Cultural influences on personality.

Key Words individualism, collectivism, Big Five, indigenous psychologies, universals

* Abstract Ecologies shape cultures; cultures influence the development of personalities. There are both universal and culture-specific aspects of variation in personality. Some culture-specific aspects correspond to cultural syndromes such as complexity, tightness, individualism, and collectivism. A large body of literature suggests that the Big Five personality factors emerge in various cultures. However, caution is required in arguing for such universality, because most studies have not included emic (culture-specific) traits and have not studied samples that are extremely different in culture from Western samples.

INTRODUCTION

Recent Annual Reviews of Psychology have had chapters dealing with personality (Wiggins & Pincus 1992, Magnusson & Toerestad 1993, Revelle 1995) and with culture (Shweder & Sullivan 1993, Bond & Smith 1996a, Cooper & Denner 1998), but not with both culture and personality. The culture and personality topic is controversial. Bruner (1974) assessed the field as a "magnificent failure." Shweder (1991) saw little that can be considered positive in this field. For instance, Shweder argued that (a) individual differences in conduct are narrowly context dependent and do not generalize across contexts. Thus, global traits do not exist. Shweder further argued that (b) early childcare practices per se do not have predictable consequences for adult character, (c) the greater the cultural variation, the smaller is the situational comparability, and that (d) "objective" conditions, such as reinforcers and other "external" stimulus events, do not predict the accommodation of an organism to its environment.

More positive evaluations have emerged recently (e.g., Lee et al. 1999a). Lee et al. (1999b) edited a book that vigorously defended the utility of culture and personality studies, summarized the history of this topic, and provided chapters about Mexican, Chinese, African, German, Indian, and Japanese personality, as well as studies for the improvement of interaction across cultures. Piker (1998) thought that Shweder's objections to previous work employed "straw dummy tactics" (p. 21).

McCrae and his colleagues (McCrae 2000, McCrae et al. 2000) also presented a view diametrically opposite to Shweder (1991). According to McCrae et al., global traits do exist. They claim that "studies of heritability, limited parental influence, structural invariance across cultures and species, and temporal stability all point to the notion that personality traits are more expressions of human biology than products of life experience" (p. 177). This view places too much emphasis on biology, and more balanced assessments of the influence of genes and environment (e.g., Maccoby 2000) suggest that personality corresponds to the area of a quadrangle, one side of which is genes and the other, environment. In short, personality emerges under the influence of both genes and environment. Furthermore, behavior is likely to be a function of not only culture and personality but also the interaction between personality and the situation. We review studies where most of the variance in behavior is a function of such interactions.

In any case, McCrae et al. (2000) argued that there are basic tendencies (neuroticism, extraversion, openness, agreeableness, and conscientiousness) that are independent of culture. For example, animal psychologists (e.g., Gosling & John 1999) have identified personality traits (such as extroversion and dominance) in some higher animals, so at a basic level, contrary to Shweder's view, such traits are likely to exist (see also Munroe 1999). Many of Shweder's other points can be criticized in similar ways, leading us to agree with Piker's (1998) comments.

Shweder (1991) proposed that cultural psychology would provide the way to think about culture and personality. He recommended "thick description" of the cultural practices without attempts at generalizations. However, the facts of phylogenetic continuity make generalizations possible. We can examine universals across cultures while admitting that the meaning that individuals give to a particular event may differ from culture to culture and must be incorporated in our understanding of the way culture is related to individual differences in behavior. Thus, in this chapter we take a position that is intermediate between Shweder (1991) and McCrae et al. (2000). We look for universal generalizations, while at the same time admitting emic (culture-specific) information.

Personality is shaped by both genetic and environmental influences. Among the most important of the latter are cultural influences. Culture is transmitted through language and the modeling of behavior when conditions permit humans to communicate through shared language, by living in the same historic period, and when they are sufficiently proximal to influence each other. The overarching model of cultural influences on personality that we have adopted in this chapter is that though biological factors have an important role in shaping personality, they do not account for most of the variance. Ecology, among other factors, shapes the culture, which in turn shapes the socialization patterns, which shape some of the variance of personality (Maccoby 2000). For example, Rohner (1986, 1999) has shown reliable links between socialization practices and personality. Both within and between cultures when parents accept their children (there is much hugging, comforting), the children become sociable, emotionally stable, have high self-esteem, feel self-adequate, and have a positive world view. When parents are rejecting (hitting, using sarcastic language, humiliating, neglecting), their children become adults who are hostile, unresponsive, unstable, immaturely dependent, and have impaired self-esteem and a negative world view.

Of course, historical factors and cultural diffusion also shape cultures, but limitations of space preclude their discussion. Broad empirical support for such a model does exist (e.g., Singelis & Brown 1995). In addition to these factors, we consider other constructs that are needed for a better understanding of the way culture influences personality.

SOME DEFINITIONS

Culture

The conceptualization of culture is by no means a simple matter. One possible way to think about culture is that "culture is to society what memory is to individuals" (Kluckhohn 1954). It includes what has worked in the experience of a society, so that it was worth transmitting to future generations. Sperber (1996) used the analogy of an epidemic. A useful idea (e.g., how to make a tool) is adopted by more and more people and becomes an element of culture (Campbell 1965). Barkow et al. (1992) distinguished three kinds of culture: metaculture, evoked culture, and epidemiological culture. They argue that "psychology underlies culture and society, and biological evolution underlies psychology" (p. 635). The biology that has been common to all humans as a species distinguishable from other species, results in a "metaculture" that corresponds to panhuman mental contents and organization. Biology in different ecologies results in "evoked culture" (e.g., hot climate leads to light clothing), which reflects domain-specific mechanisms that are triggered by local circumstances, and leads to within-group similarities and between-groups differences. What Sperber describes, Barkow et al. call "epidemiological culture."

Elements of culture are shared standard operating procedures, unstated assumptions, tools, norms, values, habits about sampling the environment, and the like. Because perception and cognition depend on the information that is sampled from the environment and are fundamental psychological processes, this culturally influenced sampling of information is of particular interest to psychologists. Cultures develop conventions for sampling information and determine how much to weigh the sampled elements from the environment (Triandis 1989). For example, people in hierarchical cultures are more likely to sample clues about hierarchy than clues about aesthetics. Triandis (1989) argued that people in individualist cultures, such as those of North and Western Europe and North America, sample with high probability elements of the personal self (e.g., "I am busy, I am kind"). People from collectivist cultures, such as those of Asia, Africa, and South America, tend to sample mostly elements of the collective self (e.g., "my family thinks I am too busy, my co-workers think I am kind") (Triandis et al. 1990, Trafimow et al. 1991).

Personality

Funder (1997) defined personality as "an individual's characteristic pattern of thought, emotion, and behavior, together with the psychological mechanisms--hidden or not--behind those patterns" (pp. 1-2). Characteristic sampling of the information in the environment, which corresponds to the sampling that occurs in different cultures, can be one of the bases of individual differences in personality.

Personality may also be conceptualized as a configuration of cognitions, emotions, and habits activated when situations stimulate their expression. Generally, they determine the individual's unique adjustment to the world. This view is supported by data that indicate the importance of the situation. For example, the authoritarian personality is characterized by submission to authorities, aggression toward people who are different, and conventionalism (Pettigrew 1999). Interestingly, Russians who are high on this trait reject laissez-faire individualism, whereas Americans who are high on this trait support this type of individualism (McFarland et al. 1992). Rejection of individualism is consistent with Russian conventionalism, whereas support for individualism is consistent with American conventionalism.

Level of Analysis

Studies that use culture as the N can provide different results than studies that use individuals as the N. Thus, below we attempt to make explicit the level of analysis that was used in a particular study.

Indigenous, Cultural, and Cross-Cultural Psychologies

Many theoretical perspectives are used when studying the relationship between culture and psychology (Cooper & Denner 1998). The most important are the indigenous, cultural, and cross-cultural perspectives. The differences in perspectives have implications for the methodology that is likely to be used in studying personality. For example, personality tests developed in one culture and translated for use in other cultures are likely to be insensitive to cultural differences and to produce distorted results (Greenfield 1997). Cultural and indigenous psychologists do not use such tests; they use mostly ethnographic methods. Cross-cultural psychologists attempt to measure the same construct equivalently in each culture with culturally sensitive methods.

Triandis (2000b) outlined several differences among these approaches and argued that all three are needed. Converging findings using these approaches are most likely to be reliable and valid. This is also the view of Marsella et al. (2000) who, after an excellent review of the history of culture and personality studies, emphasized the use of qualitative (ethnosemantic) methods in conjunction with quantitative methods. The ethnosemantic methods include (a) the elicitation of all personality terms in the particular language, (b) the organization by research participants of the terms into naturally occurring structures, (c) the derivation of the meanings (e.g., spontaneous associations) of these structures, and (d) the linking of the terms to actual behaviors. For example, researchers might use the antecedent-consequent method (Triandis 1972) ("If one is Y then one `would' or `would not' do X") to determine the link between personality terms and behaviors in different cultures. It is very likely that the emic structures obtained with these methods will have some resemblance to the etic structures obtained by Western methods. Finding such convergence allows us to compare personalities across cultures (using the etic dimensions) and also describe personalities with culturally sensitive elements (using the emic dimensions).

Church & Lonner (1998) edited a special issue of the Journal of Cross-Cultural Psychology, which utilized this convergent point of view. They included papers that linked personality and culture from the perspective of cultural (Markus & Kitayama 1998), indigenous (Ho 1998), and evolutionary psychology (MacDonald 1998). Church (2000) has provided an impressive model of culture and personality that integrated many of these approaches, especially the trait and cultural psychological approaches. According to the model, traits exist in all cultures, but account for behavior less in collectivist than in individualist cultures. Situational determinants of behavior are important universally, but more so in collectivist than in individualist cultures. Cognitive consistency among psychological processes and between psychological processes and behavior occurs universally, but is less important in collectivist than in individualist cultures.

Ecology to Culture Links

Ecology (terrain, climate, flora and fauna, natural resources) is linked to the maintenance …