In press, Personality and Social Psychological Bulletin.
Running Head: Culture and Self-Satisfaction
In contrast to the reliable effects observed with North Americans, research with Japanese has failed to detect self-enhancing biases. We considered the possibility that, owing to the need to adapt themselves to others’ expectations, Japanese are more critical of themselves than are North Americans. A comparison of actual-ideal self-discrepancies indeed revealed larger discrepancies for Japanese than for either European- or Asian-Canadians. Moreover, the magnitude of the cultural differences were larger for characteristics which participants viewed as more important, and the relation between depression scores and actual-ideal discrepancies was weaker for Japanese than for European-Canadians. The data support the notions (a) that Japanese are more likely than North Americans to be dissatisfied with themselves and (b) that these self-critical attitudes are less distressing for Japanese.
Culture, Self-Discrepancies, and Self-Satisfaction
Research on the self in North America has had a strong focus on biases in self-perceptions. This research has demonstrated that the typical (North American) individual’s self-evaluation is fraught with inaccuracies and distortions. These do not occur randomly, but rather are biased systematically toward casting the self in an unrealistically positive light. These positive self distortions have emerged across a variety of paradigms and in fact are so common that many argue that they are basic and fundamental ways of thinking, at least in North America where the bulk of this research has been conducted. These biases have been construed as information-processing errors (Miller & Ross, 1975), egocentric knowledge organizations growing out of an “intrapsychic evolution” (Greenwald, 1980), and self-protective tactics which serve to bolster the individual’s subjective well-being (Kunda, 1987, 1990; Taylor & Brown, 1988). Recently, however, the utility of self-enhancement has been challenged by a number of researchers (Colvin & Block, 1994; Colvin, Block, & Funder, 1995; John & Robins, 1994), and the biases demonstrated in these studies have been shown to be associated with psychological maladjustment and narcissism. There is thus considerable controversy over the characteristics associated with these biases, the methodologies which are appropriate to study them, and the proportion of the population embracing them. Nonetheless, across these (North American) samples there is consensus that the distortions are systematically in a positive direction (self is better than is actually the case), and that these biases are resistant to change (Krueger & Clement, 1994; Weinstein & Klein, 1995).
Despite the frequency with which self-enhancing biases are observed in the West, they do not appear to characterize the typical thinking pattern of people from Eastern cultures, particularly Japanese (Markus & Kitayama, 1991b). For example, the unrealistic optimism bias, whereby individuals view negative future life events as more likely to happen to others than to themselves (see Weinstein, 1980), is clearly less pronounced for Japanese than for North Americans (Heine & Lehman, 1995a). In some cases, in fact, Japanese seem to be unrealistically pessimistic. Neither are tendencies to view oneself as better than the majority of others evident in Japanese samples (Heine & Lehman 1997b; Markus & Kitayama, 1991a). And this is despite the findings that North Americans typically view themselves in this way (Campbell, 1986; Marks, 1984). Research on attributional biases has demonstrated routinely that Westerners tend to internalize success and externalize failure (for a review see Zuckerman, 1979), yet this self-serving pattern has not been detected with Japanese (for a review, see Kitayama, Takagi, & Matsumoto, 1995). Kitayama, Markus, Matsumoto, and Norasakkunkit (1997) found that whereas Americans are likely to view their daily experiences in terms of opportunities for self-enhancement, Japanese are likely to view such experiences in self-critical terms. Furthermore, this relative reluctance to self-enhance on the part of Japanese is not merely observable at the individual level: Japanese also exhibit fewer group-serving tendencies than do North Americans when evaluating their family members, their universities, and their cities (Heine & Lehman, 1997b; Kitayama, Masuda, & Palm, 1996). Thus far, research with Japanese has consistently failed to demonstrate any reliable positive distortions in their self-evaluations.
Importantly, the obtained cultural differences in self-enhancement do not seem simply to be the result of Japanese trying to present themselves in a more modest manner than do North Americans (for a review see Heine, Lehman, Markus, & Kitayama, 1998). Analyses of questionnaire data provide no evidence to suggest that Japanese or Asians more generally are feigning modesty in questionnaire studies (Diener, Suh, Smith, & Shao, 1995; Heine & Lehman, 1995a, 1995b; Kitayama et al., 1997). More compellingly, however, unobtrusive behavioral measures have revealed this same absence of self-enhancing and self-affirmational tendencies among Japanese (Heine, Kitayama, Lehman, Takata, & Ide, 1998; Heine & Lehman, 1997c; Heine, Takata, & Lehman, 1998). The data thus are converging on the notion that Japanese are not simply saying that they view themselves less positively than North Americans—they truly seem to feel this way.
Interpreting the Lack of Self-Enhancement for Japanese
We interpret the above cultural differences in self-enhancement from the vantage point of cultural psychology. Cultural psychology maintains that the self is born of the interaction between the person and a set of culturally derived beliefs, values, institutions, customs, and practices (Fiske, Kitayama, Markus, & Nisbett, 1997; Greenfield, 1997; Shweder, 1990). The self and attendant psychological structures and processes are thus supported by a web of cultural meanings, and likewise, the interaction of individual selves creates and sustains the cultural environment. In this way culture and self are seen to make each other up (Shweder, 1990).
Our central point is that these cultural differences in self-enhancement exist because of the ways in which the self is served in each culture. We suggest that, for the most part, the North American self is served when the individual views him or herself in an unrealistically positive light, for example, as better than most others (Taylor & Brown, 1988; but for an opposing view see Colvin et al., 1995). That Western culture places relatively greater value on individuals being adequate, competent, and self-sufficient (Markus & Kitayama, 1991b; Sampson, 1977) suggests that viewing oneself in unrealistically positive terms (i.e., as especially competent, in control, etc.) can thus be seen to bridge the gap between the individual’s actual standing and the cultural ideals, thereby authenticating the individual as a meaningful member of the culture (Heine & Lehman, 1995a). Self-enhancing biases serve to bring Westerners closer to their cultural ideals of selfhood.
In contrast, the relation between self-enhancing biases and the Japanese cultural ideals of selfhood appears to be quite different. Relatively more important cultural tasks for Japanese are to fit in harmoniously with others and to gain a sense of belongingness and interdependence with others (e.g., Bachnik, 1992; De Vos, 1985; Markus & Kitayama, 1991b). We suggest that the self is served in Japan when individuals feel that they are being accepted by their groups. This emphasis on fitting in with others suggests that it does not so critically matter how individuals evaluate how well they are doing—rather, it is more important how the groups to which they belong evaluate their performance (Spence, 1985; cf. Yamagishi, 1988). Hence, feeling good about oneself, far from hinging primarily on an individual’s personal feelings and self-evaluations, has more to do with the feelings and evaluations of others. For Japanese, it is crucial to strive to gain others’ approval.
Performing cultural tasks associated with interdependence leads Japanese to be vigilant about how they are being evaluated by their in-group members (Heine, Lehman et al., 1997; Kitayama, Markus, & Lieberman, 1995; Kitayama et al., 1997). Japanese are encouraged to focus on how their behavior affects their relations with others and how their behavior affects the overall harmony of the group. Succeeding in interdependent cultural tasks requires the individual to change and adapt him or herself to the needs of the group (Weisz, Rothbaum, & Blackburn, 1984).
The Japanese psychiatrist Takeo Doi (1973) argued that a characteristic feature of Japanese is that they maintain a perpetual sense of ki ga sumanai, that is a sense of dissatisfaction about themselves. This dissatisfaction indicates a perceived discrepancy between Japanese individuals’ current states and their aspirations. It is crucial for Japanese to dwell on their inadequacies and shortcomings—those aspects that render them vulnerable in terms of securing their groups’ approval (Heine, Lehman et al., 1997; Kitayama et al., 1995, 1997; see also Kashiwagi, 1986). Japanese are motivated to be keenly sensitive both to the ways in which they are interfering with or limiting their group, and to the ways in which they are being negatively evaluated by their group members. Using this information, they then can act accordingly and work towards rectifying their shortcomings (see also Johnson, 1993; Roland, 1988; White, 1987).
By identifying negative aspects of the self, and by making efforts to correct these, Japanese are more likely to succeed in the process of adapting themselves to their groups’ needs. So, succeeding in interdependent tasks impels Japanese to dwell on the negative aspects of their selves and to remain dissatisfied with themselves. These tendencies stand in sharp contrast to North Americans’ tendencies to emphasize their positive attributes via self-enhancement (e.g., Taylor & Brown, 1988).
The habitual dissatisfaction with oneself which we are describing is evident throughout Japanese culture. Various interpersonal scripts get played out in everyday life whereby people communicate their personal inadequacies and limitations (Heine, Lehman, et al., 1998; Marsella, Walker, & Johnson, 1973). Examples of such processes are amae (i.e., the notion that one indulges one’s sense of dependency on others; Doi, 1973; Kumagai & Kumagai, 1986), an emphasis on shame (Benedict, 1946; Creighton, 1990; Doi, 1973; Lebra, 1983), and the widespread occurrence of apologies in Japan (Barnlund & Yoshioka, 1990), to name but a few. This self-critical orientation is encouraged in the child-rearing process as well. From a young age, Japanese are taught to reflect upon their weaknesses and inadequacies (hansei suru; Johnson, 1993; Roland, 1988). In contrast to Western caretakers who tend to draw attention to children’s positive features by praising, encouraging, and complimenting them, Japanese caretakers are more likely to draw children’s attention to potentially negative features which may have to be corrected for the child to fit in more with others (Markus, Kitayama, & Heiman, 1996). This self-critical stance is institutionalized in the education system (White & Levine, 1986), with the goal of diminishing a sense of self-centeredness which can hinder the child’s ability to fit in well with others. Daily conversations also reflect this self-critical nature of Japanese. Kitayama and Karasawa (1996) found that Japanese report a greater frequency of being criticized by others, and a lower frequency of being complimented by others, than do Americans. Such features of everyday experiences in Japan suggest that Japanese develop habitual outlooks towards negative self-relevant information.
In fact, the conceptual literature on Japanese is consistent in describing a chronic self-critical outlook. Clearly, such an outlook is at odds with the well-documented self-enhancing tendencies of North Americans. In the present study, we investigated this potential cross-cultural difference in self-satisfaction.
Comparisons of Actual and Ideal Self-Assessments
We employed the framework of self-discrepancy theory to explore the notion that Japanese feel chronically more dissatisfied with themselves than do North Americans (Canadians). One of the two basic assumptions of self-discrepancy theory is that people are motivated to bring their current state in line with their ideal state (Higgins, 1989). Actual self-assessments represent how people currently view themselves, whereas ideal self-assessments indicate how people ideally want to be. We reasoned that the discrepancy between these two types of self-assessments is one way to measure individuals’ dissatisfaction with themselves.
One of our key questions was “Do Japanese view themselves to be more distant from their ideals than do Canadians?” Larger actual-ideal discrepancies for Japanese than for Canadians would be in line with the notion that Japanese focus more on their incompleteness. A large discrepancy between the way one is and the way one wants to be highlights dissatisfaction with oneself.
Higgins and colleagues (Higgins, 1987, 1989; Higgins, Klein, & Strauman, 1985) demonstrated that actual-ideal discrepancies correlate positively with depression (see also Marsella et al., 1973). Large discrepancies represent, in general, an absence of positive outcomes (i.e., the individual is not the type of person that he/she wishes to be) and are associated with dejection, sadness, and disappointment (Higgins, 1987). Such a relation between actual-ideal discrepancies and depression is consistent with the idea that large discrepancies signal individual inadequacy (see e.g., Marsella et al., 1973).
However, if viewing oneself negatively (i.e., as further away from one’s ideal self) is more a natural part of one’s everyday cultural life, such feelings should be less likely to be accompanied by stress and consequent negative affect. To the extent that one’s culture encourages actual-ideal discrepancies, not only should such discrepancies be more common, they should be less debilitating. As Kitayama, Markus, and colleagues (e.g., Kitayama et al., 1995, 1997; Markus et al., 1996) have theorized, such a focus on actual-ideal discrepancies may serve, in part, as a means for Japanese to improve themselves in order to accomplish the tasks associated with interdependence. Indeed, in previous cross-cultural research (Marsella et al., 1973; Yanagida & Marsella, 1978), Japanese-Americans living in Hawaii have been found to exhibit an attenuated relation between actual-ideal discrepancies and depression, compared to Caucasian-Americans. One of our present objectives was to build on this research, with samples of Japanese, Asian-Canadians, and European-Canadians.
Highlighting the Role of Culture
We employed two methodological devices in order to provide the most compelling test of the relation between culture and self-discrepancies. First, we sought to ensure that any obtained cultural differences were not due to differences in the cultural meaning of the traits which were employed to assess self-discrepancies. To avoid any “imposed-etics” stemming from our Western orientation (e.g., Berry, 1969), we included traits which are meaningful both to Japanese and Canadians. Towards this end we conducted a pretest to determine which traits are viewed as most important by Japanese and Canadians for succeeding in their respective cultures, and we included those traits in the actual study.
Second, to better understand the degree to which any obtained cultural differences were due to culture, and not some extraneous variable, we included a third sample which theoretically characterizes a group in between Eastern and Western cultures: Canadians of Asian descent. We collected the Canadian data in Vancouver, a city with large proportions of people of either European or Asian ancestry. This latter group, which we label “Asian-Canadians,” although heterogeneous in terms of country of origin (the majority of this sample is of Chinese descent) and length of time/number of generations in Canada, approximates a culture falling midway between the groups of European-Canadians and Japanese in terms of exposure to Western cultural values (Heine & Lehman, 1997a, 1997b; Kitayama et al., 1997). To the extent that culture mediates self-discrepancies, we anticipated that Asian-Canadians would exhibit results intermediate to European-Canadians and Japanese. Such a pattern would increase our confidence that any obtained differences were due to the cultural backgrounds of the participants.
The Japanese sample consisted of 161 students (58 females and 99 males, and 4 who did not report their gender) enrolled in an introductory urban studies course at Ritsumeikan University in Kyoto. All Japanese participants were born in Japan, and all but one had Japanese parents.
The Canadian sample consisted of 268 students enrolled in introductory psychology classes at the University of British Columbia (UBC). We partitioned this sample by ethnic background to further examine cultural differences. One hundred and fifty-one (111 females and 40 males) declared themselves to be of Asian heritage and formed the “Asian-Canadian” sample. Ninety (65 females and 25 males) declared themselves to be of European heritage, and formed the “European-Canadian” sample. The remaining 27 students were of varied ethnic backgrounds (e.g., mixed-ethnicities, Latin-American descent, African descent, etc.) and were not included in the analyses.
Participants first were asked a number of questions about 20 personality traits. Toward ensuring that our list of traits would be deemed relevant by both Canadians and Japanese, we conducted a pretest with separate samples of Canadians and Japanese to determine which traits were viewed as important. The procedure of the pre-test was as follows: First we met with several Japanese university students to discuss which traits they viewed to be most important for succeeding in Japanese culture. From this discussion a list of 20 important traits for Japanese was constructed. Another 20 traits were chosen by us to reflect important traits for succeeding in Canadian culture. This combined list of 40 traits was then given to a class of introductory psychology students at Toyama University, Japan and to a class of introductory psychology students at UBC (Canadian participants for this pre-test were limited to those who declared themselves to be of European-ancestry). Participants were asked to indicate how important they perceived each trait to be for succeeding in their own culture on a scale from 1 (not at all important) to 10 (extremely important).
The traits which were included in our main study were based on the ratings of the initial 40 traits in the pretest. We selected 20 of these traits: specifically, the 10 traits which were rated as most important by Japanese and the 10 traits which were rated as most important by Canadians. Three traits (getting along well with others, cooperativeness, and adaptability) were rated among the highest 10 traits for both cultures, so we also included the traits rated 11th most important for each culture, and the trait rated 12th most important by Canadians to reach our initial target of 20.
These 20 traits were put into three different types of statements which participants were asked to rate in terms of their accuracy on a Likert scale from 1 (Not at all accurate) to 6 (Completely accurate). To reduce potential ceiling effects (a concern particularly for the “ideal” statements), the statements were constructed using “extremely” as a modifier. On the first page, participants were asked to indicate how accurate the statements were in describing themselves (e.g., “I am extremely attractive”). On the second page, participants were asked how accurate the statements were in describing the type of person they ideally would like to be (e.g., “I would ideally like to be extremely attractive”). On the third page, participants were asked to indicate how accurate the statements were in describing the average student, same-gender as themselves, from their university (e.g., “She or he is extremely attractive”). Next, participants were asked to indicate how important they felt each of the 20 traits were for succeeding in their culture on a 1 (not at all important) to 10 (extremely important) scale. Last, participants completed Zung’s (1965) 20-item Self-Rating Depression Scale on a Likert scale from 1 (none of the time) to 5 (most or all of the time). This scale is a reliable and valid instrument for assessing depression, and it has been employed cross-culturally as well (e.g., Yanagida & Marsella, 1978; Zung, 1969).
All of the materials were originally produced in English and then translated into Japanese. Then, after an independent translator back-translated the Japanese version into English, three translators discussed and resolved any inconsistencies between the versions.
Comparability of Samples
A significant difference emerged in the average ages of the three samples, F(2, 395) = 5.66, p < .01. Post-hoc comparisons1 revealed that the Japanese sample (M = 20.4) was significantly older than the Asian-Canadian sample (M = 19.9), with the European-Canadian sample (M = 20.1) falling nonsignificantly in between. However, correlations within cultures between age and each of the dependent variables revealed a significant relation in only one instance, which will be discussed later. The samples differed with respect to gender proportion, 2[2, N = 398] = 51.1, p < .001. Relations with gender were analyzed for all the main dependent variables. No significant gender x culture interactions nor significant main effects for gender were found.
Comparisons of Self-Discrepancies
Before calculating self-discrepancies we examined how the different cultural groups evaluated themselves with respect to each of the actual self, ideal self, and average other evaluations. Participants’ evaluations were summed across the 20 traits for each of the three scales. First, an ANOVA of actual self ratings revealed a highly significant effect for culture, F(2, 388) = 63.68, p < .001. Post-hoc comparisons indicated that European-Canadians rated the traits as more characteristic of themselves than did Asian-Canadians, who in turn rated them as more characteristic than did Japanese (see Table 1). This finding is consistent with the notion that North Americans have more positive self-views than Japanese (Heine, Lehman et al., 1998). A significant cultural difference also emerged with respect to ideal-self ratings, F(2, 391) = 34.69, p < .001, and post-hoc comparisons revealed that this was due to both groups of Canadians evaluating the traits as more characteristic of their ideal selves than Japanese. This finding is interesting, and suggests that Japanese may be less inclined than Canadians to desire to possess traits to an extreme degree. Cultural comparisons of participants’ ratings of the average student from their university also revealed a significant effect, F(2, 390) = 14.27, p < .001. Post-hoc comparisons indicated that Japanese rated the traits as less characteristic of the average other student than did both groups of Canadians. This finding suggests that Japanese may be more critical of others than Canadians (see Kitayama & Karasawa, 1996).
The consistent pattern of cultural differences from the scale totals compels us to consider that Japanese may be less likely than Canadians to view traits as characteristic of any individual. Conceiving of people as a collection of traits may not be as characteristic for Japanese as it is for North Americans (e.g., Cousins, 1989; Rhee, Uleman, Lee, & Roman, 1995). As well, that the Japanese responses are consistently closer to the midpoint of the scales raises the possibility that moderacy response sets are driving the Japanese results more so than for either group of Canadians (Stening & Everett, 1984; Zax & Takahashi, 1967). The cultural differences reported here, however, are far larger than those typically observed due to such response styles (Chen, Lee, & Stevenson, 1995). The pattern of results of the scale totals, then, does not lend itself to an unambiguous interpretation.