Anyone... may be forgiven for assuming that social cognition research has uncovered universal principles of how people make sense of themselves and of their social worlds. But just how universal are these principles? The vast majority of social cognition research has been conducted in North America, and much of the rest has been conducted in Western Europe. Could it be that some or even most of the findings apply only to North Americans or, more broadly, to Westerners? Just how culture bound is social cognition?
Anthropologists have long noted remarkable differences in how people raised in Western cultures and those raised in Eastern cultures understand themselves and their social worlds (e.g., Shweder and LeVine 1984). To capture the flavor of these differences, consider the following anecdotes: In the United States, “the squeaky wheel gets the grease.” But in Japan, “the nail that stands out gets pounded down.” American parents trying to get their reluctant children to eat may urge them to think about how much more fortunate they are than all the starving children in Africa. Japanese parents faced with the same dilemma may, instead, urge their children to consider how bad the farmer who had worked so hard to produce their food would feel if they refused to eat it. An American company intent on increasing productivity asked its employees to look in the mirror before coming to work each day and tell themselves “I am beautiful” 100 times. A Japanese company with a similar intent instructed its employees to begin each day by telling a fellow employee that he or she is beautiful (these anecdotes were reported by Markus and Kitayama 1991). American children put together sports teams by appointing two captains who take turns selecting team members. Japanese children reject this method because the last picked child might feel upset. Instead, they form teams by classroom or by the Japanese equivalent of alphabetical order. American mothers want their children to grow up to be happy and successful individuals. Japanese mothers asked what they want for their children may say something like “I just want my kid to grow up so as not to be a nuisance to other people” (N.D. Kristoff, NYT 4.12.1998).
Such anecdotes reveal strikingly different understandings of the self and its relation to others. The American examples point to a sense of the self as distinct from others, and highlight the value of being different and special. In contrast, the Japanese examples point to a sense of the self as closely interrelated with and dependent on others, and highlight the value of harmoniously fitting in with one’s group. Such differing construals of the self and society may lead to substantial differences in how people think and feel about themselves and others and in how they relate to each other.
Until relatively recently, such cultural differences were explored predominantly by anthropologists whose conclusions rested, for the most part, on observational and ethnographic data. Over the last decade or so, an increasing number of social psychologists have begun investigating the impact of culture on cognition, motivation, and emotion. Their work has resulted in the emergence of a new field, which may be termed experimental anthropology; it relies on the rigorous experimental methods developed by social and cognitive psychologists to explore cultural differences. The revealed cultural differences in social cognition are interesting in their own right. In many cases, they may also shed light on topics that have long been central to social psychology such as the fundamental attribution error, dissonance reduction, and the need for self-enhancement.
For example, the fundamental attribution error, that is, people’s tendency to exaggerate the extent to which behavior is driven by underlying dispositions and to underestimate the extent to which it is driven by situational forces, has often been explained as resulting from basic perceptual processes. The perceiver, the argument goes, is bound to see the actor as a salient figure against a much less salient ground, the situation. Because of the relative salience of the actor, causal explanations focus on this actor’s disposition. Such basic perceptual processes should be universal. Therefore, if it turns out that people in non-Western cultures are less prone than Westerners to the fundamental attribution error, this would challenge the perceptual account of this phenomenon and suggest, instead, that it is rooted in people’s culture-bound beliefs and theories about what drives social behavior.
Much of the social psychological research on cultural differences has focused on East-West differences, and therefore most of this chapter will be concerned with these differences. The starting point of research on East-West differences has been the notion that the self is construed quite differently in these two cultures: North Americans and other Westerners typically view the self as independent whereas East Asians typically view the self as interdependent with others (for detailed reviews, see Markus and Kitayama 1991; Triandis 1989). After briefly outlining the basic differences between these two construals of the self, I will describe research that has investigated the implications of these differences for self-representation, attribution, and motivation.
East-West differences are presumed to be so fundamental that they affect one’s very sense of selfhood and have wide-ranging implications for the way one perceives and thinks about the self and others and the way one reacts to a wide variety of social situations. Not all cultural differences are so comprehensive, though. Two cultures may differ in their understanding of a particular class of behaviors, even though they view most other aspects of social interaction similarly. Such relatively narrow cultural differences have been found to exist within the United States between Northerners and Southerners. I will describe research pointing to differences in how threats to one’s honor are construed in the Northern and Southern regions of the United States, and will discuss the implications of these differences for interpersonal judgment and behavior.
In many Western cultures people are socialized to become unique individuals, to express their thoughts and desires, to strive to accomplish their personal goals, to self-actualize, to realize their potential. The self is viewed as an independent, autonomous, separate being defined by a unique repertoire of attributes, abilities, thoughts, and feelings (I am smart, kind, and responsible, I’m good at math, I am not very musical, I love to travel, I want to outperform my peers). One attempts to express these aspects of the self publicly and to confirm them privately through comparisons with others. It is this repertoire of internal attributes that organizes and gives meaning to one’s sense of oneself. One also has knowledge about oneself in a variety of social relations (I’m competitive with my brother, witty with my best friend, shy with members of the opposite sex), but this knowledge is not viewed as central to one’s core identity and is not quite as self-defining as one’s more global inner attributes. Such independent construals of the self are far more common in Western than in Eastern cultures, although there are considerable variations within each culture.
I once heard a Japanese scholar remind his American audience of their sanctified rights and utmost goals, “Life, liberty, and the pursuit of happiness.” Suggesting that their forefathers might have overlooked something of great importance, he asked “What about harmony?” In many non-Western societies (e.g., Japan, China, India), people are socialized to strive for harmonious relations with others, to focus on the connectedness of individuals to one another, to adjust themselves to the demands of social situations, and to try to fit in with their social group. The self is viewed as interdependent with others and is experienced as part of a social web. One’s behavior, thoughts, and feelings are seen as dependent on those of others in the relationship (I respect my parents, I am sensitive to my spouse’s needs, my family sees me as friendly, my coworkers think I travel too much, I want my company to outperform its rivals). One’s sense of self is grounded in one’s social relationships, and the most meaningful aspects of oneself are those that emerge in relation to others. As a result the self may be experienced as fluid, taking on different colorations in different social settings.
An interdependent self embraces its assigned role, focuses on social duties and obligations, and tries to figure out what others are thinking and feeling so as to best meet their expectations. One acts in accordance with one’s own wishes and desires; indeed, acting on one’s own wishes is viewed as infantile. Relationships, rather than being a means of realizing one’s own goals, become an end in themselves, so much so that others’ goals may be experienced as one’s own. Note, though, that the interdependent self need not be generally more benevolent and concerned about others’ welfare than is the independent self. The interdependent self is not attuned to the needs of all others, only to those of the in-group, and the in-group may be very narrowly defined.
To illustrate how the interactions among individuals with interdependent selves might differ from those among individuals with independent selves, Markus and Kitayama (1991) describe how events might unfold if a person has a friend over for lunch. An American might ask the friend “Hey, Tom, what would you like on your sandwich? I have turkey, salami, and cheese.” And Tom would express his preference. The assumption underlying this interaction is that Tom has an inherent right to make a choice that reflects his preferences and desires. Although this assumption is taken for granted by individuals with independent selves, it is not shared by those with interdependent selves. If a Japanese visitor were asked “What do you want on your sandwich?” there would probably be a moment of baffled silence followed by a noncommittal “I don’t know.” To a Japanese, it is the host’s responsibility to “read the friend’s mind,” figure out what the friend would like, and offer it. The friend, in turn, should accept whatever is offered with grace, and be ready to return the favor in the future. In Japan, therefore, the host might say something like “Hey, Tomio, I made you a turkey sandwich because I remember that last week you said you liked turkey more than beef.” And Tomio would graciously thank the host (Markus and Kitayama 1991, p. 229).
The experience of seeing the self as crucially connected to others, knowing and trying to do what is best for them, and allowing their goals to take precedence over one’s own, is not foreign to Westerners. It is not unusual, for example, for American parents to put their children’s needs above their own. Still, the sense of the self as interdependent is more common and pervasive in non-Western than in Western societies. Indeed, several studies have revealed systematic differences among Americans and East Asians in the contents and richness of self-knowledge, as discussed next.
Think of a friend you know very well. How similar are you to this friend in shyness? How similar is your friend to you in shyness? Remarkably, Americans give somewhat different answers to these seemingly identical questions (Holyoak and Gordon 1983). When people are asked to make comparisons between a highly familiar object and a less familiar one, their responses reveal a systematic asymmetry: The unfamiliar object is judged as more similar to the familiar one than vice versa. For example, people who know more about the USA than about Mexico judge Mexico to be more similar to the USA than the USA is to Mexico. Such asymmetries are well-explained by Tversky’s contrast model of similarity judgments (Tversky 1977). According to this model, when one is asked a question such as “how similar is the USA to Mexico?” perceived similarity decreases with the number of attributes that are unique to each country. However, similarity is decreased more by the unique attributes of the subject of the comparison (in this case, USA) than by the unique aspects of the referent (in this case, Mexico). Because we are highly familiar with the USA, we are aware that it has many unique attributes (i.e., attributes not shared with Mexico), and so we judge similarity to be relatively low. In contrast, when Mexico is the subject of the comparison (as in “How similar is Mexico to the USA), it is Mexico’s unique attributes that one focuses on. Because we know little about Mexico, we can come up with only a small number of attributes unique to it, and so we judge similarity to be relatively high.
The asymmetry obtained for Americans’ judgments about their similarity to a close friend therefore reveal something about the relative richness of their representations of self and other. Much like they might judge Mexico to be more similar to the USA than the USA is to Mexico, North Americans judge a friend to be more similar to the self than the self is to the friend (Holyoak and Gordon 1983). For example, I would say that you are more similar to me than I am to you (though you might think otherwise...). This suggests that the self is more richly represented than the other; because I know much more about my unique attributes than about yours, I judge our similarity to be lower when I am the subject of the comparison (and my unique attributes are salient) than when you are the subject of the comparison (and your unique attributes are salient).
This relatively greater richness and elaboration of self-knowledge as compared to knowledge about close others may be a by-product of a culture that promotes preoccupation with the independent self and its attributes. An interdependent culture may, instead, predispose its members toward developing an elaborate understanding of each other. Individuals whose key social tasks include fitting in with others and reading their minds to anticipate their expectations may accumulate detailed knowledge about close others. Their knowledge of others may be as rich or richer than their self-knowledge. If so, they should not show the same pattern of asymmetry in judgments about the similarity between the self and another person that North Americans tend to show.
A study reported by Markus and Kitayama (1991) supported these predictions. American students and students from India were asked to judge the similarity of the self to another person or the similarity of another person to the self. As in previous studies, the American students believed that the other was more similar to the self than the self was to the other. In contrast, as shown in figure 11.1, this pattern was reversed for the Indian students (though the reversal was not significant). This suggests that for the Indian students, the other was represented at least as richly as the self.
 |
 |
In sum, individuals with independent selves appear to have more rich and elaborate representations of the self than of other individuals. This relative advantage of self-representations over representations of others is eliminated and, possibly, even reversed for individuals with interdependent selves. Individuals with independent and interdependent selves differ not only in the relative richness of their self-knowledge but also in its contents, as discussed next.
Imagine you were asked to generate 20 different answers to the question “Who am I?” What would you say? The kinds of responses you provide and the ease with which you can generate them may depend heavily on your cultural background. To an American, this question seems natural and obvious; it permits one to express one’s stable, independent inner self. The American will likely respond with a list of global traits, abilities, and preferences. In contrast, a Japanese person confronted with the same question may find that it presupposes an unnatural separation between one’s identity and one’s social context. To answer the “Who am I?” question, the Japanese may feel compelled to supplement each response with an appropriate social context.
These ideas gained support from a study in which American and Japanese college students were asked to generate self-descriptions (Cousins 1989). The students first provided 20 responses to the global “Who am I?” question. Examination of the five attributes chosen by each student as most self-defining revealed that the Americans were considerably more likely than their Japanese counterparts to respond with global, unqualified traits (“I am honest,” “I am easygoing”), as may be seen in figure 11.2. The Japanese students, who were relatively reluctant to describe themselves in terms of global traits, were, instead, especially likely to ground themselves in social affiliations (“a Keio student”) and activities (“one who plays Mah-Jongg on Friday nights”). The Japanese students were also considerably more likely than the Americans to fall back on universal abstractions (“a living form,” “a person of the twentieth century”). In short, whereas the American students responded to the “Who am I?” question mostly with global traits, the Japanese students responded mostly with social roles, activities, preferences, and abstractions.
A very different pattern emerged when the same students were given a contextualized version of the “Who am I?” questionnaire, one that asked them to describe themselves at home, at school, and with close friends. For Americans, trait terms may lose their global, cross-situational meaning when they are restricted to a particular context, because Americans conceptualize traits as transcending situations. In contrast, for Japanese individuals, the contextualized questions ground the self in its natural social situations, and provide an opportunity for reflecting on the traits one displays in each situation. Indeed, as shown in figure 11.2, Japanese students were considerably more likely than their American counterparts to answer the contextualized questions with unqualified traits (studious, quiet, boastful). Americans, in contrast, were especially likely to provide qualified responses to these contextualized questions (e.g., “I am silly with close friends”). Cousins theorized that such qualifications to an already contextualized question were intended to convey that the self in that context need not reflect one’s true, context-free self.
In sum, the Americans were more likely than the Japanese to describe themselves in terms of unqualified, global traits when asked for global context-free self-descriptions, but were less likely than the Japanese to use such traits when asked to describe themselves in particular social contexts. Put differently, the American students were more likely to use traits to describe their global than their contextualized selves, whereas the Japanese students were more likely to use traits to describe their contextualized than their global selves. These findings that the American self is conceptualized as independent of its social milieu, as comprising global, context-free attributes. In contrast, the Japanese self is conceptualized as dependent upon social situations, as comprising context-specific attributes.
Americans’ view of the self as characterized by global attributes that transcend situations may be at the root of their tendency to attribute other people’s behavior to stable, global attributes. East Asians, who are more likely to see the self as dependent upon one’s social roles and situations, are also more likely than Americans to attribute other peoples’ behaviors to their roles and situations rather than to their underlying dispositions, as discussed next.
Take a moment to think of something an acquaintance of yours has done recently that you consider wrong. How do you explain your acquaintance’s inappropriate behavior? The kind of explanation you come up with may depend on your cultural background. There is considerable evidence that Westerners tend to explain other people’s behavior in terms of stable underlying dispositions (for a review, see Ross and Nisbett 1991). A Westerner will likely attribute an acquaintance’s wrongdoing to some ingrained personality flaw: She lied because she is dishonest, he hit because he is aggressive, she insulted because she is insensitive. Such views of social behavior as driven by stable underlying dispositions may be fostered by a culture that encourages its members to view individuals as autonomous, independent agents who are separate from their social context. In contrast, a culture that encourages its members to view people as interdependent with each other and with their social environment, as many non-Western cultures do, may give rise to very different explanations for social behavior. Non-Westerners may see other people’s behavior as determined by their social roles and interpersonal situations rather than by their personalities. An East Asian may attribute an acquaintance’s wrongdoing to some constraint imposed by the person’s role or situation; she lied because she had to defend her client’s interests, he hit because he was provoked, she insulted because she was stressed out.
Such East-West cultural differences in the way people explain everyday behaviors were demonstrated in an elegant set of studies by Joan Miller (1984). Miller recruited participants from a Western background (Americans in Chicago) and participants from an Eastern background (Hindus in Mysore, a city in South India). Each was asked to recount and explain two examples of a deviant behavior by an acquaintance and two examples of a prosocial behavior (using questions similar to the one at the start of this section). Their explanations were coded as to whether they referred to dispositions (e.g., proud, dishonest) or to context, which included interpersonal relations (e.g., “She is his aunt,” “He has many enemies”) as well as location in time and space (e.g., “it was early in the morning,” “He lives far away from school”).
There were striking differences in how Americans and Hindus explained social behavior. As may be seen in table 11.1 Americans were considerably more likely than Hindus to explain behavior as due to general dispositions whereas Hindus were considerably more likely than Americans to explain behavior as due to aspects of the context. This pattern was especially pronounced for the deviant behaviors. For these, Americans were more than three times as likely as provide dispositional than contextual explanations, whereas Hindus were more than twice as likely to provide contextual than dispositional explanations.
Interestingly, a sample of children from both cultures revealed no differences in how they explained social behavior. Cultural differences in attribution appear to emerge only in mature individuals who have been well-socialized into their culture’s view of the person.
Miller reported a couple of examples that illustrate the different kinds of explanations provided by adult Americans and Hindus. One American described the following transgression:
This involved one of the teachers I work with at school. It was a process of scheduling—something to do with scheduling. I came up with an innovative idea of organizing the scheduling, of what we should do. I talked to some of the other faculty members about it, and this first teacher picked it up and quickly went to the principal and presented it as if it were his own idea. (p. 967)
The American’s explanation for this behavior was unmistakably dispositional: “He was just a very self-absorbed person. He was interested only in himself.”
A rather similar transgression, also involving taking credit for someone else’s ideas, was described by one Hindu participant:
This involved a scholar in some other department, and she has got her PhD now. She wanted to publish four or five papers from her thesis. She produced some papers, but the thing is, her advisor, he put his name as first author and this young scholar as the second author. She was very hurt because that means usually the credit goes to the first author. (p. 968)
Rather than attributing this behavior to the advisor’s poor character, as the American had for a similar act, this Hindu participant attributed it to social role relations: “She was his student. She would not have the power to do it (publish it) by herself.”
Although these findings point to strong cultural differences in attribution, the following objection may be raised: The Americans and the Hindus may have differed not in how they explained behavior but in the kinds of behavior that they had generated; perhaps the Americans were especially likely to narrate acts that are well explained by dispositions, and the Hindus acts that are well explained by situational constraints. Maybe, if given the opportunity, even the Americans would explain the behaviors reported by the Hindus as driven mostly by situational constraints. Miller cleverly anticipated this objection and provided additional data that effectively ruled it out. In the second phase of her research, a new group of Americans listened to an experimenter read a sample of behaviors that had been generated by Hindu participants. Each of these behaviors had been explained by the Hindu participant who had originally described it as due to contextual factors rather than to dispositions (though these original explanations were not shown to the American participants). After hearing about each behavior, the Americans were asked to explain it.
Even when trying to account for behaviors attributed by the Hindus who had generated them to contextual factors, the Americans were considerably more likely to rely on dispositional explanations than on contextual ones; on average, 36 percent of their explanations were dispositional, whereas only 17 percent were based on context. It appears that Americans are predisposed to attribute social behavior to underlying dispositions, whereas Indians are predisposed to attribute behavior to situational constraints. These differences seem to arise from deep cultural differences in the way the person is conceptualized and in how the relations between individuals and their social networks are understood.
The different theories about human nature held by Americans and East Asians color not only their understanding of relatively minor transgressions of the sort reported for acquaintances in Miller’s study but also their explanations of far more deviant acts performed by strangers. Michael Morris and Kaiping Peng (1994) examined how American and Chinese individuals explain the actions of a notorious murderer. In the autumn of 1991 there were two similar, highly publicized cases of murder in the United States. In one, a Chinese graduate student at a midwestern university, who had unsuccessfully appealed his loss of an award competition and had subsequently failed to get an academic job, shot and killed his advisor, several other people, and, finally, himself. In the other case, an Irish-American postal worker in Detroit, who had lost his job, had unsuccessfully appealed the decision, and had failed to find an alternative job, shot and killed his supervisor, several other people, and, finally, himself.
Morris and Peng compared the explanations offered for these murders in an English-language newspaper (The New York Times) and a Chinese-language newspaper published in the United States (World News). All articles published in these newspapers about these two crimes were coded for the presence of dispositional and situational attributions. For the most part, the English articles were more likely than the Chinese ones to focus on both murderers’ traits (e.g., “he had a short fuse”), attitudes (e.g., “personal belief that guns were an important means to redress grievances”), and psychological problems (e.g., “darkly disturbed man who drove himself to success and destruction”). In contrast, the Chinese articles were more likely than the English ones to focus on the murderer’s interpersonal relationships (e.g., “did not get along with his advisor”), on circumstances (e.g., “had been recently fired”), on problems with Chinese society (e.g., “tragedy reflects lack of religion in Chinese culture”), and on aspects of American society (e.g., “murder can be traced to the availability of gun”). In short, Westerners favored dispositional accounts whereas Easterners favored situational ones.
The same pattern emerged in a subsequent study, in which Chinese and American-born graduate students at the University of Michigan read accounts of these two murders and rated the extent to which each shooting had been caused by a series of dispositional and situational factors which had been gleaned from the various newspaper articles (Morris and Peng 1994). The American students, on average, viewed dispositional explanations focusing on the murderer’s pathological character as more causally important than did the Chinese students. In contrast, the Chinese students viewed situational explanations focusing on American culture and circumstances (America’s individualistic values and violent movies, the recession) as more causally important than did the Americans.
These studies suggest that Westerners and Easterners subscribe to strikingly different theories about human nature, and differ markedly in their understanding of the causal underpinnings of social behavior. However, studies that involve comparisons among individuals who live in different countries, who use different languages, and who are governed by different social norms are open to a far less interesting, trivializing interpretation. Perhaps the individuals in the two cultures did not differ in their theories about human nature but, instead, differed in their understanding of the meaning and intent of the questions used by the researchers to tap their theories. Easterners, then, may hold the same theories about the causes of behavior as Westerners, but the questions they have been asked, once translated into their language and imported into their cultural context, have not adequately tapped their theories. It would be difficult to sustain this trivializing interpretation of cultural differences in attribution if it could be shown that members of the two cultures do not differ in their responses when asked about events for which they hold similar theories. Although Chinese and Americans differ markedly in their theories of social events, they should not differ in their theories about physical events. Therefore, if the two groups understand the attribution questions in the same way, they should provide comparable explanations for physical events even as they differ in their explanations of social events.
Following this logic, Morris and Peng (1994) created two sets of animated cartoons, one depicting social events, the other depicting physical events. The social cartoons all portrayed various interactions among a group of fish. Each showed a blue fish moving in different ways in relation to a group of other fish. In one cartoon, for example, the blue fish moved toward the group and then continued to swim along with it. In another, the group moved toward the blue fish, and the two parties than continued to swim separately. The physical cartoons all portrayed a ball-like object moving across a soccer field. In one cartoon, for example, the moving object stopped, started, and stopped again. In another, it gradually slowed down as if by friction.
The two sets of cartoons were shown to high school students in China and in the United States who were asked to rate the extent to which the movements of the blue fish (in the social scenarios) or of the round object (in the physical scenarios) were influenced by internal factors (e.g., hunger, for the fish, and internal pressure, for the ball) and the extent to which they were influenced by external factors (e.g., the other fish, for the social scenarios, and a person kicking the ball, for the physical ones). This methodology permitted cross-cultural comparisons of attributions about social and physical events. It also had the additional advantage of ensuring that, for both types of events, members of the two cultures were explaining identical scenarios.
Responses to the social events replicated the by now familiar pattern of cultural differences: On average, the Chinese respondents viewed internal factors as less influential than did the Americans and viewed the external factors as more influential than did the Americans. In contrast, the Americans and the Chinese shared the same understanding of the causes of physical events: On average, the two groups did not differ in the extent to which they viewed the internal factors or the external factors as influential.
The Chinese and American students, then, did not differ in their understanding of the questions posed to them by the investigators, nor did they differ in their views of the very nature of causality. Where they held similar theories, as they did for the physical events, they favored the same explanations. The differences in their preferred explanations of social events, therefore, can be assumed to reveal differences in their underlying theories about the nature of social events.
All these studies point to the same conclusion. Westerners tend to view social behavior as driven by internal, stable dispositions such as traits and attitudes. In contrast, Easterners tend to view social behavior as determined by the individual’s interpersonal relations, roles, circumstances, and cultural milieu. This conclusion has important implications for the understanding of the fundamental attribution error, namely people’s tendency to overestimate the role of dispositions in causing behavior and to underestimate the role of situations. The fundamental attribution error may be fundamental only in Western cultures, where the person is viewed as autonomous, independent, and separate from the surrounding environment. Members of Eastern cultures, where the person is viewed as intricately linked to a web of social relations and as interdependent with others, may show no similar tendency to underestimate the power of the situation.
The hypothesis that East Asians will be less prone than Americans to make the fundamental attribution error was tested in several studies (for a review, see Fiske et al. 1998). In the classic fundamental-attribution-error paradigm, participants observe a person read an essay that either supports or opposes a particular position, say a pro-Castro or an anti-Castro essay. Americans typically assume that the essay reflects the attitudes of the person who read it, even when it is clear that the person had been required to read that particular essay; for example, they believe a person who had been required to read a pro-Castro essay feels more positively about Castro than does a person who had been required to read an anti-Castro essay (for a review, see Jones 1990). In other words, they fail to appreciate the extent to which the person’s actions were influenced by the situation, namely the requirements to read that particular essay. It now appears that Japanese and Korean students also make this error in the original, unmodified paradigm. However, when the paradigm is modified so that the situational constraints are highlighted, Americans continue to make the error, whereas East Asians do not (Choi and Nisbett 1998). These studies suggest that East Asians do sometimes explain behavior in terms of dispositions, much like Westerners do. However, East Asians are more likely than North Americans to pick up on cues pointing to the importance of situational constraints. When the situation is highlighted, they set aside any dispositional explanations that they may have been entertaining in favor of more appropriate situational ones.
It appears, then, that cultural differences between Western and Eastern societies result in marked differences in how people represent, understand, and think about other people and themselves. The fundamental differences in how these two cultures view human nature also appear to influence the kinds of motives that their members strive to fulfill and the social structures through which the cultures sustain these motives....
Excerpted from Social Cognition: Making Sense of People, Chapter 11, pp. 515-533. © 1999 by The MIT Press, Cambridge, Mass. Reprinted by permission.