on the ECG wave and the B inflection on the ZCG wave

Role play was chosen to induce anger in this present study for two reasons First, anger is often provoked in interpersonal interactions in everyday life and role play provides the way to make participants engage in a certain social context, therefore, lab to field generalization could be improved Second, based on results of a meta-analysis, interpersonal stressor can elicit more pronounced cardiovascular reactivity than traditional laboratory stressors such as cold pressor and public speech (Suls & Wan, 2007)

In the present study, participants were presented with the scenario shown below, and were required to respond to a confederate who was introduced as another participant

Project Scene

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

33

Goal: Convince your partner to do her portion of the project so that

you can both get a good grade

Participants were asked to imagine they were actually engaged in the specific situation and to try to convince the confederate to change her behavior The

confederate delivered standard responses to the participant’s requests so as to

provoke anger in participants during the interaction There were 12 suggested

responses including “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

Emotion Regulation Instructions

After reading the scenario description participants were given instructions 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 the emotion Do not let the 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”

Procedure

Participants came for the experiment individually All participants were told

to refrain from alcoholic and caffeinated beverages and heavy exercise at least two hours prior to the study On arrival at the lab, each participant was told the study was about emotion regulation and negotiation results and that she would be engaging in a negotiation with another participant Then the participant was introduced to a female

34

confederate from the same race whom was introduced as “another participant” After that, the confederate left the room After participant signed a consent form, she was attached with the equipment

Before the experiment, the participant was asked to fill in a demographic questionnaire and the first emotion rating scale which served to measure the baseline data on emotional experience After that there was 10 min rest period during which baseline physiological indices were measured Following this was the role play interaction The participant was presented with the “project” scenario When the participant finished reading, instructions to suppress, express, or do reappraisal were presented

During the role play interaction the participant and the confederate were placed in separate rooms to avoid the confounding effect generated by confederate’s different nonverbal cues toward different participants The role play was done

through an intercom The role play was 3 min in length; after that, the participant was asked to complete the post-task emotion rating scale The completion of this

questionnaire was followed by a 10 min recovery period When recovery period ended, all recording apparatuses were detached Finally, the participant was asked to complete the Anger-in and Anger-out subscales of STAXI-II and reappraisal subscale

of ERQ to measure individual difference in anger expression, suppression and

reappraisal and the Portrait Value Questionnaire (PVQ) to assess cultural values

During the whole procedure, HR, BP, CO, TPR and PEP were measured continuously in the baseline, during the task as well as the recovery period

Main Analyses

Difference in cultural values and habitual anger regulatory strategies

To examine Hypothesis 1, habitual usage of anger expression, suppression and reappraisal were compared Anger expression score was obtained from the

Anger-out subscale of STAXI-II Anger suppression score was gained from the Anger-in subscale of STAXI-II And reappraisal score was obtained from the adapted reappraisal subscale of ERQ A series of ANOVAs were used to examine the

difference of habitual usage of each anger regulatory strategy between two ethnicity groups

Emotional experience

To examine Hypothesis 2, 3 ANOVAs were performed with cultural

background (Chinese vs Caucasian) and instructed anger regulatory strategies in

36

experiment (expression, suppression and reappraisal) as independent variables and experience of anger, positive emotions, and negative emotions as dependent variables respectively

Cardiovascular reactivity

Cardiovascular reactivity was calculated as the change value from the

baseline A MANOVA analysis with all cardiovascular parameters as dependent variable was run to examine the overall effects of cultural background, regulation condition and their interaction effect After that, a series of univariate ANOVAs were conducted for each parameter Cardiovascular reactivity was examined in patterns Based on prior research on CVR patterns, challenge pattern was defined as significant increased CO, decreased PEP and decreased or unchanged TPR and threat pattern was defined as less amount of decrease in PEP and less magnitude of increase in CO relative to challenge pattern, no change or increase in TPR (Tomaka, et al., 1993; Tomaka, et al., 1997; Tomaka et al., 1999)

37

CHAPTER 3 RESULTS

Cultural Representativeness of Ethnicity Groups

To ensure the cultural representativeness of the ethnic groups, cultural values were compared between the two groups using Schwartz’s (1992) Portrait Values Questionnaire Centred scores were used for each dimension with lower score

indicating greater endorsement Results indicated that Chinese (M= 0.11) endorsed the value of Conservatism more than the Caucasian group (M= 0.57), t (87) = -4.45, p=.001 However, Caucasians (M= -0.32) endorsed the values of Intellectual

Autonomy more than the Chinese group (M= -0.29), t (87) = 4.62, p=.001 There was

also a significant difference between Chinese and Caucasians in the endorsement of

Egalitarianism with Caucasians (M= -0.93) endorsing Egalitarianism more than Chinese (M= -0.60), t (87) = 3.16, p=.002 These findings are consistent with

considerations of collectivism-individualism as well as results of a previous cultural values survey with more than 35,000 respondents from 49 countries

cross-(Schwartz, 1999) These findings confirm the cultural representativeness of the

ethnicity groups in the present study

comparability of different groups

Emotional Experience at Baseline

38

A 2 (cultural background)*3(regulation condition) MANOVA with cultural background and regulation condition as IVs and three components of emotional

experience as DVs was performed to examine the baseline differences between two

cultural groups and across different conditions Results indicated that main effects of

cultural background and regulation condition were not significant: for main effect of

cultural background, Wilks’ λ= 94, F (3, 82)=1.62, p ns, for main effect of

regulation condition, Wilks’ λ= 89, F (6, 164)=1.70, p ns, interaction effect of them

was not significant either, Wilks’ λ= 97, F (6, 164)=.47, p ns Means and SDs of

experience of anger, positive emotions and negative emotions for each group are

Notes Numbers are means± standard deviation HR=heart rate, SBP= systolic blood

pressure, DBP=diastolic blood pressure, CO=cardiac output, PEP=pre-ejection period,

TPR= total peripheral resistance *p<.05 **p<.01