The Social Effects of Mixed Emotions: Black, White or Grey?
The Relationship between Emotional Ambivalence, Dialecticism and Character Inferences
Lindy Arends
Student number: 10170480
Supervisors: Bertjan Doosje / Michelle Yik
University of Amsterdam / Hong Kong University of Science and Technology Social Psychology
Content Abstract
1.Introduction 2. Method 3. Results
4. Conclusion and Discussion 5. Literature
Appendix 1: Results based on the Two-Factor Model Appendix 2: Standardized Test Results
Appendix 3: Complete Hong Kong Chinese Survey Appendix 4: Complete Dutch Survey
p.3 p.4 p.13 p.18 p.26 p.31 p.45 p.49 p.51 p.61
Abstract
Even though a growing body of research has begun to unravel the social effects of single emotions, little is known about the social effects of mixed emotions (i.e., emotional ambivalence). The present research aims to bridge this gap in the literature and examines which personality traits are inferred to people expressing emotional ambivalence. In addition, we build on literature on naïve dialecticism – a lay belief system prevalent in East Asian societies characterized by the acceptance of contradictions – and propose that ambivalent expressers are perceived more favourably in East Asian than Western countries. Hong Kong Chinese (N = 258) and Dutch (N = 151) students participated in this study and rated an ambivalent or non-emotional expresser on different characteristic traits. Moreover, dialectical thinking was measured. This study demonstrated that ambivalent expressers were perceived as less competent, less moral and more sociable than non-emotional expressers in both cultural samples. Interestingly, Hong Kong Chinese participants perceived ambivalent expressers as less competent and sociable than Dutch participants. This cultural difference could not be explained by differences in dialectical thinking, but is proposed to reflect cultural differences in display rules. Limitations and suggestions for future research will be discussed.
The Relationship between Emotional Ambivalence, Dialecticism and Character Inferences
The Syria policy of the Obama administration has been criticized for suffering from a “deeply ambivalent president” (Mazzetti, Worth, & Gordon, 2013). Obama described the situation in Syria as “one of those hellish problems every president faces, where the risks are endless and all the options are bad” (Mazzetti et al., 2013). In his Nobel Peace Prize
acceptance speech, he emphasized his mixed feelings on sending American troops to fight in the name of peace once again (Parsons & Hennigan, 2017).
Just like Obama, people are likely to experience “blends of emotions”. In fact, solely experiencing single emotions, such as pure happiness or pure anger, is actually quite rare (Scherer & Tannenbaum, 1986). The simultaneous experience of positive and negative emotions is called emotional ambivalence (Fong, 2006; Larsen, McGraw & Cacioppo, 2001), and is often referred to as mixed emotions (Aaker, Drolet, & Griffin, 2008). It has been found that emotional ambivalence is likely to occur in emotional complex situations, such as
graduation day and dormitory move-out day (Larsen, et al., 2001) or when witnessing “acts of humanity in the midst of terror” (Williams & Aaker, 2002).
Because the limited research conducted on emotional ambivalence mainly focused on its experience (e.g., Fong, 2006; Fong & Tiedens, 2002; Larsen et al., 2001), its social effects have yet to be explored. Interestingly, it has been suggested that emotional ambivalence might be an effective instrument of exerting social influence (Rafaeli & Sutton, 1991; Sinaceur, Adam, Van Kleef & Galinsky, 2013). For instance, police interrogators display contrasting emotions using a good-cop bad-cop strategy to intimidate suspects and evoke compliance (Rafaeli & Sutton, 1991). In addition, showing inconsistent emotions in negotiations has shown to result in concessions from the opposing party (Sinaceur et al., 2013).
In line with the idea that emotions serve social functions (Frijda & Mesquita, 1994; Keltner & Haidt, 1999), we propose that expressing emotional ambivalence brings about social effects. Moreover, building on naïve dialecticism – a lay belief system prevalent in East Asian societies characterized by the acceptance of contradictions in daily life
(Spencer-Rodgers, Peng, & Wang, 2010) – we take a cross-cultural perspective and suggest that the social effects of emotional ambivalence differ in East Asian and Western societies. More specifically, we argue that expressions of emotional ambivalence hold implications for personality perception, and are perceived more favourably in East Asian than Western societies due to differences in dialectical thinking.
1.1. Emotional Ambivalence: A Theoretical Background
Emotional experiences are usually complex and can often not be explained using single words. Instead of describing emotions1 as solely ‘sad’ or ‘happy’, people often use a combination of various emotions to communicate their state of mind more accurately (Diener & Iran-Nejad, 1986; Oatley & Duncan, 1994). Even though it might seem counter-intuitive, it is not unusual for people to simultaneously experience opposing emotions such as happiness and sadness or hope and fear (Larsen et al., 2001; Bee & Madrigal, 2013).
Neuro-psychological research has provided evidence that it is indeed possible for positive and negative emotions to occur simultaneously: positive and negative emotions have been found to be separate dimensions that correspond to different neurological substrates (Cacioppo & Gardner, 1999).
Research on emotional ambivalence primarily focused on its eliciting conditions. In addition to the aforementioned graduation day and dormitory move-out day (Larsen et al., 2001), emotional ambivalence has been found to occur while watching movies, such as “Life
1
Emotions are short-lived, affective and intense responses to relevant environmental stimuli with a clear cause (Frijda, 1986). Emotions are accompanied by physical reactions (Levenson, Ekman, & Friesen, 1990), subjective experiences (Scherer & Tannenbaum, 1986) and action tendencies (Frijda, Kuipers, & ter Schure, 1989).
is Beautiful”. Whereas prior to watching this humorous and tragic movie only 10% of the participants reported to feel a combination of happiness and sadness, this percentage
increased to 44% directly after watching the movie (Larsen et al., 2001). Similarly, people are likely to experience emotional ambivalence in organizational contexts. For instance,
employees reported mixed feelings towards their supervisors because they impose high
demands but also provide emotional support (Pratt & Doucet, 2000) and women in high-status positions reported happiness and sadness because they successfully reached their career goals but could not adhere to stereotypical gender-roles (Fong & Tiedens, 2002). Finally,
experimental studies successfully provoked feelings of emotional ambivalence using inductions such as nostalgia (Wildschut, Sedikides, Arndt, & Routledge, 2006) and “disappointing wins” and “relieving losses” (Larsen et al., 2001).
While research on emotional ambivalence is in its early stages, attitudinal
ambivalence has received considerable research attention2 (e.g., Armitage & Conner, 2000; Nordgren, van Harreveld, & van der Pligt, 2006; van Harreveld et al., 2009). Attitudinal ambivalence refers to the simultaneous positive and negative evaluation of an attitude object (Eagly & Chaiken, 1993; Jonas, Broemer, & Diehl, 2000). These evaluations can be cognitive and affective in nature, but may also arise due to discrepancies between cognitive and
affective attitudes (Lavine, Thomsen, Zanna, & Borgida, 1998) and established and newly endorsed attitudes (Petty, Tormala, Brinol, & Jarvis, 2006). Attitudinal ambivalence has been associated with topics such as abortion (Schneider et al., 2015), organ donation (Cohen, 2010), fast food (Gillebaart, Schneider, & De Ridder, 2016), physical exercise (Sparks,
Harris, & Lockwood, 2004) and minority groups (Pacilli, Mucchi-Faina, Pagliaro, Mirisola, & Alparone, 2013).
2 Ambivalence generally started to be a research interest after the realization that bipolar measures (i.e., Likert-scales) failed to distinguish between indifference and ambivalence, because in both cases respondents will report
There is a tendency in social-psychological literature to view ambivalence as being disadvantageous. For instance, both attitudinal and emotional ambivalence have been found to produce feelings of tension and conflict (Aaker et al., 2008; Fong, 2006; Larsen, McGraw, Mellers, & Cacioppo, 2004; Petty, Brinol, & Johnson, 2012; Rydell, McConnell, & Mackie, 2008; Van Harreveld et al., 2009), presumably due to the violation of the human need to be consistent in thoughts, feelings and behaviour (Festinger, 1957; Proulx, Inzlicht, & Harmon-Jones, 2012).Moreover, ambivalent attitudes have been found to be less accessible in memory (Bassili, 1996), weakly predictive of behaviour (Armitage & Conner, 2000; Conner et al., 2002) and prone to persuasion attempts (Armitage & Conner, 2000).
Yet, the cognitive consequences of ambivalence are not merely negative. Attitudinal ambivalence has been found to increase systematic information processing while making decisions (Maio, Greenland, Bernard, & Esses, 2001) and emotional ambivalence is suggested to increase creative thinking (Fong, 2006) and judgment accuracy (Rees et al., 2013).
1.2. The Social Effects of Emotional Ambivalence: Character Inferences
As research on emotional ambivalence primarily focused on the personal experiences and cognitive consequences, the social effects of emotional ambivalence have largely
remained unexamined. This is surprising, considering that emotions are not solely a private experience: emotions serve social functions (Frijda & Mesquita, 1994; Keltner & Haidt, 1999). After all, if emotions would only matter to the individual, why would they show on our faces?
The social–functional approach to emotions posits that emotions provide information to observers (Keltner & Haidt, 1999; Van Kleef, 2009). For instance, emotional expressions3 communicate one’s social motives and behavioural intentions, which in turn may influence
3Emotional expressions can be displayed in multiple ways, such as facial expressions (Keltner, Ekman, Gonzaga, & Beer, 2003), posture (Riskind, 1984), gaze (Adams & Kleck, 2003), tone of voice (Scherer, Johnstone, & Klasmeyer, 2003) and verbal expressions (Reilly & Seibert, 2003).
the observer’s cognitions, attitudes and behaviours (Ekman, 1993; Hess & Fischer, 2013). To illustrate this point, a boss expressing anger due to a missed deadline might induce fear in his employees, which causes them to work harder on the next deadline. Emotional expressions can thus be viewed as a powerful tool of social influence (Van Kleef, Van Doorn, Heerdink, & Koning, 2011).
In a similar vein, it has been suggested that emotions may play a crucial role in impression management. A growing body of research indicates that people form first
impressions based on observed emotional expressions (Hareli & Hess, 2010; Karawasa, 2001; Knutson, 1966; Sinaceur & Tiedens, 2006; Tiedens, 2001). For example, it has been found that people expressing anger are perceived to be dominant and competent (Clark, Pataki & Carver, 1996; Karawasa, 2001; Tiedens, 2001), but also less friendly and warm (Sinaceur & Tiedens, 2006). In contrast, people expressing sadness are perceived to be likeable and warm, but also submissive, weak and incompetent (Tiedens, 2001). These findings provide important information that can be useful in a variety of domains and contexts, such as negotiation
(Sinaceur & Tiedens, 2006) and leadership (Gooty, Connelly, Griffith, & Gupta, 2010; Van Kleef, Homan, Beersma, & Van Knippenberg, 2010).
However, this research does not touch upon the question how people expressing emotional ambivalence will be perceived. Emotional ambivalence is conceptualized as tension and conflict in facial expressions, posture, gaze and tone of voice (Sincoff, 1990; Rothman, 2011) and has been found to be accurately recognized by adults (Ekman, Friesan, & Ellsworth, 1972; LaPlante & Ambady, 2000). Contrary to emotional neutrality and single emotions, emotional ambivalence thus communicates a conflicted state of mind which
conveys deliberation (Rothman, 2011) and unpredictability (Bushman & Holt-Lunstad, 2009; Drolet & Morris, 2000). While there is general agreement that expressions of positive
unfavourable first impressions (Ames & Johar, 2009; Clark et al., 1996; Clark & Taraban, 1991), little is known about the personality traits attributed to people simultaneously expressing positive and negative emotions. To bridge this gap in the current literature, the current study intends to develop an understanding of the personality perception of emotionally ambivalent expressers.
Building on the stereotype content model (Fiske, Cuddy, & Glick, 2007; Leach, Ellemers, & Barreto, 2007), we propose that ambivalent expressions hold implications regarding the expresser’s perceived character in terms of competence, morality and sociability4. First, given that emotional ambivalent expressions convey deliberativeness, ambivalent expressers are likely seen as indecisive and unable to enact their intent, hereby indicating a lack of knowledge (Marsh & Rothman, 2013). Moreover, ambivalent leaders have been found to be perceived as less powerful (Rothman, 2011) and people who appear inconsistent are seen as being less credible (Goffman, 1969). Hence, ambivalent expressions are likely to reduce perceptions of competence. We therefore predict:
Hypothesis 1. People who express emotional ambivalence are perceived to be less competent than people who do not express emotions.
In addition, ambivalent expressions likely reduce perceptions of morality. Prior research suggests that the intentions of ambivalent expressers remain unclear, which makes them seem unpredictable and hesitant in interactions (Bushman & Holt-Lunstad, 2009; Drolet & Morros, 2000). This might be perceived as if the expresser is hiding something, which may undermine their trustworthiness (Mayer, Davis, & Schoorman, 1995). Moreover, ambivalence has been associated with hypocrisy in organizational contexts (Meyerson & Scully, 1995),
4 Competence refers to someone’s capability to enact their intent and is comprised of traits related to intelligence and efficiency. Morality pertains to the perceived correctness of someone’s social behaviour and consists of traits related to trustworthiness and sincerity. Last, sociability concerns the ability to connect with others and consists of traits related to likeability and kindness (Fiske et al., 2007; Leach et al., 2007).
possibly because emotional ambivalence is sometimes suspected to be an insincere impression management strategy. Hence, we hypothesize the following:
Hypothesis 2. People who express emotional ambivalence are perceived to be less moral than people who do not express emotions.
Finally, ambivalent expressions should positively affect sociability-related
perceptions. Because ambivalent expressions convey deliberation, ambivalent expressers are likely to be perceived as open to different perspectives (Rothman, 2011). Consequently, ambivalent expressers might be perceived as approachable and understanding. Indeed, research suggests that expressions of emotional ambivalence are sometimes evaluated positively. Pillaud, Cavazza and Butera (2013) found that participants who had to present themselves in a positive (compared to negative) manner were more likely to express contrasting emotions on controversial issues, presumably to appear nuanced and emphatic. Therefore, we predict the following:
Hypothesis 3. People who express emotional ambivalence are perceived to be more sociable than people who do not express emotions.
1.3. Culture and Emotional Ambivalence: The Role of Dialecticism
However, drawing on literature about the cultural difference of naïve dialecticism, we argue that the social effects of emotional ambivalence may differ culturally. Naïve
dialecticism is a lay belief system prevalent in East Asian countries characterized by the acceptance and expectation of contradictions in daily life. Dialectical thinkers view the world and events as inherently contradictory, dynamic and fundamentally interconnected (Peng & Nisbett, 1999; Spencer-Rodgers et al., 2010). This worldview is often contrasted to Western folk epistemology characterized by “linear” thinking, which aims to reconcile contradictions in daily life and views the world and events as fixed and unchanging (Festinger, 1957; Peng & Nisbett, 1999).
Accordingly, dialectical and non-dialectical thinkers might interpret ambivalent expressions in a different manner. Dialectical thinkers are likely to view emotions such as happiness and sadness as mutually dependent, coevolving and existing in a state of balance (Choi & Choi, 2002). In contrast, non-dialectical thinkers are likely to view emotions as discrete and opposing phenomena that do not co-occur (Goetz, Spencer-Rodgers, & Peng, 2008). Indeed, emotional ambivalence has been found to be more prevalent in East Asian than Western societies (Goetz et al., 2008; Schimmack, Oishi, & Diener, 2002) and it has been suggested that differences in dialectical thinking account for this cultural difference (Hui, Fok, & Bond, 2009; Lu, Hamamura, Doosje, Suzuki, & Takemura, 2016; Miyamoto, Uchida, & Ellsworth, 2010). Moreover, it has been shown that dialectical thinkers hold favourable attitudes towards ambiguous information (Williams & Aaker, 2002).
Building on these findings, we expect that people from East Asian cultures, where dialectical thinking is prevalent, perceive ambivalent expressers as more favourable than people from Western cultures, where “linear” thinking is prevalent. Therefore, we hypothesize:
Hypothesis 4. People from East Asian cultures will infer more positive character traits to ambivalent expressers than people from Western cultures;
Hypothesis 4a. People from East Asian cultures perceive ambivalent expressers as more competent than people from Western cultures;
Hypothesis 4b. People from East Asian cultures perceive ambivalent expressers as more moral than people from Western cultures;
Hypothesis 4c. People from East Asian cultures perceive ambivalent expressers as more sociable than people from Western cultures.
Finally, we expect that these cultural differences are accounted for by differences in dialectical thinking. Hence, we predict:
Hypothesis 5. Dialecticism mediates the cultural difference in character inferences between Hong Kong Chinese and Dutch participants.
1.4. Overview of the present research
In summary, the current research aims to examine (1) how emotionally ambivalent expressers are perceived in terms of competence, morality and sociability characteristics; (2) whether there are cultural differences in the way emotionally ambivalent expressers are perceived; and (3) whether dialecticism accounts for this cultural difference.
In order to do so, we manipulate emotional ambivalence using a description of an ambivalent expresser (and non-emotional expresser) and assess how people perceive them in terms of competence, morality and sociability characteristics, as well as dialectical thinking in Hong Kong (representing an East Asian culture) and the Netherlands (representing a Western culture). We predict that ambivalent expressions reduce perceptions of competence and morality, but positively affect sociability perceptions in both cultural samples. Moreover, we predict that ambivalent expressers are perceived as more competent, moral and sociable in Hong Kong relative to the Netherlands, and propose that differences in dialectical thinking account for this cultural difference.
The current research contributes to the existing literature in three ways. First, it broadens the social-functional approach to emotion by shifting focus from discrete emotions to emotional ambivalence. Second, it extends research on emotional ambivalence by
exploring its social effects instead of the individual experience. Finally, this is – to our knowledge – the first research that cross-culturally compares the social effects of emotional ambivalence. Importantly, given the fact that we live in a global economy, this research might help to better understand the social effects of emotions in different cultures. In turn, this might prevent potential cultural misunderstandings.
Method Participants
The total sample consisted of 409 students (258 Hong Kong Chinese, 151 Dutch). To ensure that all participants were representative for the Hong Kong Chinese and Dutch culture, nine participants have been excluded from further analysis5. Our final sample consisted of 252 Hong Kong Chinese students from the Hong Kong University of Science and Technology (54% females, Mage = 20.32, SD = 1.49) and 148 Dutch students from the University of
Amsterdam (74% females, Mage = 20.74, SD = 2.08). The two cultural samples did not differ significantly in age, t (208.63) = -1.81, p = .07, but they did differ significantly in gender, χ2
(1) = 15.48, p < .0016. Participants in Hong Kong completed this study in the lab and
received monetary compensation (40HKD) whereas participants in the Netherlands completed this study online and received course credits. Course credits could not be awarded in Hong Kong because the academic year had ended at the time of research conduction.
Design and procedure
A 2 (culture: Chinese versus Dutch) x 2 (emotion: emotional ambivalence versus control) between-subjects design has been employed. In Hong Kong, students signed up prior to participation. Upon arriving, participants were welcomed and took seat in front of a
computer. Once all participants arrived, they were instructed to start the research and remain seated until allowed to leave. In the Netherlands, students completed the survey online. Because previous research indicates that response sets of participants show similar patterns in web-based and laboratory-based research (Riva, Teruzzi, & Anolli, 2003), it is unlikely that the slight difference in procedure affected our results. In both cultural samples, the study was
5 Participants who were not born in Hong Kong (the Netherlands) and/or lived in Hong Kong (the Netherlands) less than 10 years were excluded.
framed to be about “first impressions and emotional expressions”. First, all participants were introduced to the study after which they signed the informed consent agreement.
Subsequently, participants were randomly assigned to the ambivalent or control condition and read information about X, who was framed to be an emotionally ambivalent or non-emotional expresser. After completing the manipulation checks, participants rated to what extent they believed different characteristic traits applied to X. Participants proceeded with the Dialectical Self Scale and the Independent and Interdependent Self-Construal Scale. At the end of the survey, participants indicated their demographic information and were given the opportunity to comment on the research. Finally, they were debriefed and thanked for participation. It took an average of 12.4 and 11.4 minutes to complete the study for Hong Kong Chinese and Dutch participants, respectively.
Measures and materials
Because bilingual participants have been found to score differently when tested in their native or second language (Bond, 1983; Ross, Xun, & Wilson, 2002), the present study was conducted in the mother tongue of the participants: traditional Chinese and Dutch. All measures have been translated using the Brislin model for back-translation (Brislin, 1970; Jones, Lee, Phillips, Zhang, & Jaceldo, 2001), unless stated otherwise.
Emotion manipulation. Participants were asked to imagine to work at a company of their choice, where they would soon be introduced to a new colleague (X), with whom they were going to work closely. The context of the introduction was chosen carefully, as it has been found that ambivalent emotions are commonly experienced in organizational settings (Fong & Tiedens, 2002; Pratt & Doucet, 2000). After, participants read a vignette adapted
from Belkin and Rothman7 (2017). In the ambivalent condition, the following vignette was introduced:
“When interacting with peers, X is known to often display mixed feelings, appearing rather torn and conflicted in most circumstances. Typically, everyone who communicates with X notices that X’s facial features, tone of voice, and gestures show conflicting feelings. Moreover, even when discussions touch upon sensitive and important issues, X typically expresses ambivalence (i.e., tension and conflict about where he/she stands).”
Participants in the control condition read the following vignette:
“When interacting with peers, X is known to show no emotions, staying neutral and non-emotional in most circumstances. Typically, everyone who communicates with him/her notices that X’s facial features, tone of voice and gestures show no emotions. Moreover, even when discussions touch upon sensitive and important issues, X typically does not express any emotions.”
Manipulation check. To ensure that the vignettes successfully manipulated the desired emotion, participants were asked to recall what kind of emotions X typically expressed. Participants could choose between positive emotions, negative emotions, mixed emotions and no emotions.
Control variable: perceived gender of X. Participants were asked to what gender they thought X belonged (coded as 1 = male; 2 = female). We decided to control for this variable because emotional neutrality, competence and morality are primarily associated with
masculinity and emotional ambivalence and sociability with femininity (Kranz, Pröbstle, & Evidis, 2017).
7
In the original vignettes, the androgynous name Pat was used. However, considering the uncommon nature of this name in Hong Kong as well as the Netherlands, we decided to change the name to ‘X’.
Character inferences. Building on the stereotype content model (Fiske et al., 2007), participants rated to what extent they believed several characteristic traits applied to X on a scale from 1 (absolutely no) to 7 (absolutely yes). The items were taken from scales used in prior research (Brambilla, Rusconi, Sacchi, & Cherubini, 2011). The competence subscale consisted of the traits intelligent, competent, skilful and inefficient and incapable (reverse coded; Chronbach’s α = .89 for Hong Kong, α = .70 for the Netherlands). The sociability subscale consisted of the traits friendly, warm, helpful and unkind and unlikeable (reverse coded; Chronbach’s α = .66 for Hong Kong, α = .76 for the Netherlands). Finally, the morality subscale included the traits sincere, righteous and respectful and untrustworthy and dishonest (reverse coded; Chronbach’s α = .68 for Hong Kong, α = .74 for the Netherlands).
Dialecticism8. Participants completed the 32-item version of the Dialectical Self Scale (DSS, Spencer-Rodgers et al., 2016). Participants rated their agreement on items such as “When I hear two sides of an argument, I often agree with both” and “I often find that my beliefs and attitudes will change under different contexts” on a scale from 1 (strongly
disagree) to 7 (strongly agree) scale. A commonly used traditional Chinese version was used (Spencer-Rodgers et al., 2016). Alpha reliabilities for the DSS were α = .80 for Hong Kong Chinese participants and α = .77 for Dutch participants.
Self-construals. Because dialecticism is more prevalent in collectivistic than in individualistic cultures (Peng & Nisbett, 1999), self-construals as proxies for cultural
orientation were included as control variables. Participants filled in the 31-item Self-Construal Scale (SCS) developed and revised by Singelis (1994, 2000). The SCS consists of two factors; independent (Chronbach’s α = .70 for Hong Kong as well as the Netherlands) and
interdependent (Chronbach’s α = .76 for Hong Kong, α = .71 for the Netherlands) self-construals. People with dominant independent self-construals define themselves in terms of
Likert-internal attributes, such as traits and abilities. In contrast, people with dominant
interdependent self-construals define themselves based on their relationship with others. It has been suggested that someone’s dominant self-construal is mostly determined by
individualistic and collectivistic cultural contexts (Markus & Kitayama, 1991). Sample items included: “I enjoy being unique and different from others in many respects” (independent self-construal) and “I will sacrifice my self-interest for the benefit of the group I am in” (interdependent self-construal). Each item was rated on a Likert scale from 1 (strongly disagree) to 7 (strongly agree). Independent and interdependent self-construals were not related in the Dutch sample (r = .09, p = .32), but they were weakly positively related in the Hong Kong Chinese sample (r = .17, p < .05). A modified traditional Chinese translation of Li and colleagues (2006) and a Dutch translation of Nezlek, Schaafsma, Safron and Krejtz (2012) was used for Hong Kong Chinese and Dutch participants respectively.
Demographics. Demographic information on age, gender (coded as 1 = male, 2 = female) and cultural background (country of birth for themselves and their parents’) was asked.
Results
Manipulation check. To ensure that our manipulation was successful, we checked whether participants in the ambivalent condition correctly recalled that X typically expressed emotional ambivalence, and participants in the neutral condition correctly recalled that X typically did not express any emotions. A total of 40 Hong Kong Chinese and 17 Dutch participants failed the manipulation check and have been excluded from our analysis. Our final sample consisted of 212 Hong Kong Chinese and 131 Dutch participants.
Factor analysis. Next, we conducted a principal component analysis (PCA) with varimax rotation on the total dataset, to identify the underlying structure of the characteristic traits. Our data did not confirm the expected three-factor solution (competence, sociability and morality) as found in previous research (Brambilla et al., 2011; Leach et al., 2007). Instead, we found four factors yielding an eigenvalue greater than 1, accounting for a total of 65.71% of the total variance. However, no patterns in line with the three theoretical factors were found and repeating this procedure for the two language samples yielded different results. Because the stereotype content model originally consisted of the scales competence and warmth (Fiske et al., 2007), we checked whether a two-factor solution would offer a solution. We found two factors in 10 items, using a cut-off point of <.5 for factor loadings and <.4 for communalities (Field, 2009). This solution largely replicated the factors competence and sociability of the proposed three-factor model. Nevertheless, we decided to stick to the three theoretical variables, considering their acceptable to high reliabilities (ranging from Chronbach’s α = .66 to α = .89) and its strong theoretical background (Brambilla et al., 2011; Leach et al., 2007). More information and results based on the two-factor solution can be found in Appendix 1.
Descriptive statistics. Descriptive test results of competence, morality and sociability for
homoscedasticity have been met, but some competence, morality and sociability data violated assumptions of normality.9 However, an analysis of Variance (ANOVA) remains robust when assumptions are broken (Glass, Peckham, & Sanders, 1972) and violation of its assumptions merely affects the false positivity rate (Lix, Keselman, & Keselman, 1996). Accordingly, we did not adjust our data to achieve normal distribution.
Table 1. Means (M) and standard deviations (SD) on competence, morality and sociability in the neutral and ambivalent condition, for Hong Kong (HK) and the Netherlands (NL).
Control variable: perceived gender of X. Building on previous research (Belkin & Rothman, 2017; Kranz et al., 2017), we checked whether the perceived gender of X was statistically related to condition, competence, morality and sociability. In line with previous research, perceived gender of X differed significantly in the two conditions, χ2(1) = 28.69, p < .001, indicating that X was more often considered to be female in the ambivalent (41%) than in the neutral (15%) condition. Moreover, male X’s were perceived as more competent (M = 4.25, SD = .97) than female X’s (M = 3.87, SD = .87), t (341) = 3.28, p < .001, 95%
Confidence Interval (CI) = [0.15; 0.60]. Male X’s were also perceived as more moral (M = 4.09, SD = .77) than female X’s (M = 3.90, SD = .82), t (341) = 2.00, p < .05, 95% CI
9 Test results for the Shapiro-Wilk test for competence data; D(65) = .95, p <.05 (NL) in the neutral condition and D(103) = .96, p < .01 in the ambivalent condition (HK). For morality data; D(109) = .93, p <.001 (HK) in the neutral condition and D(103) = .97 p < .05 (HK) in the ambivalent condition. Finally, for sociability data;
D(109) = .97, p <.05 (HK) and D(65) = .95, p <.05 (NL) in the neutral condition.
HK NL
1. Competence Neutral condition M = 4.87 SD = .81 M = 4.37 SD = .76 Ambivalent condition M = 3.48 SD = .73 M = 3.76 SD = .73 2. Morality Neutral condition M = 4.19
SD = .68 M = 4.22 SD = .79 Ambivalent condition M = 3.93 SD = .77 M = 3.75 SD = .87 3. Sociability Neutral condition M = 3.40
SD = .65 M = 3.23 SD = .75 Ambivalent condition M = 3.78 SD = .64 M = 4.06 SD = .70
[.00; .37]. No such differences occurred on sociability data, t (341) = -1.33, p = .19, 95% CI [-.29; .06]. Hence, we controlled perceived gender of X as a potential influential variable, but all results remained unchanged after adding this variable in the model.
Dialecticism and self-construals. Consistent with the existing literature (Markus & Kitayama, 1991; Spencer-Rodgers et al., 2010), Hong Kong Chinese participants reported significantly higher scores on the DSS (M = 4.29, SD = .48) than Dutch participants (M = 3.76, SD = .49), t (341) = 9.73, p < .001, 95% CI [.42; .63], representing a large Cohen’s effect size value (d = 1.07). Furthermore, results on the self-construal scale also replicated prior research (Harb & Smith, 2008; Hui et al., 2009). Hong Kong Chinese participants scored higher on interdependent self-construal (M = 4.76, SD = .58) than Dutch participants (M = 4.50, SD = .58), t (341) = 4.02, p < .001, 95% CI [.13; .39], indicating a small to medium-sized effect (d = .45). In addition, Dutch participants scored higher on independent self-construal (M = 4.84, SD = .57) than Hong Kong Chinese participants (M = 4.58, SD = .56), t (341) = -4.11, p <.001, 95% CI [-.38; -.13], also representing a small to medium-sized effect (d = .46). Thus, participants in Hong Kong were more dialectical and interdependent, and less independent compared to participants in the Netherlands.
Main analysis. We predicted that the ambivalent expresser is perceived as less competent (Hypothesis 1), less moral (Hypothesis 2) and more sociable (Hypothesis 3) than the neutral expresser in both cultural samples. Moreover, we expected that Hong Kong Chinese
participants are overall more positive about the ambivalent expresser than Dutch participants (Hypothesis 4). Specifically, we predicted that Hong Kong Chinese participants perceive the ambivalent expresser as more competent (Hypothesis 4a), more moral (Hypothesis 4b) and more sociable (Hypothesis 4c) than Dutch participants. On competence, sociability and morality data three separate factorial ANOVA’s were run with condition (ambivalent versus neutral) and culture (Hong Kong versus the Netherlands) as between-subject variables. To
account for cultural differences in response tendencies (Van de Vijver & Leung, 1997), standardized Z-scores were computed for raw scores in the competence, sociability and morality dataset (see Appendix 2). The standardized scores replicated the raw scores reported below and our findings are therefore not affected by effects of cultural response bias.
Competence. Results of the two-way ANOVA yielded a significant main effect of condition on competence. As expected, participants in both cultural samples rated the
ambivalent expresser as less competent (M = 3.59, SD = .74) than the neutral expresser (M = 4.68, SD = .83), F (1, 339) = 138.15, p <.001, 95% CI [-.1.16; -.83], partial eta-squared (ηp2) = .29. No main effect of culture on competence was found; F (1,339) = 1.72, p = .19, 95% CI [-.05; .28], ηp2=.01. Furthermore, a significant interaction between condition and culture was found, F (1, 339) = 20.65, p <.001, ηp2 = .06, indicating that the main effect of condition on competence was stronger for Hong Kong Chinese than Dutch participants (see Figure 1). An analysis of simple main effects revealed that, contrary to our expectations, Hong Kong Chinese participants evaluated the ambivalent expresser as less competent than Dutch participants, F (1,339) = 5.19, p <.05, 95% CI [-.51; -.04], ηp2 = .02. Moreover, Hong Kong Chinese participants evaluated the neutral expresser as more competent than Dutch
participants, F (1, 338) = 17.26, p <.001, 95% CI [.26; .73], ηp2
= .05. Thus, Hypothesis 1 has been confirmed, but we did not find support for Hypothesis 4a.
Figure 1. Interaction effect (mean ± SEM) between condition and culture on competence data (on a scale from 1 to 7).
Morality. In line with our expectations, participants in both cultural samples rated the ambivalent expresser as less moral (M = 3.86, SD = .81) than the neutral expresser (M = 4.20, SD = .72), F (1, 339) = 17.91, p < .001, ηp2 = .05, 95% CI [-.53; -.19]. No main effect of culture on morality, F (1, 339) = 0.78, p = .38, ηp2 = .00, 95% CI [-.09; -.24] and no interaction effect between condition and culture on morality, F (1, 339) = 1.38, p = .24, ηp2
= .00 was found. Moreover, analysis of simple main effects revealed that there were no cultural differences in the ambivalent (p = .15) and neutral condition (p = .84). Thus, we found support for Hypothesis 2, but not for Hypothesis 4b.
Sociability. As expected, participants in both cultural samples rated the ambivalent expresser as more sociable (M = 3.88, SD = .68) than the neutral expresser (M = 3.34, SD = .69), F (1, 339) = 63.82, p < .001, ηp2 = .16, 95% CI [.45; .75]. No main effect of culture on sociability was found, F (1, 339) = 0.60, p = .44, 95% CI [-.21; .09], ηp2=.00. However, a significant interaction between condition and culture on sociability was found, F (1, 339) = 8.75, p < .01, ηp2= .03, indicating that the effect of condition on sociability was stronger for Dutch than Hong Kong Chinese participants (see Figure 2). An analysis of simple main effects revealed that, contrary to our expectations, Hong Kong Chinese participants evaluated
1 2 3 4 5 Neutral Ambivalent Condition Competence Hongkong The Netherlands
95% CI [-.49; -.07], ηp2 = .02. No cultural differences were found for the neutral expresser (p = .12). Thus, we found support Hypothesis 3, but not for Hypothesis 4c.
Figure 2. Interaction effect (mean ± SEM) between condition and culture on sociability data (on a scale from 1 to 7).
Correlation analysis. Subsequently, we further examined the relationship between emotional ambivalence, dialecticism and character inferences. To do so, correlations were assessed to check whether competence, morality, sociability, dialecticism, culture,
independence, interdependence, perceived gender of X, gender and age were statistically related in the ambivalent condition (see Table 2). Competence, morality and sociability were positively related, which replicates prior research and represents a halo-effect (Fiske et al., 2007). Moreover, in line with our findings thus far, we found a negative connection between culture and competence (r = -.18, p < .05) as well as between culture and morality (r = -.20, p < .01), suggesting that the ambivalent expresser was perceived to be less competent and moral in Hong Kong compared to the Netherlands. Moreover, there was a strong significant
correlation between culture and dialecticism (r = .47, p <.01), indicating that Hong Kong Chinese participants scored higher on dialecticism than Dutch participants, replicating our earlier findings. In a similar vein, dialecticism correlated negatively with independent self-construals and positively with interdependent self-self-construals, replicating prior research (Hui
1 2 3 4 5 Neutral Ambivalent Condition Sociability Hongkong The Netherlands
et al., 2009). Finally, there seemed to be a positive connection between dialecticism and competence (r = .13, p = .08). Additionally, we examined the correlations between
dialecticism and competence in both cultural samples; in Hong Kong, we found a significant positive correlation (r = .28, p < .001) and in the Netherlands, we found a marginally
significant positive correlation (r = .21, p = .09). This suggests that participants scoring high on dialecticism perceived the ambivalent expresser as (somewhat) more competent than participants scoring low on dialecticism.
Table 2. Correlations between competence, morality, sociability, culture, dialecticism, independence and interdependence, gender and age in the ambivalent condition.
1 2 3 4 5 6 7 8 9 1. Competence - 2. Morality .19* - 3. Sociability .18* .54** - 4. Culture -.18* .11 -.20** - 5. Dialecticism .13 .05 -.10 .47** - 6. Independence -.06 -.03 .01 -.17* -.36** - 7. Interdependence -.02 .03 -.06 .26** .30** .16* - 8.Perceived gender X .05 -.03 .03 -.00 .02 -.00 .09 - 9. Gender .07 -.04 .03 -.16* -.20** .10 -.05 -.16* - 10. Age -.05 .02 -.07 -.02 -.02 .09 -.05 -.09 -.19*
Note: Gender and perceived gender of X were coded as 1 for male and 2 for female, culture was coded as 1 for the Netherlands and 2 for Hong Kong. N = 103 for Hong Kong and N = 68 for the Netherlands. *Correlation is significant, * p < .05 ** p < .01 (two-tailed).
Mediational effect of dialecticism in the ambivalent condition. We did not find support for our hypothesis that the ambivalent expresser is perceived more favourably by Hong Kong Chinese than Dutch participants (Hypothesis 4). In fact, the opposite pattern was found on competence and sociability data: Hong Kong Chinese participants perceived ambivalent expressers to be less competent and sociable than Dutch participants. Following our theoretical model, the final series of analyses were conducted to determine whether
dialecticism accounts for this cultural difference (Hypothesis 5). To do so, we conducted a mediation analysis on competence and sociability data using bootstrapping (Preacher &
competence data10. A bootstrapping procedure with 5,000 resamples and a bias corrected confidence interval indicated that the confidence interval for the proposed indirect (mediated) effect had a lower limit [LL] of .07 and an upper limit [UL] of .36. Given the fact that the confidence interval did not contain zero, mediation is suggested. Yet, to establish full mediation the direct effect of culture on competence should become non-significant when dialecticism is controlled for. Interestingly, the opposite pattern was found: adding
dialecticism to the model actually increased the effect of culture on competence from β = -.27 to β = -.4711. These findings demonstrate a suppressor effect: the negative effect of culture on competence is stronger when dialecticism is accounted for. This indicates that (Hong Kong Chinese) participants scoring high on dialecticism evaluated the ambivalent expresser to be more competent, therefore weakening the negative effect of culture on competence. The path coefficients are reported in Figure 1.
Figure 1. Standardized regression coefficients for the relationship between culture and competence as mediated by dialecticism in the ambivalent condition. The standardized regression coefficient between culture and competence, controlling for dialecticism, is in parentheses. **p< .01.
10Assumptions of multicollinearity, linearity and homoscedasticity and normality of errors have been met.
11Running the same mediational analysis with independent- as well as interdependent self-construals as potential influential variables did not yield different results.
Dialecticism Competence Culture 1 = NL, 2 = HK β = .39** β = -.27 ** (β = -.47**) β =.52**
Conclusion and Discussion
The current research intended to develop an understanding of the social effects of emotional ambivalence. Drawing on the social-functional approach to emotions (Keltner & Haidt, 1999; Van Kleef, 2009), we demonstrated the important role of emotional expressions in personality perception. More specifically, we demonstrated the undermining effect of expressing emotional ambivalence on competence and morality perceptions and the beneficial effect of expressing emotional ambivalence on sociability perceptions. In line with our
expectations, we found that people who express emotional ambivalence are perceived as less competent, less moral and more sociable than people who do not express emotions. These results hold in an East Asian country (Hong Kong) as well as a Western country (the Netherlands).
Moreover, we took a cross-cultural perspective and proposed that expressions of emotional ambivalence are perceived more favourably in East Asian countries (Hong Kong) than in Western countries (the Netherlands), due to cultural differences in dialectical thinking (Peng & Nisbett, 1999). Even though the Hong Kong Chinese participants were indeed more dialectical than the Dutch participants, our results indicated the opposite pattern: ambivalent expressers were perceived as less competent and less sociable in Hong Kong, and no cultural differences emerged for morality perceptions. In other words: expressions of emotional ambivalence were generally perceived as less favourable in Hong Kong relative to the Netherlands.
Finally, while exploring the role of dialecticism in the cultural differences of emotional ambivalence on personality perception, we found that the cultural difference of emotional ambivalence on competence perceptions was suppressed by dialecticism. This means that emotionally ambivalent expressers were perceived as less competent in Hong Kong relative to the Netherlands, and even less competent when dialecticism was accounted
for. Put differently: the undermining effect of expressing emotional ambivalence on competence perceptions in Hong Kong became even more profound when controlled for dialecticism.
This finding provides modest support for our expectation that dialecticism positively affects the social effects of emotional ambivalence. However, this suppression effect only occurred on competence perceptions. Suggestively, this might be because dialectical and non-dialectical thinkers may value competence perceptions in a different manner. For example, expressions of emotional ambivalence convey deliberation (Rothman, 2011) as well as indecisiveness (Marsh & Rothman, 2013). It might be that dialectical thinkers value deliberation as a predictor for competence, whereas non-dialectical thinkers value decisiveness. It could therefore be argued that dialectical thinkers perceive ambivalent expressers to be considerate and thoughtful, which they ought to be important predictors for competence. Future studies should examine the strength of this potential explanation.
The question remains: how can we explain that expressions of emotional ambivalence are perceived less favourably in Hong Kong relative to the Netherlands? A possible
explanation is that these results reflect cultural differences in display rules. It has been suggested that people in collectivistic societies are less likely to express emotions in order to maintain social harmony. As a result, people are relatively low in self-disclosure and more likely to supress their emotions (Matsumoto, Yoo, & Fontaine, 2008). In contrast, people in individualistic societies are known to express their emotions more freely (Butler, Lee, & Gross, 2007; Matsumoto et al., 2008). Indeed, the Dutch culture is known to be a prototypical individualistic culture (Fischer, Manstead, & Rodriquez Mosquera, 1999; Hofstede, 1980) and Hong Kong is classified as a collectivistic culture (Oyserman, Coon, & Kemmelmeier, 2002). Therefore, it may be that expressions of emotional ambivalence are considered to be
inappropriate in Hong Kong, and perceived less favourably as a consequence. Again, this explanation needs further studies to determine its value.
There are several limitations of this study that suggest potential areas for future research. First, emotional ambivalence is often conceptualized to as the co-occurrence of positive and negative emotions (e.g., Fong, 2006; Larsen et al., 2001). However, it could be argued that emotional ambivalence may also result from experiencing contrasting emotions of the same valence (Pratt & Doucet, 2000). For example, fear and sadness are both considered to be negative emotions but the former is high in arousal and the latter is low in arousal (Russell & Barrett, 1999). Similarly, emotions might differ in action tendencies: anger triggers approach tendencies whereas embarrassment triggers avoidance tendencies (Elliot, Eder, & Harmon-Jones. 2013). It is thus worth considering a broader definition of emotional ambivalence and investigate the social effects of contrasting emotions in terms of arousal level and action tendencies as well.
Second, our study did not capture contextual information. Participants rated an unknown person based on a short description regarding their typical emotional expressions. However, in real life people form first impressions based on a variety of factors, such as facial expressions (Keltner et al., 2003), posture (Riskind, 1984) and tone of voice (Scherer et al., 2003). Similarly, the social effects of emotional ambivalence may depend on the social context. For instance, expressing emotional ambivalence might be appropriate and therefore perceived as favourable during emotionally complex situations (Larsen et al., 2001), but inappropriate and thus unfavourable during first encounters and business meetings (Pratt & Doucet, 2000). To enhance our understanding of the social effects of emotional ambivalence, further research should examine whether (1) the current findings replicate when emotional ambivalence is manipulated in a more direct manner, for instance using video images; and (2) whether the current findings generalize in different contexts.
A third suggestion for further research emphasizes the importance of comparing the social effects of emotional ambivalence to positive and negative emotions. The present research used non-emotional expressions as a comparison standard for a combination of positive and negative emotional expressions. However, it could be argued that this
comparison standard does not provide a neutral baseline. For example, it has been found that people who do not express emotions are perceived as cold and inhuman (McCord, Joseph, & Grijalva, 2014). It is therefore important to understand how expressions of emotional
ambivalence relate to more consistent positive and negative emotions.
Moreover, while we examined the character traits inferred to ambivalent expressers, the exact mechanisms underlying these character inferences remain open to speculation. For example, it is possible that competence, morality and sociability traits are not equally desirable in East Asian and Western countries. Indeed, it has been suggested that some competence related traits are less desirable in East Asian countries (Littrell, 2002). Future studies should therefore directly measure the mechanisms underlying the inferred
characteristics traits, such as their social desirability.
In addition, this research is among the first to link dialectical thinking to social perception domains. Because our results seem to suggest that dialectical thinking might indeed affect the social effects of emotional ambivalence, we encourage other researchers to follow suit. Nonetheless, it is important to note that the common measure for dialectical thinking (the DSS) has specifically been developed in the domain of self-perception (Spencer-Rodgers et al., 2016). Because it has been suggested that dialectical thinking is best measured in specific domains (Spencer-Rodgers et al., 2016), it might be worthwhile to adapt measures of dialectical thinking to social perception domains as well.
Finally, we acknowledge that our findings require further investigation across different cultures and populations. Hong Kong has been under British colonial rule, which means that
the society is relatively westernized (Zhao & Chen, 2008). It would therefore be interesting to see whether our results replicate in more traditional countries, such as Mainland China and Japan. In addition, we must be cautious when generalizing our results to the general population: students represent a homogenous and biased sample of the overall population (Mook, 2001). Still, the use of university students allowed for greater comparability and the fact that we found cross-cultural differences is promising, because it could be argued that students are generally less traditional than their countries as a whole (Heine, 2012).
In conclusion, the present research is the first to shed light on the social effects of emotional ambivalence. We took a cross-cultural perspective and demonstrated the important role of emotional expressions in personality perception. Importantly, our findings deepen our understanding of the cultural differences in the social effects of emotional expressions. In a world that becomes increasingly interconnected, this understanding is crucial to prevent potential cultural misunderstandings. Still, the question remains – are the social effects of emotional ambivalence black (negative), white (positive) or rather grey (negative and
positive)? Our results indicate that the social effects of emotional ambivalence are not solely positive or negative, however, more research is needed to get a clearer picture of this grey area.
Literature
Aaker, J., Drolet, A., & Griffin, D. (2008). Recalling mixed emotions. Journal of Consumer Research, 35(2), 268-278. doi: 10.1086/588570
Adams Jr, R. B., & Kleck, R. E. (2003). Perceived gaze direction and the processing of facial displays of emotion. Psychological Science, 14(6), 644-647. doi: 10.1046/j.0956-7976.2003.psci_1479.x
Ames, D. R., & Johar, G. V. (2009). I'll know what you're like when I see how you feel: How and when affective displays influence behavior-based impressions. Psychological Science, 20(5), 586-593. doi: 10.1111/j.1467-9280.2009.02330.x
Armitage, C. J., & Conner, M. (2000). Attitudinal ambivalence: A test of three key hypotheses. Personality and Social Psychology Bulletin, 26(11), 1421-1432. doi: 10.1177/0146167200263009
Bassili, J. N. (1996). Meta-judgmental versus operative indexes of psychological attributes: The case of measures of attitude strength. Journal of Personality and Social Psychology, 71(4), 637-653. doi: 10.1037/0022-3514.71.4.637
Bee, C. C., & Madrigal, R. (2013). Consumer uncertainty: the influence of anticipatory emotions on ambivalence, attitudes, and intentions. Journal of Consumer Behaviour, 12(5), 370-381. doi: 10.1002/cb.1435
Belkin, L. Y., & Rothman, N. B. (2017). Do I Trust You? Depends on What You Feel: Interpersonal Effects of Emotions on Initial Trust at Zero‐Acquaintance. Negotiation and Conflict Management Research, 10(1), 3-27. doi: 10.1111/ncmr.12088
Bond, M. H. (1983). How language variation affects inter-cultural differentiation of values by Hong Kong bilinguals. Journal of Language and Social Psychology, 2(1), 57-66. doi: 10.1177/0261927X8300200104
primary role of morality (vs. sociability and competence) in information gathering.
European Journal of Social Psychology, 41(2), 135-143. doi: 10.1002/ejsp.744
Brislin, R. W. (1970). Back-translation for cross-cultural research. Journal of Cross-Cultural Psychology, 1(3), 185–216. doi: 10.1177/135910457000100301
Butler, E. A., Lee, T. L., & Gross, J. J. (2007). Emotion regulation and culture: Are the social consequences of emotion suppression culture-specific?. Emotion, 7(1), 30-
48. doi:10.1037/1528-3542.7.1.30
Bushman, B. B., & Holt-Lunstad, J. (2009). Understanding social relationship maintenance among friends: Why we don't end those frustrating friendships. Journal of Social and Clinical Psychology, 28(6), 749-778. doi: 10.1521/jscp.2009.28.6.749
Cacioppo, J. T., & Gardner, W. L. (1999). Emotion. Annual Review of Psychology, 50(1), 191-214. doi: 10.1146/annurev.psych.50.1.191
Chen, F. F., Jing, Y., & Lee, J. M. (2014). The looks of a leader: Competent and trustworthy, but not dominant. Journal of Experimental Social Psychology, 51(2014), 27-33. doi: 10.1016/j.jesp.2013.10.008
Choi, I., & Choi, Y. (2002). Culture and self-concept flexibility. Personality and Social Psychology Bulletin, 28(11), 1508-1517. doi: 10.1177/014616702237578. Clark, M. S., Pataki, S. P., & Carver, V. H. (1996). Some thoughts and findings on self-
presentation of emotions in relationships. In G. O. Fletcher & J.
Fitness (Eds.), Knowledge structures in close relationships: A social psychological approach (pp. 247–274). New York, NY: Lawrence Erlbaum Associates.
Clark, M. S., & Taraban, C. (1991). Reactions to and willingness to express emotion in communal and exchange relationships. Journal of Experimental Social Psychology, 27(4), 324-336. doi: 10.1016/0022-1031(91)90029-6.
Individuals' Decisions to Become Organ Donors. Health Communication, 25(8), 758- 769. doi:10.1080/10410236.2010.521923.
Colquitt, J. A., Scott, B. A., & LePine, J. A. (2007). Trust, trustworthiness, and trust
propensity: a meta-analytic test of their unique relationships with risk taking and job performance. Journal of Applied Psychology, 92(4), 909-927. doi: 10.1037/0021-9010.92.4.909
Conner, M., Sparks, P., Povey, R., James, R., Shepherd, R., & Armitage, C. J. (2002). Moderator effects of attitudinal ambivalence on attitude–behaviour
relationships. European Journal of Social Psychology, 32(5), 705-718. doi:
10.1002/ejsp.117
Diener, E., & Iran-Nejad, A. (1986). The relationship in experience between various types of affect. Journal of Personality and Social Psychology, 50(5), 1031-1038. doi:
10.1037/0022-3514.50.5.1031
Drolet, A. L., & Morris, M. W. (2000). Rapport in conflict resolution: Accounting for how face-to-face contact fosters mutual cooperation in mixed-motive conflicts. Journal of
Experimental Social Psychology, 36(1), 26-50. doi: 10.1006/jesp.1999.1395
Dueck, C. (2015). The Obama doctrine: American grand strategy today. New York, NY: Oxford University Press.
Eagly, A. H., & Chaiken, S. (1998). Attitude structure and function. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.). Handbook of Social Psychology (pp. 269-322). Boston, MA: McGraw-Hill.
Eisenberg, N., Fabes, R.A., Miller, P.A., Fultz, J., Shell, R., Mathy, R.M., & Reno, R.R. (1989). Relation of sympathy and distress to prosocial behavior: A multimethod study. Journal of Personality and Social Psychology, 57(1), 55-66. doi:
Ekman, P. (1993). Facial expression and emotion. American Psychologist, 48(4), 384-392. doi: 10.1037/0003-066X.48.4.384
Elliot, A. J., Eder, A. B., & Harmon-Jones, E. (2013). Approach–avoidance motivation and emotion: Convergence and divergence. Emotion Review, 5(3), 308-311.doi:
10.1177/1754073913477517
Field, A. (2009). Discovering statistics using SPSS (4th ed.). London: Sage.
Festinger, L. (1957). A theory of cognitive dissonance. New York, NY: Row, Peterson. Fischer, A. H., Manstead, A. S. R., & Rodriquez Mosquera, P. M. (1999). The role of honour
related vs. individualistic values in conceptualizing pride, shame, and anger: Spanish and Dutch cultural prototypes. Cognition and Emotion, 13, 149–179. doi:
10.1080/026999399379311
Fiske, S. T., Cuddy, A. J. C., & Glick, P. (2007). Universal dimensions of social cognition: Warmth and competence. Trends in Cognitive Sciences, 11(2), 77–83. doi:
10.1016/j.tics.2006.11.005
Fong, C. T. (2006). The effects of emotional ambivalence on creativity. Academy of
Management Journal, 49(5), 1016-1030. doi: 10.5465/AMJ.2006.22798182
Fong, C. T., & Tiedens, L. Z. (2002). Dueling experiences and dual ambivalences: Emotional and motivational ambivalence of women in high status positions. Motivation and Emotion, 26(1), 105-121. doi: 10.1023/A:1015198209285
Frijda, N. H. (1986). The emotions: Studies in emotion and social interaction. New York, NY: Cambridge University Press.
Frijda, N. H., Kuipers, P., & ter Schure, E. (1989). Relations among emotion, appraisal, and emotional action readiness. Journal of Personality and Social Psychology, 57(2), 212-228. doi: 10.1037/0022-3514.57.2.212
S. Kitayama & H. R. Markus (Eds.), Emotion and Culture (pp. 51–88). Washington, DC: American Psychological Association.
Gillebaart, M., Schneider, I. K., & De Ridder, D. T. (2016). Effects of trait self‐control on response conflict about healthy and unhealthy food. Journal of Personality, 84(6), 789-798. doi: 10.1111/jopy.12219
Glass, G. V., Peckham, P. D., & Sanders, J. R. (1972). Consequences of failure to meet assumptions underlying the fixed effects analyses of variance and covariance. Review of Educational Research, 42(3), 237-288. doi: 10.3102/00346543042003237
Goffman, E. (1969). The insanity of place. Psychiatry, 32(4), 357-388. doi: 10.1080/00332747.1969.11023600
Gooty, J., Connelly, S., Griffith, J., & Gupta, A. (2010). Leadership, affect and emotions: A state of the science review. The Leadership Quarterly, 21(6), 979-1004. doi:
10.1016/j.leaqua.2010.10.005
Hamamura, T., Heine, S. J., & Paulhus, D. L. (2008). Cultural differences in response styles: The role of dialectical thinking. Personality and Individual Differences, 44(4), 932-942. doi: 10.1016/j.paid.2007.10.034
Harb, C., & Smith, P. B. (2008). Self-construals across cultures: Beyond
independence—interdependence. Journal of Cross-Cultural Psychology, 39(2), 178-197. doi: 10.1177/0022022107313861.
Heine, S. J. (2012). Cultural Psychology (2nd ed.). New York, NY: Norton.
Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Beverly Hills, CA: Sage.
Hui, C. M., Fok, H. K., & Bond, M. H. (2009). Who feels more ambivalence? Linking dialectical thinking to mixed emotions. Personality and Individual Differences, 46(4), 493-498. doi: 10.1016/j.paid.2008.11.022
Jonas, K., Broemer, P., & Diehl, M. (2000). Attitudinal ambivalence. European Review of
Social Psychology, 11(1), 35-74. doi: 10.1080/14792779943000125
Jones, P. S., Lee, J. W., Phillips, L. R., Zhang, X. E., & Jaceldo, K. B. (2001). An adaptation of Brislin’s translation model for cross-cultural research. Nursing Research, 50(5), 300-304. doi: 10.1097/00006199-200109000-00008
Keltner, D., Ekman, P., Gonzaga, G. C., & Beer, J. (2003). Facial expression of emotion. In R. J. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Series in affective science. Handbook of affective sciences (pp. 415-432). New York, NY: Oxford University Press.
Keltner, D., & Haidt, J. (1999). Social functions of emotions at four levels of
analysis. Cognition & Emotion, 13(5), 505-521. doi: 10.1080/026999399379168 Klopfer, F. J., & Madden, T. M. (1980). The middlemost choice on attitude items:
Ambivalence, neutrality, or uncertainty?. Personality and Social Psychology
Bulletin, 6(1), 97-101. doi: 10.1177/014616728061014
LaPlante, D., & Ambady, N. (2000). Multiple messages: Facial recognition advantage for compound expressions. Journal of Nonverbal Behavior, 24(3), 211-224.
10.1023/A:1006641104653
Larsen, J. T., McGraw, A. P., & Cacioppo, J. T. (2001). Can people feel happy and sad at the same time? Journal of Personality and Social Psychology, 81(4), 684-696. doi: 10.1037/0022-3514.81.4.684.
Larsen, J. T., McGraw, A. P., Mellers, B. A., & Cacioppo, J. T. (2004). The agony of victory and thrill of defeat: Mixed emotional reactions to disappointing wins and relieving losses. Psychological Science, 15(5), 325-330. doi: 10.1111/j.0956-7976.2004.00677.x.
the determination of attitudes and behavior: The moderating role of affective-cognitive ambivalence. Journal of Experimental Social Psychology, 34(4), 398-421. doi:
10.1006/jesp.1998.1357.
Leach, C. W., Ellemers, N., & Barreto, M. (2007). Group virtue: the importance of morality (vs. competence and sociability) in the positive evaluation of in-groups. Journal of
Personality and Social Psychology, 93(2), 234-249. doi: 10.1037/0022-3514.93.2.234.
Levenson, R. W., Ekman, P., & Friesen, W. V. (1990). Voluntary facial action generates emotion‐specific autonomic nervous system activity. Psychophysiology, 27(4), 363-384. doi: 10.1111/j.1469-8986.1990.tb02330.x.
Li, H. Z., Zhang, Z., Bhatt, G., & Yum, Y. O. (2006). Rethinking culture and self-construal: China as a middle land. The Journal of Social Psychology, 146(5), 591–610. doi: 10.3200/SOCP.146.5.591-610.
Littrell, R. F. (2002). Desirable leadership behaviours of multi-cultural managers in China. Journal of Management Development, 21(1), 5-74. doi:
10.1108/02621710210413190
Lix, L. M., Keselman, J. C., & Keselman, H. J. (1996). Consequences of assumption violations revisited: A quantitative review of alternatives to the one-way analysis of variance F test. Review of Educational Research, 66(4), 579-619. doi:
10.2307/1170654
McCord, M. A., Joseph, D. L., & Grijalva, E. (2014). Blinded by the light: The dark side of traditionally desirable personality traits. Industrial and Organizational
Psychology, 7(1), 130-137. doi: 10.1111/iops.12121
Maio, G. R., Greenland, K., Bernard, M., & Esses, V. M. (2001). Effects of intergroup ambivalence on information processing: The role of physiological arousal. Group
Processes & Intergroup Relations, 4(4), 355-372. doi:10.1177/1368430201004004005.
Matsumoto, D., Yoo, S. H., & Fontaine, J. (2008). Mapping expressive differences around the world: The relationship between emotional display rules and individualism versus collectivism. Journal of Cross-Cultural Psychology, 39(1), 55-74. doi:
10.1177/0022022107311854
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98(2), 224-253. doi: 10.1037/0033 295X.98.2.224
Marsh, J. K., & Rothman, N. B. (2013). The ambivalence of expert categorizers. In M. Knauff, M. Pauen, N. Sebanz & I. Wachsmuth (Eds.), Proceedings of the 35th annual conference of the cognitive science society (pp. 984–989). Austin, TX: Cognitive Science Society.
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709-734. doi: 10.5465/AMR.1995.9508080335
Mazzetti, M., Worth, R. F., & Gordon, M. R. (2013, October 22). Obama's Uncertain Path Amid Syria Bloodshed. Retrieved March 19, 2017, from:
http://www.nytimes.com/2013/10/23/world/middleeast/obamas-uncertain-path-amid syria-bloodshed.html
Meyerson, D. E., & Scully, M. A. (1995). Crossroads tempered radicalism and the politics of ambivalence and change. Organization Science, 6(5), 585-600. doi:
10.1287/orsc.6.5.585
Mook, D. G. (2001). Psychological research: The ideas behind the methods. London: Norton. Miyamoto, Y., Uchida, Y., & Ellsworth, P. C. (2010). Culture and mixed emotions: Co
occurrence of positive and negative emotions in Japan and the United States. Emotion, 10(3), 404-415. doi: 10.1037/a0018430
Nezlek, J. B., Schaafsma, J., Safron, M., & Krejtz, I. (2012). Self-construal and the intra-and interethnic social interactions of ethnic minorities. Journal of Cross-Cultural
Psychology, 43(4), 614-627. doi: 10.1177/0022022111399647
Nordgren, L. F., Van Harreveld, F., & Van Der Pligt, J. (2006). Ambivalence, discomfort, and motivated information processing. Journal of Experimental Social Psychology, 42(2), 252-258. doi: 10.1016/j.jesp.2005.04.004
Oatley, K., & Duncan, E. (1994). The experience of emotions in everyday life. Cognition &
Emotion, 8(4), 369-381. doi: 10.1080/02699939408408947
Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128(1), 3-72. doi: 10.1037/0033-2909.128.1.3
Pacilli, M. G., Mucchi-Faina, A., Pagliaro, S., Mirisola, A., & Alparone, F. R. (2013). When affective (but not cognitive) ambivalence predicts discrimination toward a minority group. The Journal of Social Psychology, 153(1), 10-24. doi:
10.1080/00224545.2012.701251
Parsons, C., & Hennigan, W. J. (2017, January 13). President Obama, who hoped to sow peace, instead led the nation in war. Retrieved March 19, 2017, from:
http://www.latimes.com/projects/la-na-pol-obama-at-war/
Peng, K., & Nisbett, R. E. (1999). Culture, dialectics, and reasoning about contradiction. American Psychologist, 54(9), 741-754. doi: 10.1037/0003-066X.54.9.741
Petty, R. E., Briñol, P., & Johnson, I. (2012). Implicit ambivalence. In B. Grawonski, & F. Strack (Eds)., Cognitive consistency: A fundamental principle in social cognition (pp.
