Abstract
Previous research has shown that enthusiastic responses to personal good news are associated with positive relationship qualities, whereas more muted or ambivalent responses tend to have negative effects on relationships. This process, called capitalization, has been studied almost exclusively in Western cultures. The present research examined capitalization in three East Asian cultural groups (Mainland China, Hong Kong, and Taiwan), as well as in the United States. Whereas enthusiastic responses were associated with positive relationship outcomes in all groups, muted and critical responses showed significant moderation by culture. In three North American samples, these responses were negatively associated with relationship qualities. However, muted and critical responses revealed strong and positive correlations with relationship outcomes in Mainland China, non-significant associations in Hong Kong, and significant negative (but still weaker than in North America) associations in Taiwan. These findings point to the importance of examining relationship processes in the context of culture.
One of the ways in which romantic partners build resilience in their relationships involves celebrating each other’s good fortune when something good happens to one of them (Gottman, 1999). Langston (1994) proposed this process as one means of capitalizing on positive events: sharing good news so that its benefits are extended. In the years since Langston’s seminal paper, dozens of studies have documented the personal and relational benefits that accrue when partners respond positively to the sharing of good news (see Gable & Reis, 2010; Peters et al., 2018, for reviews). This research has resulted in a growing literature, such that capitalization is now described as a core construct in many relationship science textbooks (e.g., Miller, 2019), as a self-help strategy for improving personal happiness (Lyubomirsky, 2008), and as an intervention for distressed couples (e.g., Conoley et al., 2015; Woods et al., 2015).
As an example, imagine that a young graduate student, Jason, has just had a paper accepted for publication at a top journal. By sharing this news with his partner, Josie, Jason provides an opportunity for Josie to respond enthusiastically, adding to his joy, or alternatively, to dampen his pleasure by expressing disinterest. The term capitalization refers to interactions in which one partner’s sharing of personal good news 1 (a capitalization attempt) leads, in the most favorable case, to a supportive response. Existing research shows that sharing good news with a partner who responds enthusiastically is associated both with intrapersonal benefits to the capitalizer, such as increased positive affect, subjective well-being, and self-esteem (e.g., Demir & Davidson, 2013; Gable et al., 2004; Lambert et al., 2012; Monfort et al., 2014; Otto et al., 2015; Reis et al., 2010), and with interpersonal benefits, such as higher relationship satisfaction, commitment, and a sense of acceptance, gratitude, and couple identity (e.g., Donato et al., 2014; Gable et al., 2004; Hershenberg et al., 2016; Kashdan et al., 2013; Lambert et al., 2012; Pagani et al., 2015; Pagani et al., 2020). These benefits even apply to the partners who are responding enthusiastically (e.g., Hicks & Diamond, 2008; Monfort et al., 2014; Reis et al., 2014; Smith, 2012), perhaps in part because the recipients of capitalization support are likely to respond in kind when their partners share a positive event (Kaczmarek et al., 2021). Thus, the weight of evidence supports the conclusion that supportive responses to the disclosure of positive personal events are robustly associated with relational well-being.
The consistency of existing evidence notwithstanding, this body of work has an important limitation: Nearly all studies have been conducted with Western participants (a few exceptions are noted below). There is good reason to believe that the capitalization process may not work in the same way in East Asian cultures. As we elaborate below, capitalization, when effective in the Western context, tends to involve explicit communication, yet support processes among East Asians are more likely to involve implicit, or indirect, communication (Kim et al., 2008). Also, in East Asian cultures, displays of personal accomplishments may have negative consequences for social relationships (Choi et al., 2019) and may even be considered hubristic (e.g., Miyamoto & Ma, 2011), so that capitalization attempts may not evoke a favorable response.
The present research was designed to fill this gap. In particular, we focused on perceived responses to capitalization attempts, inasmuch as prior research has shown that these perceptions are the proximate influence on personal and interpersonal outcomes. In three Western and three East Asian samples, we examined two inter-related questions about capitalization in romantic relationships. The first question explored differences in the prevalence of different perceived responses to capitalization attempts. The second question investigated the possibility of cultural variations in the relational implications of different responses to capitalization attempts; that is, do the distinct types of capitalization responses have different associations with relationship quality in Western and East Asian cultures?
Distinctive types of response to capitalization attempts
Gable et al. (2004) used Rusbult et al.’s (1982) conceptual scheme for categorizing responses to a partner’s destructive behavior (i.e., accommodation) to identify four distinct types of responses to a partner’s capitalization attempts. These categories vary along two dimensions: whether the partner’s response is perceived to be constructive or destructive and whether it is perceived to be active or passive. Active-constructive responses involve explicit interest and engagement with the positive event, such as by expressing enthusiasm, pride, and happiness for the partner’s good fortune. Examples include elaborating on the positive implications of the event and showing genuine excitement about it. Passive-constructive responses are also positive but subdued, such as when the responder quietly and tersely conveys approval but says little about the event’s implications for the person or their relationship. An active-destructive response is one in which the responder appears to undermine the discloser’s pleasure, such as by focusing on the event’s negative implications, pointing out its downside, or minimizing its value or significance. Passive-destructive responses involve minimal acknowledgment of the event, such as by displaying little or no interest in the account or by quickly changing the subject.
Existing research supports this categorical scheme and the distinctive correlates of these four types of responses (Gable et al., 2004; Pagani et al., 2013), as measured by the Perceived Responses to Capitalization Attempts scale (PRCA; Gable et al., 2004). Evidence consistently associates the perception of a partner’s active-constructive response with diverse positive personal and relationship outcomes, whereas the perception of an active-destructive or passive-destructive response tends to be associated with a variety of negative personal and relationship outcomes (see Gable & Reis, 2010; Peters et al., 2018, for reviews). On the other hand, evidence regarding the correlates of perceived passive-constructive responses is mixed. Whereas some studies show negative associations between these responses and relationship outcomes (e.g., Gable et al., 2004, Study 2; Lambert et al., 2012), others show positive (Bemis, 2008; Carson, 2015) or no significant associations (Gable et al., 2004, Study 3; Greenwell, 2016; Pagani et al., 2013; Pagani et al., 2020). This inconsistency may reflect the fact that in a Western context, passive responses, even if nominally positive in valence, sometimes fail to convey the level of pride and engagement that would-be capitalizers seek.
Capitalization in the East Asian context
The idea of capitalization originated in a Western context, and to date it has been studied primarily in Western cultures, so that the relevance of this process for non-Western cultures is unclear. Several lines of theorizing lead us to propose that capitalization processes may unfold differently in East Asian cultures than in the United States. For one, East Asian cultures traditionally value emotional self-control and humility to a greater extent than Western cultures do (Chen et al., 2009; Smith et al., 2019), which would discourage individuals from sharing their personal good fortune or accomplishments with others, for fear of appearing shallow, immodest, emotionally immature, or unconcerned with group harmony (Chen et al., 2009; Wang et al., 2010). Moreover, affect valuation theory (Tsai, 2007, 2017) proposes that compared to European-Americans, East Asians value high-arousal positive emotions less and low-arousal positive emotions more. Several studies have shown that these values shape preferences for activities and social interactions (e.g., Sims et al., 2018; Tsai et al., 2006), which suggests that East Asians should demonstrate less of the active-constructive response and more of the passive-constructive response than European-Americans.
Another line of reasoning comes from research on social-support seeking, which has shown that when negative or stressful events occur, East Asians are less likely than European Americans to explicitly ask members of their social network for support. This lesser likelihood has been attributed to the relatively greater desire among East Asians to avoid burdening others, to preclude the possible embarrassment that asking for help might entail, and to the expectation that significant others will be aware of one’s need for assistance without being asked (Kim et al., 2008; Taylor et al., 2004). Similar logic has been used to explain why explicit support-seeking should be less common in cultures that place greater emphasis on interdependent self-construals, as opposed to conceptualizing the self in independent terms (Markus & Kitayama, 2010).
The reluctance among East Asians to explicitly ask for support may have particular relevance to capitalization following positive events. A tendency toward interdependent self-construals typically leads East Asians to focus on the thoughts and feelings of others in their social group and to emphasize blending in (Markus & Kitayama, 2010). Calling attention to personal accomplishments may embarrass others in one’s group by implying superiority and thereby interfering with the group’s harmony. In support of this idea, Choi et al. (2019) found that Koreans were less likely than European Americans to share their positive events with others. Moreover, East Asians are less likely than European Americans to display positive emotions such as pride and instead are more likely to dampen their expression (e.g., Butler et al., 2009; Eid & Diener, 2001; Matsumoto & Kupperbusch, 2001; Miyamoto & Ma, 2011).
An additional way to conceptualize the difference between East Asian and Western responses to capitalization attempts is based on the relatively greater reliance on dialectical thinking in East Asian cultures (Hamamura et al., 2008; Spencer-Rodgers et al., 2018). Dialectical thinking, or a tolerance of contradictory beliefs, suggests a preference for moderation and modesty, acknowledging that certain life occurrences can simultaneously have good and bad aspects. In the context of capitalization, dialectical thinking would involve activation of a more mixed set of informational and emotional goals, and a corresponding moderation of emotional responses. If so, we would expect passive-constructive and active-destructive responding to be relatively more characteristic of satisfying relationships in East Asian cultures than in North America. In contrast, because Western culture tends to place greater value on uniformly enthusiastic and affectively positive responses, sharing of positive events in satisfying relationships should be more closely associated only with the active-constructive response.
Only a few studies have examined capitalization outside of Western cultures. As noted above, Choi et al. (2019) found that Korean university students were less likely than a comparison sample of European Americans to share their positive events with others, primarily because of the perceived negative social consequences of doing so. Demir et al., (2013) found that active-constructive responses to capitalization attempts were positively associated with personal happiness in a sample of Turkish university students, although the correlation was somewhat weaker than in a comparison sample of American college students. While replicating the four-factor structure of the PRCA instrument in a large sample of Chinese college students, Guo et al., (2018) found that, like Westerners, active-constructive were positively related to emotional well-being and passive-destructive responses were negatively correlated with emotional well-being. However, unlike Westerners, passive-constructive and active-destructive responses were unrelated to emotional well-being. Finally, in a pair of unpublished studies, relative to European Americans, Asians and Asian-Americans reported that their partners were more likely to use passive-constructive and active-destructive responding, and that these responses were unrelated to their relationship intimacy and satisfaction (Hamilton et al., 2015; Ip & Lee, 2015). 2
In the research to be described, we used multiple examples of each cultural group (Americans, East Asians) to foster greater confidence in our findings. Cross-cultural studies that rely on a single exemplar of each group might reflect unique characteristics of those groups, rather than the cultural constructs under investigation (Boer et al., 2018; Milfont & Klein, 2018). That is, much like the assumption of outgroup homogeneity can bias lay observers, the assumption of cultural homogeneity can lead researchers to erroneous generalizations (Oishi, 2004). Because culture subgroups may differ from one another in ways that might substantially affect processes under investigation (Leung & Cohen, 2011), drawing valid conclusions about a cultural difference requires sampling more than a single example of each. In other words, any differences observed between one American and one East Asian sample might be due to unique factors about those groups rather than the cultures they are intended to represent. By studying three examples of each culture, we minimize the possibility of this misattribution. This strategy has the added advantage of allowing us to identify subgroup differences that may be the focus of further research (Cohen, 2007).
The present research
The goal of the present research was to provide a more comprehensive comparison of capitalization processes in European-American and East Asian cultures. In order to ensure that our findings would not be limited to a single exemplar of each culture, we conducted our study in three American settings (a Northeastern university, a Southern university, and in a national sample of adults) and three East Asian settings (crowd-sourced samples in Mainland China and Taiwan, and a university sample in Hong Kong). We included adult, non-college participants in order to be better able to generalize results to the culture at large (Fischer & Poortinga, 2018). Furthermore, as described above, we studied three examples of each cultural group to enhance confidence that the differences we observed reflect culture-general processes and not the uniqueness of particular subgroups.
The first aim of this research was to examine in each group the relative prevalence of the four perceived responses to capitalization attempts. Our main prediction was that the passive-constructive and active-destructive response types would be more common in the East Asian than in the Western samples (Hypothesis 1). This prediction was based on two principles (detailed above). First, East Asians more commonly seek and receive support through implicit means, which seems most nearly representative of the passive-constructive response. Second, because individuals from East Asian cultures are more likely to dampen displays of positive affect (such as pride, which is particularly relevant to capitalization), we expected that the recipients of a capitalization attempt would be more likely to enact either of these two types of response so as to discourage future displays of positive affect. We did not expect differences regarding active-constructive and passive-destructive responses.
For exploratory purposes, we examined differences among the three United States samples, particularly to determine whether individuals in longer-established relationships are more likely to report their partners using any of the capitalization responses. Also, we compared the Northeastern and Southern samples of university students. Past research has shown that individuals in honor-based cultures, such as in the southern United States, are more likely to emphasize paying respect to the accomplishments (and failures) of the self and others (Leung & Cohen, 2011). We also expected that there would be differences among the three Chinese samples. Past research has found differences among these three groups in social-psychological processes such as affect and interpersonal conflict (e.g., Yeh & Chen, 2004; Yik, 2010), which likely reflects at least in part increasing Westernization in self-construal and value systems in Hong Kong and Taiwan (Bond & Cheung, 1983; Chan & Lee, 1995; Triandis et al., 1990; Yeh et al., 2013). If so, we would expect a mixed pattern of results in the Hong Kong and Taiwan samples, partially resembling the Mainland China pattern and partially resembling the North American pattern.
The second aim of this research was to examine associations between perceived partner responses to capitalization attempts and romantic relationship outcomes. We assessed relationship outcomes in three domains: relationship satisfaction, intimacy, and perceived partner responsiveness. Although usually correlated in romantic relationships, these three outcomes provide somewhat different perspectives on the features of successful relationships and therefore may help to establish convergent validity in our findings. Relationship satisfaction is commonly considered as an indication of successful relationships. Intimacy is relevant because it indexes partners’ sense of closeness and emotional connection (Reis & Shaver, 1988). Perceived partner responsiveness gauges the extent to which individuals feel understood and appreciated by their partners, as well as whether these partners respond to their needs (Reis et al., 2004). Although perceived partner responsiveness is typically evaluated in the context of negative events and stressors, research has shown that perceived responsiveness following disclosures of positive events may be a stronger predictor of the long-term health of relationships (Gable et al., 2006).
For the three U.S. samples, we predicted a pattern of correlations that would replicate prior studies, namely that active-constructive responses would be positively correlated with relationship outcomes, whereas passive-constructive, active-destructive, and passive-destructive responses would be negatively associated with relationship outcomes (Hypothesis 2). For the Chinese samples, we predicted a similar result for active-constructive responses, namely a positive correlation with relationship outcomes (Hypothesis 3a). However, as discussed earlier, because passive-constructive and active-destructive responses are more nearly consistent with the idea of implicit support and dampening expressions of positive affect, we expected that these two strategies would be positively associated with relationship outcomes (Hypotheses 3b and 3c, respectively). We also expected that passive-destructive responses would be negatively correlated with relationship outcomes in the three Chinese samples, primarily because this type of response is indicative of distance and disengagement between relationship partners (Hypothesis 3d). As mentioned above, we anticipated that there might be differences in these associations among the three Chinese samples and will examine them for exploratory purposes.
Method
Participants
We determined the minimum number of participants in each of the six groups we sampled by specifying a medium effect size, alpha = .05, and power = .90, which indicated 109 participants per group (G*Power; Erdfelder et al., 1996). However, we left the survey website open for the remainder of the semester, which resulted in a total of 4522 participants. The minimum requirement for participation was being in a committed romantic relationship of at least 3 months duration. Three samples were recruited in the United States. Sample 1 comprised 199 university students at a northeastern university, recruited from the Psychology Department participant pool. Sample 2 included 153 university students from a southern university, recruited from the Psychology Department participant pool. The third sample comprised 889 adults who volunteered for this study on ResearchMatch, a national pool of adults who have registered interest in research participation. The remaining three samples were collected in East Asia. The fourth sample, consisting of 2315 adults, was recruited in mainland China with an online survey link distributed via Tencent QQ and WeChat groups, the top two popular local social messaging and media smartphone applications. Sample 5 was composed of 458 adults in Taiwan, who participated through an online survey link posted on Dcard, a locally popular social networking website. The sixth sample consisted of 508 Hong Kong college students recruited from the Psychology Department participant pool and campus mass mail and flyers.
Descriptive Statistics for the Six Subsamples.
Note. Age and relationship length are expressed in years. For race, other includes Native Americans and Native Pacific Islanders. For gender, participants were only given the options “male” and “female.” In significance tests, df vary from (5, 2680) to (5, 2701) due to missing data. Sample sizes are Mainland China = 886–915, Hong Kong = 313–315, Taiwan = 369, Northeast U.S. = 180, Southern U.S. = 117, Research Match = 811–814. Additional descriptive statistics are reported in Table S1 (supplemental material).
Procedure and materials
Participants were recruited for a study of romantic relationships. Announcements were sent through the recruitment source and linked participants to an information sheet describing the study, followed by the survey itself. The survey was administered on Qualtrics (for the U.S. and Hong Kong samples), Tencent Survey (for Mainland China), and Google Sheet (for the Taiwan sample).
The survey included the following measures, listed in the order they appeared in the survey. Internal consistency coefficients (Cronbach’s alpha) for all measures broken down by sample are presented in Tables S2-S7 (supplemental online material).
Perceived responses to capitalization attempts
We used the 12-item measure developed and validated by Gable et al., (2004). This measure has four subscales, one each for active-constructive, passive-constructive, active-destructive, and passive-destructive responding. Participants were asked to imagine how their romantic partner would react if they came home with good news. A sample item for the active-constructive subscale is, “My partner usually reacts to my good fortune enthusiastically”; for the passive-constructive subscale, “My partner says little, but I know he/she is happy for me”; for the active-destructive subscale, “He/she points out the downsides of the good event”; and for the passive-destructive subscale, “Sometimes I get the impression that he/she doesn’t care much.”
Relationship satisfaction
We used the 16-item version of the Couples Satisfaction Inventory, developed and validated by Funk and Rogge (2007). Sample items are, “In general, how often do you think that things between you and your partner are going well?” and “My relationship with my partner makes me happy.”
Perceived partner responsiveness
This construct was assessed with the 12-item perceived partner responsiveness measure, developed and validated by Reis and colleagues (Reis et al., 2018). Sample items are “My partner is responsive to my needs” and “My partner understands me.”
Intimacy
We used 23 items from the Personal Assessment of Intimacy in Relationships scale (Schaefer & Olson, 1981). Sample items are “I share very few of my partners’ interests” (reverse-scored) and “My partner helps me clarify my thoughts.”
Demographic questions
Following the primary measures, we asked participants demographic questions about their and their partner’s gender, relationship status, race, ethnicity, and age.
Attention checks
To screen out inattentive responders, two attention check items were embedded among the survey questions (Maniaci & Rogge, 2014). The first attention check asked participants “If you are paying attention, please select 1.” The second attention check asked, “To show that you are careful while answering, please select “Disagree” for this item.”
Translation and back-translation protocol
All measures were translated to Chinese by a senior graduate student in Psychology, and then back-translated by another senior graduate student in Psychology. Next, all authors discussed the meaning and the accuracy of the back-translated items and revised inappropriate ones accordingly, until complete agreement was reached for all items. Finally, the translated measures were administered to a pilot group of 10 Chinese college students, who reported no difficulties understanding the meaning of the items. Appendix A in the supplemental online material includes a full copy of the English and Chinese versions of the survey.
Results
Data analytic framework and preliminary analyses
To test our hypotheses, we constructed a set of five planned orthogonal contrasts to represent membership in the 6 participant groups. By entering these contrasts simultaneously in regression analyses, our analyses controlled the family-wise Type I error rate (Cohen et al., 2003). Contrast 1 compared the United States (U.S.) and East Asian samples, testing Hypothesis 1. Contrast 2 compared the U.S. adult sample (Research Match) and the two groups of U.S. university students. Contrast 3 contrasted the U.S. northeastern and southern university samples. Contrast 4 compared participants from Mainland China to those in Hong Kong and Taiwan. The final contrast, Contrast 5, compared participants from Hong Kong and Taiwan.
In preliminary analyses, we examined whether gender interacted significantly with the five planned orthogonal contrasts in predicting any of the seven main measures (i.e., relationship satisfaction, intimacy, perceived responsiveness, and the 4 capitalization scores). Out of these 35 effects, only one was significant. 3 For parsimony, we did not include gender interactions in the main model tests reported below.
Because we predicted differences in how the measures would correlate across the samples, we do not include a table of correlations for the full study. Instead, Table S2-S7 in the supplemental online materials presents these correlations separately for each of the six study subsamples. 4
Tests of the hypothesis and exploratory questions
Results of Significance Tests for the Five Planned Contrasts on Perceived Responses to Capitalization Attempts.
Note. B coefficients are unstandardized. Standard errors are in parentheses. Statistically significant results are bolded. Df for the B coefficients range from (1, 2702) to (1, 2703). Df for the overall model range from (5, 2702) to (5, 2703).
Means and Standard Deviations for Main Study Measures.
Notes. Standard deviations are in parentheses. Sample sizes are Mainland = 915, Hong Kong = 312–315, Taiwan = 369, Northeast U.S. = 180, Southern U.S. = 117, Research Match = 807–813. Ns vary due to missing data.
For the next contrast, comparing U.S. older adults and university students, only active-constructive responses yielded a significant result, with active-constructive responses being more common among university students than older adults. Furthermore, in contrast 3, we found that passive-constructive, active-destructive, and passive-destructive responding were more common in southern than in northeastern U.S. university students. This result indicates that southern U.S. students perceived more common usage of these less constructive responses than northeastern U.S. students did.
The contrast comparing participants from mainland China to participants from Hong Kong and Taiwan yielded significant results for all four measures. Participants from mainland China endorsed all four responses more strongly, with the largest difference occurring for active-destructive responses. Finally, only the active-destructive measure demonstrated a significant difference for the final contrast, with this response being perceived as more prevalent in the Taiwanese than in the Hong Kong sample.
Does the association between capitalization and relationship outcomes vary across groups?
To examine this question, we performed a series of hierarchical regression analyses. At the first step, we entered the five planned orthogonal contrasts as a set. Next, at the second step, we entered one of the capitalization measures (centered). At the third step, we entered as a set five product terms representing interactions between the capitalization measure and the five contrasts. If this latter set provided a significant increase in the variance accounted for, we went ahead and examined whichever individual terms were significant. (This procedure, a Fisher protected test, controls for the family-wise Type I error rate; Cohen et al., 2003.) Although we analyzed each relationship outcome separately, to establish convergent validity in our findings, we focus primarily on variables that yielded similar results across the three measures. The capitalization measures were examined separately because we expected a different pattern of results for each of them.
Group differences in relationship outcomes
Results of Significance Testing for Contrasts and Active-Constructive Responses.
Notes. Statistically significant results (p < .05) are bolded. AC: Active- constructive. Sample sizes are Mainland = 915, Hong Kong = 312–315, Taiwan = 369, Northeast U.S. = 180, Southern U.S. = 117, Research Match = 807–813. Ns vary due to missing data.
Active-constructive responses
The bottom section of Table 4 reports results when the active-constructive scale was added to the analysis described above, along with its interaction with each of the five contrasts. As expected, overall, higher scores on active-constructive responding were strongly associated with more positive relationship outcomes across all samples, consistent with Hypotheses 2 and 3a.
More informatively, the set of five interaction terms (i.e., active-constructive × the five planned contrasts) was also significant for all three relationship outcomes. To identify these effects, we used simple-slope tests following the procedure outlined by Aiken and West (1991). From these analyses, the most consistent result involved contrast 4, such that that active-constructive responding was more strongly correlated with satisfaction, intimacy, and responsiveness in the Mainland China sample (Bs = 9.397, 0.404, and 0.815, Fs(1, 2687–2695) = 283.34, 304.60, and 400.04, respectively) than in the Hong Kong (Bs = 6.910, 0.262, and 0.483, Fs(1, 2687–2695) = 56.71, 48.39, and 52.79, respectively) and in the Taiwan (Bs = 7.963, 0.324, and 0.577 Fs(1, 2687–2695) = 120.35, 115.86, and 118.47, respectively) samples. Additionally, there were significant interactions for contrast 2, such that the active-constructive scale was more strongly associated with intimacy and responsiveness in the older-adult Research Match sample (Bs = 0.429 and 0.841, Fs(1, 2694/2695) = 492.80 and 610.76, respectively) than in the samples of university students from the Northeast (Bs = 0.290 and 0.512, Fs(1, 2694/2695) = 38.72 and 38.79, respectively) and Southern (Bs = 0.341 and 0.739, Fs(1, 2694/2695) = 25.79 and 38.88, respectively) U.S. regions.
In summary, active-constructive responding related positively to relationship outcomes in all groups. Contrast 1, comparing the U.S. and East Asian samples, did not yield significant interactions with active-constructive responding.
Passive-constructive responses
Results of Significance Testing for Perceived Passive-Constructive, Active-Destructive and Passive-Destructive Responses to Capitalization Attempts.
Note. Statistically significant results (p < .05) are bolded. PC: Passive-constructive. AD: Active-destructive. PD: Passive-destructive.
Overall, higher scores on passive-constructive responding were associated with less positive relationship outcomes. As predicted, these results were qualified by several significant interactions. Simple slopes tests to decompose the interaction of contrast 4 with passive-constructive responding revealed that passive-constructive responses were positively correlated with relationship satisfaction, intimacy, and responsiveness in the Mainland China sample (Bs = 4.859, 0.170, and 0.416, Fs(1, 2686–2694) = 62.00, 45.25, and 81.98, respectively), consistent with Hypothesis 3b. However, these correlations were non-significant in the Hong Kong sample (Bs = 0.408, −0.051, and 0.059, respectively, all Fs < 1.27) and negative in the Taiwan sample (Bs = −2.554, −0.099, and −0.121, Fs(1, 2686–2694) = 6.99, 6.24, and 2.81, respectively). Likewise, as predicted in Hypothesis 2, these associations were negative and significant in all three U.S. samples, which did not differ significantly among one another: Research Match (Bs = −4.919, −0.252, and −0.453, Fs(1, 2686–2694) = 97.88, 153.35, and 148.36, respectively), Northeastern (Bs = −5.431, −0.230, and −0.328, Fs(1, 2686–2694) = 23.72, 25.22, and 15.62, respectively), and Southern (Bs = −5.709, −0.286, and −0.352, Fs(1, 2686–2694) = 19.81, 29.46, and 13.55, respectively) samples.
In summary, passive-constructive responding was positively associated with relationship outcomes in Mainland China, uncorrelated with relationship outcomes in Hong Kong, and negatively correlated with relationship outcomes in Taiwan and the United States.
Active-destructive responses
Results for significance tests involving active-destructive responses are shown in the middle section of Table 5. As expected, overall, higher scores on active-destructive responding were significantly and negatively associated with all three relationship outcome measures.
As with passive-constructive responding, these results were qualified by several significant interactions. Simple slopes tests to decompose the interaction of contrast 4 with active-destructive responding revealed that active-destructive responses were positively correlated with relationship satisfaction, intimacy, and responsiveness in the Mainland China sample (Bs = 2.556, 0.094, and 0.294, Fs(1, 2687–2694) = 18.67, 15.10, and 44.76, respectively), as predicted in Hypothesis 3c. These correlations were non-significant in the Hong Kong sample (Bs = 0.845, −0.009, and 0.012, respectively, all Fs(1,2687–2694) < 0.878) and Taiwan samples (Bs = 0.294, 0.022, and 0.010, respectively, all Fs(1, 2687–2694) < 0.710). Furthermore, these associations were all significant and negative in the three U.S. samples, consistent with Hypothesis 2, though the associations were significantly stronger in the older-adult Research Match sample (Bs = −8.189, −0.393, and −0.679, Fs(1, 2687–2694) = 226.36, 317.14, and 283.58, respectively) than in the Northeastern (Bs = −6.712, −0.293, and −0.463, Fs(1, 2687–2694) = 30.32, 34.82, and 26.24, respectively) or Southern (Bs = −4.701, −0.280, and −0.357, Fs(1, 2687–2694) = 9.58, 20.51, and 10.04, respectively) samples. The latter two groups did not differ significantly.
In summary, active-destructive responding was positively associated with relationship outcomes in Mainland China, uncorrelated with relationship outcomes in Hong Kong and Taiwan, and negatively correlated with relationship outcomes in the United States (more strongly so among older adults than university students).
Passive-destructive responses
Results for significance tests involving passive-destructive responses are shown in the bottom section of Table 5. As expected, overall, higher scores on active-destructive responding were significantly and negatively associated with all three relationship outcome measures.
Passive-destructive responses were negatively associated with relationship outcomes in all six samples, as predicted by Hypotheses 2 and 3d, although the associations were significantly weaker in the three East Asian than in the three U.S. samples. These effects were smallest in the Mainland China (Bs = −7.213, −0.368, and −0.542, Fs(1, 2687–2694) = 237.42, 396.28, and 252.49, respectively) and the Hong Kong (Bs = −7.038, −0.306, and −0.544, Fs(1, 2687–2694) = 68.55, 85.54, and 79.22, respectively) samples, and significantly more negative in the Taiwan sample (Bs = −10.547, −0.446, and −0.759, Fs(1, 2687–2694) = 210.14, 240.37, and 204.60, respectively). In the U.S. samples, the associations were more strongly negative in the older-adult Research Match sample (Bs = −12.639, −0.552, and −1.050, Fs(1, 2687–2694) = 865.02, 1066.19, and 1134.25, respectively) than in the Northeastern (Bs = −11.474, −0.446, and −0.746, Fs(1, 2687–2694) = 101.91, 98.52, and 81.02, respectively) or Southern (Bs = −11.800, −0.483, and −0.858, Fs(1, 2687–2694) = 80.85, 86.50, and 80.31, respectively) samples, the latter two of which did not differ significantly from each other.
In summary, although passive-destructive responding was negatively associated with relationship outcomes in all six samples, the magnitude of these associations was greatest among older Americans, next most in the American university and Taiwan samples, and weakest (though still significant) in the Hong Kong and Mainland China samples.
Discussion
We begin with a brief summary of our major findings. With regard to mean differences in reports of the four response styles to capitalization attempts, there was no overall difference in active-constructive responding between East Asian and North American samples. However, the three other forms of perceived responding—passive-constructive, active-destructive, and passive-destructive—were more common in the East Asian than in the North American samples.
We included three measures of relationship outcomes (relationship satisfaction, intimacy, perceived responsiveness), and in almost all cases, they yielded similar findings, so for convenience we refer to these measures generically, as favorable relationship outcomes. Active-constructive responding was positively associated with favorable relationship outcomes in all samples. This association was significantly stronger in Mainland China than in Hong Kong or Taiwan, and among older adults than college students in North America. For both passive-constructive and active-destructive responding, we found a significant positive association with favorable relationship outcomes in Mainland China, a non-significant association in Hong Kong, and a significant negative association in Taiwan. These associations were also significantly negative in the North American samples. Passive-destructive responding predicted significantly more negative relationship outcomes in all subsamples, with the largest (i.e., most negative) associations occurring in the North American and Taiwanese samples.
Theoretical implications
Noteworthy in these findings is the positive association in East Asian samples between favorable relationship outcomes and two forms of capitalization responses—passive-constructive and active-destructive—that have consistently shown negative associations in Western samples. The former result, consistent with our hypothesis, can be explained by affect valuation theory, which indicates that East Asians place greater value on low-arousal positive affects than European-Americans do (Tsai, 2007; 2017). Both results fit with the idea that East Asian cultures emphasize emotional self-control while discouraging calling attention to oneself (Eid & Diener, 2001; Markus & Kitayama, 2010; Miyamoto & Ma, 2011). In other words, when individuals recount their personal accomplishments to relationship partners, their response may be positive but muted or may even point out the event’s downside, presumably to remind the individual about the importance of blending in and not calling attention to oneself. Our results show that in the East Asian context, this can happen without impairing relationship quality. In contrast, among North Americans, muted and downside-highlighting responses were consistently and strongly linked to more problematic relationships. This finding may reflect relatively greater levels of modesty in East Asian cultures (Chen et al., 2009). It is also consistent with the relatively greater prevalence of dialectical thinking in East Asian cultures (Hamamura et al., 2008; Spencer-Rodgers et al., 2018). Perhaps their more dialectical perspective allows East Asian capitalizers and their partners to be simultaneously aware of, and comfortable with, the idea that all good events have potential drawbacks.
This finding was not consistent across the three East Asian samples that we studied, however. Passive-constructive and active-destructive responding were not significantly correlated with relationship outcomes in Hong Kong, and were even negatively related to relationship outcomes in Taiwan, as they were in the North American samples (though the magnitude of these effects was still appreciably weaker than in the three North American samples). A plausible interpretation is that the latter two groups generally have had greater exposure to Western ideas, and consequently may have adopted a more mixed set of values, affording greater flexibility in domain-specific knowledge structures (Hong et al., 2000). Of course, other culture-specific variations may also contribute to explaining this difference, and future research including other national-cultural groups is needed to explore this possibility (cf., Bond & Van de Vijver, 2011; Vignoles et al., 2016). On a methodological level, the differences that we found among three East Asian samples point to the value of conducting cross-cultural research in multiple exemplars of each culture or society, so that the obtained findings are truly culture-general, rather than being attributable to unnamed culture-specific or society-specific factors (Boer et al., 2018; Milfont & Klein, 2018).
Passive-destructive responses were associated with poorer relationship outcomes in all of the samples we studied, although these associations were somewhat stronger in the North American samples than in the East Asian samples. Passive-destructive responding reflects disinterest and disengagement, signs that a relationship is waning in both individualistic and interdependent cultures. One reason why people describe their positive events to relationship partners is to share their positive feelings with significant others, thereby building relationship capital (Gottman, 1999; Langston, 1994; Reis et al., 2010). Passive-destructive responses undermine such efforts by suggesting that the partner being told has little or no interest in the event, and by implication, in the capitalizer. In this circumstance, relationships are unlikely to thrive. In a similar vein, active-constructive responding was strongly linked to favorable relationship outcomes in all six samples. One of the most central characteristics of relationship closeness is “including the other in the self” (Aron et al., 2013), and perceiving a partner’s shared sense of pleasure in one’s personal good news provides clear evidence of this.
Although the pattern of associations was generally similar in the three North American samples, two results bear mention. First, compared to college students, in the relatively older adult sample, active-constructive responding was more strongly related to favorable relationship outcomes, and active-destructive responding was more strongly related to unfavorable relationship outcomes. This result may indicate more influential consequences of these active forms of responding in long-term, established, higher-interdependent relationships compared to relatively new, and mostly less committed, relationships. Further research is needed to better understand how capitalization may differentially affect new versus established relationships. Second, we observed no significant interactions involving the Northeastern and Southern samples. In part, this may reflect some degree of homogeneity in individuals who attend universities. Or, more conceptually, it may indicate that the “culture of honor” idea that distinguishes southern and northern U.S. cultures (Leung & Cohen, 2011), which focuses primarily on threats, may have little relevance to how people in relationships respond to each other’s good news.
These results indicate that general theories of capitalization (e.g., Peters et al., 2018), rather than simply classifying certain responses as relationship-enhancing and others as relationship-harming, need to consider their culturally grounded meaning. Social support theories recognize that the meaning of support communications depends on cultural values, often referenced implicitly rather than explicitly, such that East Asians are more likely to avoid calling attention to themselves (e.g., Kim et al., 2008). To date, capitalization theories have not done similarly but our results indicate how they can do so–that is, by recognizing that relationship-promoting responses need not seek to directly boost the capitalizer’s pride or pleasure but may instead take a more nuanced approach that acknowledge both sides. Furthermore, within-culture variations also mattered in our work, suggesting that other cultural factors may also influence the capitalization process–for example, ethnicity and social class (consistent with the finding that communication processes that are maladaptive for middle-class U.S. couples may be adaptive for socioeconomically disadvantaged couples; Ross et al., 2019). Accounting for cultural variations will enable theories of capitalization to provide a deeper understanding of what works for whom in which context.
On a more pragmatic level, our results suggest that interventions based on teaching couples how to converse more effectively about their personal good fortune (e.g., Conoley et al., 2015; Woods et al., 2015) should design their programs to accord with the cultural and even subcultural values of their participants. That is, although the active-constructive approach was associated with positive outcomes in all of the groups we studied, and therefore should not be discouraged, other approaches were also beneficial for East Asians. For example, with East Asians, it would be important to recognize that pointing out the downsides of a happy event is not necessarily detrimental to a relationship. In other words, it would be best to devise interventions that are more flexibly adapted to cultural circumstances and preferences.
Limitations
Several limitations of this research deserve attention. First, our study relied entirely on self-reports, and may have been influenced by motivated reinterpretation. Although people’s interpretations of their relationship circumstances are worthy of study in their own right, it would be desirable to replicate this research with direct observation of capitalization conversations. Relatedly, because it seems possible that the responses we observed would influence people’s willingness to share good news. it would be valuable to see if there are cultural differences in the frequency with which people share good news with their partners. Second, our work was intentionally focused on romantic relationships, but people often capitalize with other partners, particularly parents, other relatives, close friends, and acquaintances. It will be important in future studies to determine if the culture differences we observed also extend to these other types of social roles and relationships. Third, we did not provide direct evidence to explain the differences obtained among the Mainland China, Hong Kong, and Taiwan samples. Future research is needed to directly examine potential explanations for these noteworthy differences. Fourth, we did not collect information about our participants’ socioeconomic status, disability, or sexual orientation, and the possible impact of these variables should be considered in future work. Finally, it will also be important to examine capitalization in other cultural and societal contexts beyond East Asia and North America. Aside from helping to identify the conceptual factors underlying cultural differences in capitalization, multi-cultural investigations can also contribute to understanding of individual differences (Smith & Bond, 2019).
Conclusion
Recent research has identified the capitalization process—how people share their good news with partners, and how those partners respond—as a highly valued element of interaction between relationship partners. This research is the first to systematically and comprehensively document how this process varies between East Asia and North America. Although capitalization may be a universal feature of romantic relationships, the manner in which it plays out varies as a function of culture. Better understanding of these cultural variations will contribute to deeper insights about the role that relationships play in human behavior.
Supplemental Material
Supplemental Material - Are you happy for me?: Responses to sharing good news in North America and East Asia
Supplemental Material for Are you happy for me?: Responses to sharing good news in North America and East Asia by Harry T Reis, Sisi Li, Yan Ruan, David C de Jong and Fen-Fang Tsai in Journal of Social and Personal Relationships
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Open research statement
This research was not pre-registered. The data used in the research are publicly available at https://dataverse.harvard.edu/privateurl.xhtml?token=31fe5024-7cdd-418b-b0bd-f635239f590b or by email to the first author,
or can be obtained by email to the first author,
Supplemental material
Supplemental material for this article is available online.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
