Culture and Applied Nonverbal Communication

Culture and Applied Nonverbal Communication

As this volume has indicated there is now a rich literature examin­ing the nonverbal behaviors in many applied settings. These stud­ies continue to document exactly how important nonverbal behaviors are in real life situations and that they have real life con-sequences as well.

In addition to this literature there is a large basic research litera­ture examining the influence of culture on nonverbal behaviors. This literature is important because it informs us of the ways in which nonverbal behaviors and communication processes in general can be similar and different across cultures. They provide a platform by which many basic studies of nonverbal behaviors and communica­tion have occurred in the past, and will occur in the future.

The domain of cross-cultural research on applied nonverbal be­havior, however, is in its infancy, and to date there have only been a handful of studies that have been published in peer-reviewed jour­nals. The goal of this chapter is to encourage such research to blos­som. To do so we first discuss a conceptual understanding and definition of culture, and then how culture influences the encoding and decoding of nonverbal behaviors. We then discuss several meth­odological issues concerning cross-cultural research that research­ers should be aware of. At the end of this chapter we briefly describe an example of an applied study of nonverbal behaviors from our labo ratory. We hope that this information becomes some of the nutrients needed for future research to take root and grow.

WHAT IS   CULTURE? Human Nature

In order to understand and define culture it is inevitable to start with some assumptions about human nature. The view of human nature that provides the best platform to account for not only pancultural uni­versals but also culture-specifics is that of evolutionary psychology. This perspective suggests that people have evolved a set of motives and strivings that are ultimately related to reproductive success (Buss, 2001). Reproductive success and other biological functions such as eating and sleeping are biological imperatives if people are to survive.

In the evolutionary psychology perspective survival is related to the degree to which people can adapt to their environments and to the con-texts in which they live. Over the history of time people must have had to solve a host of distinct social problems in order to adapt and thus achieve reproductive success. These social problems include negotiat­ing complex status hierarchies, forming successful work and social groups, attracting mates, fighting off potential rivals of food and sexual partners, giving birth and raising children, and battling nature (Buss, 1988, 1989, 1991, 2000, 2001). In fact we need to do these things in our everyday lives today as well. Thus universal biological imperatives have become associated with a universal set of psychological problems that people need to solve in order to survive.

That is, all individuals and groups of individuals have a universal problem of how to adapt to their environments in order to deal with their universal biological needs and functions and the imperative of re-productive success. Thus all individuals and groups of individuals must create ways to deal with these universal problems. These ways can be very specific to each group, because the context in which each group lives-the physical environment, the social factors, and the types and sizes of their families and communities-are different. The ways that each group develops then become each group's culture.

Culture

Culture is created as people have adapted to their environments in or-der to survive. In our view, culture is the product of the interaction be­tween universal biological needs and functions, universal social problems created to address those needs, and the context in which people live. Culture results from the process of individuals' attempts to adapt to their contexts in addressing the universal social problems and biological needs.

In the past there have been many attempts at defining exactly what those biological and social needs are, and the aspects of culture that address them. For example, Malinowski suggested that all individuals had universal basic needs related to metabolism, reproduction, bodily

Similarly social scientists have been interested in culture and how it influences people for well over 100 years. Consequently there have been many different definitions of culture over the years, with similari­ties as well as differences (Berry, Poortinga, Segall, & Dasen, 1992; Jahoda, 1984; Kroeber & Kluckholn, 1952/1963; Linton, 1936; Rohner, 1984; Triandis, 1972). In our work we define culture simply as a shared system of socially transmitted behavior that describe, define, and guide people's ways of W.

The Characteristics of Culture

Culture touches on all aspects of our lives. It involves subjective and ob­jective elements (Triandis, 1972). It explains differences in the types of foods we eat and how we eat them. It explains the clothes we wear and our home life. We use culture to describe our activities, values, attitudes, opinions, and beliefs, and to describe our communities, religion, and even our government. We use culture to explain similarities within and differences between groups of people (Tooby & Cosmides, 1992).

The subjective elements of culture are psychological. Culture influ­ences many psychological processes, such as attitudes, beliefs, norms, opinions, values, and behaviors. Culture in this sense is like a syn­drome, a constellation of separate but interrelated psychological com­ponents that collectively characterize a condition (Triandis, 1994).

Culture is always changing, even slowly. It is not a static entity, but a living, breathing one. at we commonly know as 'the generation gap' is a cultural difference as it refers to different ways of life and being for people who are raised in different periods of time (Pipher, 1998). Many

Malinowski's Conceptualization of Basic Needs and Cultural Responses

Basic Needs                               Cultural Response

countries around the world including the U.S. have undergone cultural changes across time and will continue to do so in the future as our ways of life change.

Culture exists on multiple levels. Individuals are part of small groups, and smaller groups are part of larger and even larger groups. Each group can have its own culture and in this way culture can exist on many levels. This is true for different ethnic and community groups that live in a large country like the U.S., as well as among different de­partments, sections, and work units of large companies.

Culture enhances survival. Cultures provide rules for living, tell people how to interact, and how to work and play with each other. Culture provides a hierarchy for decision-making and sets the stan­dards for group cooperation and divisions of labor. With culture there is some order to life; without culture there is chaos. Even peo ple who think they have no culture have a culture; it is just the culture to believe they have no culture. Culture is often difficult to perceive because we do not recognize alternative possibilities without having experienced them. That's why people learn as much about their own culture as they do about others when they travel to or live in new cul tures. Of all the possible things people could do, culture helps to limit what we should do in order to survive in the environment in which we live (Poortinga, 1990).

Culture is communicated across generations. This ensures that many aspects of culture are durable. Beliefs and attitudes that be-come popular from time to time and that are shared by many people may be what we know of as 'popular culture,' but the culture we are concerned with here is more stable across time. Many elements of culture are communicated across generations, including the rules we learned when we were children, the holidays and cultural activities we celebrate, and the foods we eat at home.

Culture both enables behavior, allowing it to be created or in-vented, while at the same time it constrains or restricts behavior (Adamopoulous & Lonner, 2001). On one hand, individualism, for example, fosters uniqueness, independence, autonomy, and creativ­ity. It provides the cultural framework within which behaviors can be invented. Jazz musicians, writers, poets, artists, rock stars and even disciplines like psychology can thrive and flourish in such an environment (Buss, 2001). On the other hand, culture also provides rules for constraining behavior. Laws exist in every country and cul ture of the world, and these laws define what is right and wrong, ac­ceptable and not, in every land. Cultures also provide for social sanctions against inappropriate behavior. In many cultures, for in-stance, shame is used as a powerful and important social sanction that limits behavior and keeps everyone in line. In many Asian cul­tures the concept of 'face' is important, and keeping and protecting one's face is as important as invention is in individualistic cultures (Oetzel et al., 2001).

Universal and Culture-Specific Psychological Processes

While cultures can be unique to the groups of individuals that live in them and the contexts in which they live, they all must deal with the same set of biological needs and functions and universal social prob lems. Thus it is very possible and in many cases very likely that the ways in which they are addressed are the same, even though the cul tures may be different. That is, universal biological needs and social problems can lead to similar solutions across cultures, especially over time in our evolutionary history.

For this reason many aspects of our psychology-our mental pro­cesses and behaviors-are universal, that is, common to all people of all cultures and backgrounds. For example all humans appear to have some degree of specific fears, such as to snakes, spiders, heights, and darkness because these types of fears have led in our evolutionary history to greater probability of survival (Seligman & Hager, 1972, cited in Buss, 2001). All people have a tendency to per ceive their own ingroup as heterogeneous, fully recognizing the indi vidual differences that exist in that group, while they perceive other groups as more homogeneous, assuming less diversity within the group (Linville & Jones, 1980; Triandis, McCusker, & Hui, 1990). People also seem to have a natural proclivity to fears of strangers and outgroup members, which may be a universal basis for ethnocentrism, prejudice, aggression, and even war (Buss, 2001). The differences in how we treat ingroup and outgroup members are likely rooted in our evolutionary history because such distinctions were useful in the past to our reproductive success. Other universal psychological processes, such as incest avoidance, facial expres­sions of emotion, division of labor by sex, revenge and retaliation, mate selection and sexual jealousy, self-enhancement, and person­ality can be traced to the core aspect of a universal human nature based on biological imperatives and universal social problems of adaptation and living.

But many psychological processes are also culture specific. Differ ent cultures have developed different ways of dealing with the biologi cal imperatives and universal social problems based on their contexts. Language is a good example of a very culture-specific behavior. Each culture has its own language, with its own set of vocabulary, syntax, grammar, phonology, and pragmatics. The need to have language may be a pancultural universal problem; and having a language may be a universal solution to this problem. But the specific way in which each culture solves this problem-that is develops its own language-is different in every culture.

Culture is a pretty fuzzy construct with a pretty fuzzy definition. There are no hard and fast rules of how to determine what a culture is or who belongs to that culture. But its influence on psychology and nonverbal behavior cannot be denied.

CULTURE AND NONVERBAL BEHAVIORS

Culture has a pervasive and profound influence on verbal and nonver­bal encoding and decoding processes, which we discuss in this section.

Cultural Influences on Encoding Nonverbal Behaviors

Culture exerts considerable influence over the verbal languages that we speak, from the syntax of a language to its pragmatics. But just as culture influences our verbal languages, culture also exerts consider-able influence over our nonverbal languages. People of all cultures learn to use nonverbal behaviors-facial expressions, gestures, dis­tance, gaze, and postures-as part of their communication repertoire, but people in each culture learn to use them in different ways. All hu mans are born with the capacity to form all types of sounds; culture dictates how we shape and mold those sounds into particular lan­guages. In the same way, culture shapes and molds nonverbal behav­iors into each culture's nonverbal language.

Some kinds of nonverbal behaviors are common to many cultures, such as greeting behaviors (for example, the eyebrow raise), whereas others differ radically (for example, touching behaviors; Keating, 1976). Developmental research has suggested that rules governing nonverbal behavior are as old as verbal languages, and that children learn their cultural rules governing nonverbal behaviors as they learn the rules of vocal expression and acquire verbal language (Von-Raffler Engel, 1981). If this is the case, it is no wonder that the cultural rules of nonverbal behavior are well ingrained in us by the time we are adults, and that we use them without much second thought.

There are many examples in the literature of cultural differences in encoding nonverbal behaviors. For instance, the universality of facial expressions of emotion is no longer debated in psychology; people all around the world, despite differences in culture, have the ability to ex press anger, contempt, disgust, fear, happiness, sadness, and surprise in the same ways (Ekman, 1999; Matsumoto, 2001). Yet all people learn rules that govern how to manage and modify these universal emotional expressions based on social circumstances. These rules, called cultural display rules (Ekman & Friesen, 1969), are learned early on and are an important part of the socialization and encultur ation process (Saarni, 1979). We know that compared to Americans, Japanese, Poles, and Hungarians are likely to express more positive and less negative emotions toward ingroup members, and more nega­tive and less positive emotions to outgroups (Biehl, Matsumoto, & Kasri, in press; Matsumoto, 1990). Also compared to Americans, Ger­mans tend to minimize their expressions of negative emotions by deamplifying or neutralizing them, but not by qualifying them with a smile (Koopmann & Matsumoto, 2003).

Emblems and gestures also differ across cultures (Morris, Collett, Marsh, & O'Shaughnessy, 1980), as when a Japanese person nods his or her 'yes' and says 'hai' (literally, yes) and does not necessarily mean 'yes.' There are cultural differences in gaze and visual attention, touch­ing, and interpersonal space as well. Well known is the 'diplomatic dance that occurs when Americans are uncomfortable when they inter-act with some peoples of middle eastern cultures, who have learned to interact with others at a distance at which they can feel your breath.

Cultural Influences on Decoding Nonverbal Behaviors

Culture affects the decoding process in several ways. Here we summa­rize three sets of psychological processes related to decoding that are affected by culture.

Cultural Filters, Ethnocentrism, Emotions, and Value Judgments. As we grow up, we learn cultural rules of appropriate communicative encod­ing with respect to both verbal and nonverbal behaviors. When we are little, these rules are constantly reinforced by parents, friends, teach­ers, and other enculturation agents. Many rules are also transmitted and reinforced by organizations and institutions (as in our study of language through the school system). As we get older, we need to be re-minded less about these rules, and their use requires less conscious effort. The inevitable result is unique, culture-specific ways in which communication-verbal and nonverbal-occurs.

As we grow, we also learn how to perceive signals and interpret mes­sages; that is, we learn cultural rules of appropriate decoding as well. Because we share a set of encoding and decoding rules with people of our culture, we develop a st of expectations about communication. These rules and expectations form a basis of tacit understanding that need not be spoken each time we, as adult members of the same cul­ture, communicate with one another.

Not only do we have certain expectations about the communication process; we have also learned emotional reactions associated with those expectations. These reactions can range from acceptance and pleasure to outrage, hostility, and frustration. Our emotions, in turn, are intimately tied to value judgments, which we often make without a second thought. These judgments seem only natural because they are rooted in our upbringing; they are the only types of judgments we have learned to make. Emotions and values serve as guidelines in helping us form opinions about others and ourselves.

A recent set of studies in our laboratory highlights these relationships (Matsumoto, Choi, Hirayama, Domae, & Yamaguchi, 2003). Across three studies American and Japanese observers were shown either neu­tral, low intensity or high intensity emotional expressions and were asked to judge how strong the external display of the expressor was, and how much emotion they really thought the expressor was feeling. These

are, in effect, judgments of other's display rules, as observers made judgments of how much emotion was expressed in relation to how much they thought was actually felt. In addition the observers com­pleted measures of their own display rules or emotion regulation pro cesses. In all three studies there were significant culture by rating type interactions. For instance, there was no difference between the two rat­ing types on neutral expressions for Americans while the Japanese rated internal experience higher than external display. In every instance in which a culture by rating type interaction occurred, the difference dis appeared when the observers' own display rules or emotion regulation scores were controlled, indicating that the differences occurred entirely because of differences in display rules.

Thus, decoding rules, and their associated emotions and value judgments, form the basis of the 'filters' that we use in seeing the world. As we become more enculturated, we add more layers to those filters. These filters are like lenses that allow us to perceive the world in a certain way. By the time we are adults, we share the same filters with others in our cultural group. They become part of our self, insepa rable and invisible, and are a normal part of our psychological compo­sition because of the way we have been enculturated.

Culture and Stereotypes. Stereotypes are generalizations about peo­ple, particularly about their underlying psychological characteristics or personality traits. Stereotypes are inevitable products of normal psy­chological processes, including selective attention, appraisal, concept formation and categorization, attributions, emotion, and memory. Ste reotypes are invaluable mental aids, helping us organize information about the world. They are important in helping us interact with others in our world, and are especially important in communication.

Stereotypes prime our expectations. We may selectively attend to events that support our stereotypes, and ignore, albeit unconsciously, events and situations that challenge them. Negative attributions may reinforce negative stereotypes. Even when we perceive events contrary to stereotype, we may convince ourselves that the stereotype is correct. Such dismissals can occur quickly, without much conscious thought or effort, and are resilient to emotion.

These psychological processes-including selective attention, attri­bution, and emotion-are all part of our self-concept. They reinforce the cultural knowledge we have learned from many years of encultur­ation, and thereby reinforce our sense of self. As we confirm our ste­reotypes, therefore, we reinforce our self-concept. Stereotypes are thus an integral part of the package of psychological processes, and are intimately tied to our emotions, values, and core self.

Culture and Social Cognition. Culture influences how we interpret the actions of others-that is, our attributions regarding others. Amer icans, for example, tend to draw inferences about other people's inter-

nal states or dispositions that supposedly underlie or even cause their behavior. This bias is known as fundamental attribution error (Ross, 1977). Cross-cultural research has shown that this bias may not exist in other cultures. For instance such dispositional explanations were common for Americans but much less so for the Hindus (Miller, 1984); the Hindus provided explanations in terms of the actor's duties, social roles, and other situation-specific characteristics (see also Shweder & Bourne, 1984). Other attributional tendencies, such as self-serving bias and defensive attributions are also manifested differently in dif­ferent cultures.

In summary, culture plays a large role in decoding signals during communication episodes-first, because of the close relationship be­tween cultural rules governing encoding and decoding, and second, because of cultural influences in the development of ethnocentrism, stereotyping, and social cognition. Cultural decoding rules are inti­mately associated with emotions and value judgments, which collectively form our self-concepts.

CONDUCTING APPLIED RESEARCH ON NONVERBAL BEHAVIOR
ACROSS CULTURES

In this section we consider methodological issues that underlie the conduct of cross-cultural research on nonverbal behaviors. There are many different types of cross-cultural studies and space limitations do not permit a full discussion of many of them; interested readers are re­ferred to Matsumoto (2003) for a fuller discussion of the problems and possible solutions. Here we focus exclusively on cross-cultural com­parisons, and only on a few issues within that, namely sampling, mea­surement, and data equivalence. We begin with a discussion of the concept of equivalence.

Equivalence (and Bias)

In reality, there are only a few issues specific to the conduct of cross-cultural research that set it apart from general experimentation; the same problems and solutions that are typically used to describe issues concerning experimental methodology in general can and should be applied to most, if not all, cross-cultural comparisons. Most of the is-sues raised in this section, therefore, with the notable exception of lan­guage issues, are generally true of 'good' experimentation in mono-cultural studies as well. Cross-cultural research, however, has been useful in highlighting them.

One concept that is of crucial importance in the conduct and evalua­tion of all aspects of cross-cultural comparison is that of equivalence and its corresponding construct, bias. (Bias is generally viewed as non-equivalence. For this reason, we will generally refer to one.) Equiv­alence in cross-cultural research can be defined as a state or condition

of similarity in conceptual meaning and empirical method between cultures that allows comparisons to be meaningful. In a strict sense, the greater the non-equivalence (thus bias) of any aspect of a cross-cul tural study in meaning or method across the cultures being compared, then the less meaningful the comparison. Lack of equivalence in a cross-cultural study creates the proverbial situation of comparing ap­ples and oranges. If and only if, however, the theoretical framework and hypotheses have generally equivalent meaning in the cultures be ing compared, and the methods of data collection, management, and analysis have equivalent meanings, only then are the results from that comparison meaningful.

Of course, this is true in any between-group comparison study. Still, it is important to remember that the perfectly equivalent cross-cul­tural study is an impossibility; there will always be some aspect of the comparison that is not perfectly equivalent to each other. Thus, it is probably more accurate to suggest that for cross-cultural compari­sons to be valid and meaningful, they have to be 'equivalent enough.' The difficult part of this concept, however, that frustrates students and researchers alike is that there is no direct method, no mathematical formula, no easy way, to determine what is 'equivalent enough.' Some-times a study may have a lot of little non-equivalences, but still be meaningful. Sometimes a study may have one fatal non-equivalence, and thus be meaningless. These issues differ from study to study, and we cannot tell you here what the fatal flaw will always be. As usual, ex­perience and conscientiousness are probably two of the largest teachers.

The issues described here, and in most descriptions of experimen tation, therefore, are the ideals. The closer to the ideals the study is, the more valid the comparison (of course, this may also mean that it is farther from reality).

Sampling Equivalence

Sampling Adequacy. Researchers need to insure that the partici­pants in their study are adequate representatives of the cultures that they are supposed to represent. More often than not researchers gen erally assume that people who happen to fit into the categorical label of culture as operationalized (e.g., by nationality) are 'good' representa tives of that particular culture. In doing so there is an unacceptable as­sumption of homogeneity among the participants with regard to culture that can, in its worse sense, only serve to perpetuate stereotyp ic impressions and interpretations based on the findings. When differ­ences are found, researchers assume that the differences are 'cultural' because they assume that the samples are representatives of culture.

While this issue is relatively straightforward and easy to under-stand, in practice it is extremely difficult to achieve. In its strictest

sense, proper addressing of this issue would require the following steps: (a) the researcher would have to be able to theoretically define exactly what the cultures are that are being tested; (b) the researcher would have to be able to access a pool of individuals from the larger population that embodied those characteristics; (c) the researcher would have to randomly sample from that larger population; and (d) the researcher would have to measure those social, cultural, and psy­chological characteristics in their participants and empirically dem­onstrate that their culture manipulations occurred as intended.

Unfortunately, this is a tall order that is not, and perhaps cannot, be filled currently because of the limitations to our abilities to theorize about, and subsequently measure, culture on the individual level, and our inability to randomly access all members of any given cultural pop ulation. Given that we cannot currently achieve this ideal, the real is sue facing researchers concerns the degree to which they understand how far from this ideal they are, and how much they use this informa­tion to temper their interpretations. In a practical sense, a sound cross-cultural comparison would entail the collection of data from multiple sites within the same cultural group, either in the same study or across studies, to demonstrate the replicability of a finding across different samples within the same culture.

Non-Cultural, Demographic Equivalence. Researchers need to insure that the differences they obtain in a study are due to culture and not to any other non-cultural demographic variables on which the samples may differ. That is, researchers need to make sure the samples they compare are equivalent on variables such as gender, age, SES, educa­tional level, religious orientation, geographic area (e.g., rural vs. urban), and such. If they are not equivalent on non-cultural, demo-graphic variables, then those variables on which they are not equiva­lent may confound the comparison.

The conceptual problem that arises in cross-cultural research, which is not as apparent in mono-cultural studies, is that some non cultural demographic characteristics are inextricably intertwined with culture such that researchers cannot hold them constant across sam­pies in a comparison. Religion is such an example. There are differ­ences in the meaning and practice of religions across cultures that make them inextricably bound to culture oftentimes. Holding religion constant across cultures does not address the issue, because being Catholic in the U.S. just does not mean the same thing as being Catho­lic in Japan or Malaysia. Randomly sampling without regard to reli­gion will result in samples that are different not only on culture, but also on religion (to the extent that one can separate the two's influ­ences). Thus, presumed cultural differences often reflect religious dif ferences across samples as well. The same is also true oftentimes for SES, as there are vast differences in SES across cultural samples from around the world.

Measurement Equivalence

Conceptual Equiualence. Researchers need to insure that the psy­chological variables being measured in their studies are conceptually equivalent across the cultures being compared. Different cultures may conceptually define a construct differently. Common examples of con­structs that have widely divergent meanings across cultures include such topics as intelligence, self-concept, personality, or emotion. Clearly, just because something has the same name in two or more cul­tures does not mean that it refers to the same thing in those cultures (Wittgenstein, 1953, cited in Poortinga, 1989). If a concept means dif­ferent things to people of different cultures, then there is a lack of equivalence in the definition of the construct, and comparisons of cul­tures based on non-equivalent constructs will lack meaning. Re-searchers wishing to compare cultures on psychological constructs, therefore, have the onus of demonstrating, either empirically or con­ceptually, that the constructs themselves are equivalent across the cul­tures being compared.

Empirical Equivalence. Even if a construct is conceptually equivalent across cultures, reliable and valid measurement of it may take differ­ent forms across cultures. Concretely this requires that researchers use measures that have been empirically demonstrated to reliably and validly measure the construct of interest in the cultures being studied. Simply taking an existing test developed in one culture and translating it for use in other cultures is not methodologically adequate, although this has often been the case previously. Cross-cultural validations of-ten require extensive testing in the target cultures in order to establish a reasonable amount of reliability and validity parameters, especially with regard to convergent and predictive validity. Questionnaires that involve multiple scales and items will need to have been tested to es­tablish the cross-cultural equivalence of item and scale meaning, espe­cially concerning equivalence in factor structures and item loadings.

For example, one applied cross-cultural study examined how emo­tion displays of French and American political leaders on TV affect vot­ers in France and the U.S., respectively (Masters & Sullivan, 1989). They showed videoclips of political leaders of one's own country displaying three types of emotion (happy, anger, fear) to French and American judges, and made ratings of the behavior of the political leaders, self-re­ports of their own emotional responses, and attitudes toward politics, leaders, and the media. The measures were back-translated (see be-low), and the French version was pretested to confirm that scales were used in the same way with French participants as it was the American participants, offering some evidence of convergent validity. Self-report of emotional response was factor analyzed separately for each culture and was found to have similar factor structures providing support that measures and procedure in both cultures were equivalent.

Still, factor equivalence is only one step in establishing the empiri cal equivalence of measures across cultures. These are not easy issues to deal with, and cross-validation is not as easy as it seems. Some writ ers have suggested that tests of psychological abilities are inherently incomparable across cultures. Greenfield, for example, argues that constructs such as intelligence and cognitive ability are inherently symbolic products of a culture (Greenfield, 1997). As such, the con­structs and tests of it presuppose a certain cultural framework in the first place in order to be valid. As these frameworks are not usually universally shared, cross-cultural comparisons of ability and intelligence therefore become meaningless.

Similar questions may exist concerning the equivalence in con­struct and operation of values. Peng and others, for example, have ar­gued that common methods for assessing values, which include providing participants with a list of values and asking them either to rate them or rank them in order of importance, may not be valid across cultures because of implicit social comparison processes (Peng, Nisbett, & Wong, 1997). They suggested that such methods may be in valid because of cultural differences in the meanings of specific value items, and because of the possibility that some value judgments are based on inherent social comparisons with others instead of making a direct inference about a private, personal value system. In order to in­vestigate this possibility, these researchers examined four different value survey methods, including the traditional ranking, rating, and attitude scaling procedures, as well as a behavioral scenario rating method. The only method that yielded reasonable validity estimates was the behavioral scenario rating method, which is the most unorthodox of all measures tested.

Poortinga has suggested that when a measure has high content va lidity in all cultures being tested (i.e., it has been shown to mean the same thing in all cultures), and when the construct being measured is in a psychological domain that is similar or identical across cultures (e.g., color schemes, pitch scale for tones), valid comparisons are gen­erally possible (Poortinga, 1989). When unobservable psychological traits and attributes of individuals are being measured, comparison may be possible as long as equivalence in the conceptual meaning of the psychological domain and its measurement in all participating cul tures have been established. Other than these two situations, all other research situations, according to Poortinga (1989), preclude valid comparison across cultures.

Linguistic Equivalence. Researchers need to insure that the research protocols used in their studies are linguistically equivalent across the cultures being compared. While most other methodological issues de scribed in this chapter pertain to all group difference research, mono or cross-cultural, this issue is one of the few that is specific to cross-cultural research.

Cross-cultural research often cannot be conducted solely in one lan­guage, because the samples being tested are frequently comprised of two or more distinct language groups. There are generally two proce­dures used to establish linguistic equivalence. One is known as back translation (Brislin, 1970), which involves taking the research proto­col in one language, translating it to the other language(s), and having someone else translate it back to the original. The second approach is to utilize the committee approach, in which several bilingual infor­mants collectively translate a research protocol into a target language on a consensual basis. Actually, a third approach is also available, which is a combination of the first two approaches.

Regardless of the approach, a major caveat for researchers here is that 'closest semantic equivalent' does not mean 'the same.' Getting protocols that are 'the same,' in fact, is probably impossible. Even if the words being used in the two languages are the agreed upon transla­tions, there is no guarantee that those words have exactly the same meanings, with the same nuances, across cultures. There is also the additional problem to deal with which concerns the difference between linguistic and cultural equivalence. That is, you can have a protocol that is linguistically equivalent to its original in another language, but that just does not make sense in the target language. In this case, the researcher needs to make a decision concerning whether to go with the literal translation, which may be awkward and difficult to interpret but is the closest semantic equivalent, or to go with the cultural transla­tion, which will make sense but is not linguistically equivalent.

Data Equivalence

Cultural Response Sets. When analyzing data, researchers need to be aware of the possible existence of cultural response sets, and if they do exist deal with them. Cultural response sets are tendencies for members of a culture to use certain parts of a scale when responding. For example, participants of culture A in a two-culture comparison may tend to use the entire scale, whereas participants of culture B may tend to use only a part of the scale (e.g., the middle). These tendencies may exist for several reasons, including cultural differences in atti­tudes and values regarding self-expression of personal opinions. There have been numerous suggestions in the past that members of collectivistic cultures hesitate using the extreme end points of a scale in congruence with a cultural hesitation to 'stick out,' resulting in the use of the middle of a scale. There have also been some studies that have shown tendencies for members of some cultural groups to use the endpoints. Bachman and O'Malley, for example, found such evi­dence in extreme response styles among African Americans (Bachman & O^'Malley, 1984), and Min and colleagues found similar evidence for Hispanics (Marin, Gamba, & Min, 1992). If they exist, cultural re­sponse sets may confound between-culture differences because it is

difficult to know whether differences are occurring because of re­sponse sets or because of 'meaningful' differences in real scores on the target variables of interest.

Effect Size Analyses. Cultural differences in mean values on any scale do not readily predict how individuals are different between cultures. Statistical significance does not mean 'practical' signifi­cance in a realistic or pragmatic sense, especially because statistical significance is so dependent on sample size. One mistake that re-searchers and consumers of research alike make when interpreting group differences is that they assume that most people of those groups differ in ways corresponding to the mean values. Thus, if a statistically significant difference is found between Americans and Japanese, for instance, on emotional expressivity such that Ameri­cans had statistically significantly higher scores than the Japanese, people often conclude that all Americans are more expressive than all Japanese. This, of course, is a mistake in interpretation that is fueled by the field's fascination and single-minded concern with sta­tistical significance and perhaps with cultural myths that are easy to perpetuate.

In reality, there are statistical procedures available that help to de termine the degree to which differences in mean values reflect mean­ingful differences among individuals. The general class of statistics that do this is called effect size statistics, and when used in a cross-cul tural setting, Matsumoto and his colleagues called them cultural effect size statistics (Matsumoto, Grissom, & Dinnel, 2001). It is beyond the scope of this chapter to present them in detail; Matsumoto et al. (2001) present four such statistics that they deemed most relevant for cross-cultural analyses, with reanalyses from two previously published studies as examples. Whether cross-cultural researchers use these or others, it is incumbent on them to include some kind of effect size anal ysis when comparing cultures so that informed readers can determine the degree to which the differences reported reflect meaningful differences among people.

Dealing With Non-Equivalent Data. Despite the best attempts to es­tablish equivalence in theory, hypothesis, method, and data manage­ment, cross-cultural comparisons are often inextricably, inherently, and inevitably non-equivalent. That is, it is impossible to create any cross-cultural study that means exactly the same thing to all partici­pating cultures, both conceptually and empirically. What cross-cul­tural researchers often end up with are best approximations of the closest equivalents in terms of theory and method in a study.

Thus, researchers are often faced with the question of how to deal with non-equivalent data. Poortinga (1989) outlined four different ways in which the problem of non-equivalence of cross-cultural data can be handled:

Preclude Comparison. The most conservative thing a researcher could do is to not make the comparison in the first place, concluding that such a comparison would be meaningless.

Reduce the Non-Equivalence in the Data. Many researchers engage in empirical steps to identify equivalent and non-equivalent parts of their methods, and then refocus their comparisons solely on the equivalent parts. For example, to compare perceptions of teacher immediacy in U.S. and Japan, Neuliep asked American and Japanese university students to complete the Verbal Immediacy Behaviors Scale, the Nonverbal Immedi­acy Measure, and various ratings of attitudes toward content of course, attitudes toward teacher, likelihood of using behaviors taught in class, likelihood of taking another class by same teacher, and own perception of how much they learned (Neuliep, 1997). The 20-item Verbal Immediacy Behaviors Scale and the 14-item Nonverbal Immediacy Measure were factor analyzed separately after each item was standardized within each culture to eliminate cultural differences. The factor analysis reduced the Verbal Immediacy Scale to 14 items and Nonverbal Immediacy Measure to 9 items and the scales with reduced items were analyzed in the study.

Interpret the Non-Equivalence. A third strategy is for the researcher to interpret the non-equivalence as an important piece of information concerning cultural differences.

Ignore the Non-Equivalence. While this is what most cross-cultural researchers should not do, this is in fact what many end up doing. Poortinga (1989) suggests that this is because many researchers hold onto beliefs concerning scale invariance across cultures, despite the lack of evidence to support such beliefs.

Obviously, how a researcher handles the interpretation of his or her data, given non-equivalence, is dependent on his or her experience and biases, and on the nature of the data and the findings. Because of the lack of equivalence in much cross-cultural research, researchers are often faced with many gray areas in interpreting findings from their cross-cultural studies. This is, of course, to be expected, because the study of culture is neither black nor white. Culture itself is a complex phenomenon that is replete with gray, and we see that in research every day and in the journals. It is the objective and experienced researcher who can deal with the gray area in creating sound, valid, and reliable interpretations that are justified on the basis of the data. And it is the astute consumer of that research who can sit back and judge those in­terpretations relative to the data in their own minds and not be swayed by the arguments of the researchers.

The Need for Unpackaging Studies

The field has come to increasingly recognize the limitations of the tra­ditional cross-cultural comparison in which two or more cultures are

CULTURE AND APPLIED NONVERBAL COMMUNICATION

compared on one or more target dependent variables. The problem with this approach is that 'culture' is really only a label that summa­rizes many concrete and specific aspects of a group's way of life. As such, it is impossible for us to know in a typical cross-cultural com parison exactly what about cultures produced the differences we ob­served, and why.

To address this issue, researchers have begun to identify specific, concrete, and measurable psychological variables that they believe represent at least some of the contents of culture most pertinent to their variables of interest and to include them in their cross-cultural comparisons. 'Culture,' then, as a global construct is replaced by these specific, measurable variables, which are called context vari­ables. Analyses are then directed to examine the degree to which these context variables actually account for the cultural differences. In this sense, the context variables are akin to nuisance variables in traditional experimentation, and the approach is exactly that of stud­ies of covariance, as the context (nuisance) variables are treated as covariates in Analyses of Covariance or hierarchical multiple regres­sion schemes. These types of cross-cultural studies are called unpackaging studies.

A number of examples of unpackaging studies can now be found in the literature. Bond and Tedeschi, for example, give an excellent re-view of cross-cultural studies on aggression both with and without unpackaging (Bond & Tedeschi, 2001). Singelis and his colleagues use the concept of self-construals to unpackage cultural influences on self-esteem and embarassability (Singelis, Bond, Sharkey, & Lai, 1999). Matsumoto and his colleagues have used the concepts of indi­vidualism-collectivism and status differentiation to unpackage cul­tural differences in cultural display rules and judgments of emotion. (Matsumoto et al., 2002; Matsumoto, Takeuchi, Andayani, Kouznetsova, & Krupp, 1998) Most recently, they have also shown that display rules of emotional expression mediate cultural differ­ences in judgments of the emotional regulation of others (Matsumoto et al., 2003).

Unpackaging studies force researchers to think about cultures in ways that they did not in the past, breaking them down to spe­cific, measurable constructs in considering how they influence the target variables of interest. Thus they force theoretical develop­ments in our understanding of culture. They also allow us to exam­ine the specific degree to which the hypothesized context variables actually do account for between culture differences. When they do not account for 100% of the differences, they force us to think about other ways in which culture influences our target constructs, helping us to refine our theoretical understanding of culture. We urge researchers interested in examining cultural effects on non-verbal behaviors to consider the design and conduct of unpack-aging studies.

APPLIED RESEARCH IN OUR LABORATORY: DOES THE ABILITY TO RECOGNIZE EMOTIONS PREDICT INTERCULTURAL ADJUSTMENT?

For several years our laboratory has been engaged in exploring the psy chological skills that predict successful intercultural adjustment. Our previous studies have focused on the development of the Intercultural Adjustment Potential Scale (ICAPS; Matsumoto, LeRoux, Bernhard, & Gray, 2001; Matsumoto et al., in press; Matsumoto, LeRoux, Ratzlaff et al., 2001), the only measure to date that can reliably and validly pre dict adjustment in a wide range of immigrants and sojourners. Across all studies the most important predictor of adjustment has consis­tently been shown to be emotion regulation (ER). We define ER as the ability of individuals to manage, modify, and use their emotions to-ward constructive outcomes. In fact we view ER as the 'gatekeeper' skill of adjustment-that ER is a necessary component of adjustment because the utilization of it allows for the use of other knowledge, skills, and abilities to help people navigate the trials and tribulations of intercultural adaptation.

The importance of emotion regulation in intercultural adjustment raises the possibility that emotional intelligence (EI) is important for adjustment because ER is considered to be a single component of EI (Mayer, Salovey, Caruso, & Sitarenios, 2001). EI, which is 'ability to recognize the meanings of emotions and their relationships and to use them as a basis in reasoning, problem solving and enhancing cognitive activities' (Mayer et al., 2001), is comprised of four skills; ER, emotion recognition in self and others, understanding of emotion, and utiliza tion of emotion to facilitate thinking (Ciarrochi, Chan, & Bajgar, 2001; Ciarrochi, Chan, & Caputi, 1999; Mayer et al., 2001). These four skills form a hierarchy, with emotion recognition in the bottom forming the basis for the rest of the skills (Izard, 2001; Mayer et al., 2001). In other words, the other three skills are only possible when emotion recogni­tion is first achieved. Therefore, emotion recognition ability (ERA) is conceptually a more primary construct than emotion regulation, because people need to recognize emotions before they can engage in emotion regulation.

Based on this notion, and based on the knowledge from previous studies that ER predicts intercultural adjustment, we hypothesized that ERA should predict intercultural adjustment also. If it does, we subsequently hypothesized that ERA mediates the relationship be­tween ER and adjustment.

To test these notions we recruited 63 international students at San Francisco State University within the first month of their first semester at the university. They completed six subjective measures of inter-cultural adjustment that included the Beck Anxiety Inventory (BAI; Beck & Steer, 1993), the Beck Depression Inventory (BDI; Beck, Steer, & Brown, 1996), the Beck Hopelessness Scale (BHS; Beck & Steer, 1988), the Culture Shock Questionnaire (CSQ; Mumford, 1998), the

Homesickness and Contentment Scale (HCS; Shin &Abell, 1999), and the Satisfaction with Life Scale (SWLS; Diener, Emmons, Larsen, & Griffin, 1985). In addition, they completed the ICAPS and the neuroticism scale of the Big Five Inventory (John, 1989) to measure ER. ERA was measured by using the Japanese and Caucasian Brief Af­feet Recognition Test (JACBART), in which expressions of seven uni­versal emotions (anger, disgust, fear, happiness, sadness, surprise, and contempt) are presented for 1/5 of a second embedded within a one second neutral expression of the same individual (Matsumoto et al., 2000).

The results confirmed our hypotheses. We performed a series of hi­erarchical multiple regressions on each of the adjustment scales, en­tering demographic variables that significantly predicted each adjustment scale on the first step (thus eliminating the effects of non-cultural demographic variables), and then JACBART ERA scores on the second. ERA added significant and unique variance to the predic­tion of four adjustment scales above and beyond that already pre­dicted by demographics: ERA of contempt predicted the BAI, ERA of sadness and disgust predicted culture shock, ERA of contempt pre­dicted homesickness, and ERA of sadness predicted contentment.

Because ERA was correlated with ER and ER was correlated with all adjustment indices, we next examined the degree to which ERA medi­ated the relationship between ER and adjustment by computing an-other series of hierarchical multiple regressions, entering demographics on the first step, significant ERA scores on the second, and then ER scores on the third. ER contributed unique variance to the prediction of all four adjustment scores predicted by E. Com­parison of the effect sizes associated with the predictive validity of ER on adjustment with and without ERA in the equation indicated that ERA accounted for between 9.09 and 19.26% of the association between ER and adjustment.

These results are the first to demonstrate that ERA can reliably pre­dict adjustment in immigrants and sojourners and are significant in their own right. Moreover because ERA is a trainable skill these find­ings open the door to the possibility that the potential for positive ad­justment outcomes may improve if immigrants and sojourners receive such training early on in their sojourns. We sincerely hope that future applied research may investigate these, and other, possibilities.

CONCLUSION

The goal of this chapter was to encourage cross-cultural research on applied nonverbal behaviors to take root, grow, and blossom, and to provide some of the basic information that could form some of the im portant nutrients that would allow this to occur. We first discussed a conceptual understanding and definition of culture, and then how cul ture influences the encoding and decoding of nonverbal behaviors. We

then discussed several methodological issues concerning cross-cul­tural research that researchers should be aware of. Immediately above we briefly describe an example of an applied study of nonverbal behav­iors from our laboratory.

Cross-cultural research in applied settings, like any good research in applied settings, is definitely not easy. Yet the potential rewards in information and knowledge and the possibilities of intervening posi tively in many people's lives, not only in one culture but in many, seem to be a benefit that far outweighs those difficulties. We sincerely hope that the information we have provided in this chapter wi11 encourage researchers in the field to conduct such studies in the future. As the world becomes increasingly smaller and the need for people across cultural lines to get along better becomes more and more apparent, the role of such research becomes increasingly larger.

REFERENCES

Adamopoulous, J., & Lonner, W. J. (2001). Culture and psychology at a cross-road: Historical perspective and theoretical analysis. In D. Matsumoto (Ed.), The Handbook of Culture and Psychology (pp. 11-34). New York: Oxford University Press.

Bachman, J. G., & O'Malley, P. M. (1984). Black-white differences in self-es teem: Are they affected by response styles? American Journal of Sociology, 90(3), 624-639.

Beck, A. T., & Steer, R. A. (1988). Beck Hopelessness Scale. San Antonio, TX: The Psychological Corporation.

Beck, A. T., & Steer, R. A. (1993). Beck Anxiety Inventory. San Antonio, TX: The Psychological Corporation.

Beck, A. T., Steer, R.A., & Brown, G. K. (1996). BDI-II: Beck Depression Inven‑

tory Manual (2nd ed.). San Antonio, TX: The Psychological Corporation. Berry, J. W, Poortinga, Y. H., Segall, M. H., & Dasen, P R. (1992). Cross-cul‑

tural psychology: Research and applications. New York: Cambridge Uni‑

versity Press.

Biehl, M., Matsumoto, D., & Kasri, F. (in press). Culture and emotion. In U. Gielen & A. L. Communian (Eds.), Cross-cultural and international di­mensions of psychology. Trieste, Italy: Edizioni Lint Trieste S.r.l.

Bond, M., & Tedeschi, J. (2001). Polishing the jade: A modest proposal for im­proving the study of social psychology across cultures. In D. Matsumoto (Ed.), The Handbook of Culture and Psychology (pp. 309-324). New York: Oxford University Press.

Brislin, R. (1970). Back translation for cross-cultural research. Journal of Cross-Cultural Psychology, 1, 185-216.

Buss, D. M. (1988). The evolution of human intrasexual competition: Tactics of

mate attraction. Journal of Personality & Social Psychology, 54(4), 616-628. Buss, D. M. (1989). Sex differences in human mate preferences: Evolutionary

hypotheses tested in 37 cultures. Behavioral & Brain Sciences, 12(1), 1-49. Buss, D. M. (1991). Evolutionary personality psychology. Annual Review of

Psychology, 42, 459-491.

Buss, D. M. (2000). The evolution of happiness. American Psychologist, 55(1), 15-23.

Buss, D. M. (2001). Human nature and culture: An evolutionary psychological perspective. Journal of Personality, 69(6), 955-978.

Ciarrochi, J., Chan, A. Y. C., & Bajgar, J. (2001). Measuring emotional intelligence in adolescents. Personality and Individual DIerences, 31(7), 1105-1119.

Ciarrochi, J. V., Chan, A. Y. C., & Caputi, P. (1999). A critical evaluation of the emotional intelligence construct. Personality and Individual Dierences, 28(3), 539-561.

Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction

with life scale. Journal of Personality Assessment, 49, 71-75.

Ekman, P (1999). Basic emotions. In T. D. a. T. Power (Ed.), The handbook of

cognition and emotion (pp. 45-60). Sussex, United Kingdom: John Wiley

and Sons, Ltd.

Ekman, P., & F^riesen, W. (1969). The repertoire of nonverbal behavior: Catego­ries, origins, usage, and coding. Semiotica, 1, 49-98.

Greenfield, M. P (1997). You can't take it with you. American Psychologist, 52, 1115-1124.

Izard, C. E. (2001). Emotional intelligence or adaptive emotions? Emotion, 3, 249-257.

Jahoda, G. (1984). Do we need a concept of culture? Journal of Cross-Cul­tural Psychology, 15(2), 139-151.

John, O. (1989). The BFI-54. Unpublished test; Institute of Personality and So‑

cial Research, Department of Psychology, University of California, Berkeley. Keating, C. (1976). Nonverbal aspects of communication. Topics in Culture

Learning, 4, 12-13.

Koopmann, B., & Matsumoto, D. (2003). American-German Differences in

Emotional Display Rules. Paper presented at the 2003 Western Psychologi‑

cal Association Annual Convention. Vancouver, British Columbia, Canada. Kroeber, A. L., & Kluckholn, C. (1952/1963). Culture: A critical review of con‑

cepts and definitions. Cambridge, MA: Harvard University.

Linton, R. (1936). The study of man: An introduction. New York: Appleton. Linville, P W, & Jones, E. E. (1980). Polarized appraisals of out-group mem‑

bers Journal of Personality and Social Psychology, 38, 689-703. Malinowski, B. (1927, 1961). Sex and repression in a savage society. Cleve‑

land, OH: World Publishers.

Malinowski, B. (1944, 1960). A scientc theory of culture and other essays. New York: Oxford University Press.

Min, G., Gamba, R. J., & Min, B. V. (1992). Extreme response style and ac-quiescence among Hispanics: The role of acculturation and education. Journal of Cross-Cultural Psychology, 23(4), 498-509.

Masters, R. D., & Sullivan, D. G. (1989). Nonverbal displays and political lead‑

ership in F^rance and the United States. Political Behavior, 11, 123-156. Matsumoto, D. (1990). Cultural similarities and differences in display rules.

Motivation & Emotion, 14(3), 195-214.

Matsumoto, D. (2001). Culture and emotion. In D. Matsumoto (Ed.), The Handbook of Culture and Psychology (pp. 171-194). New York: Oxford University Press.

Matsumoto, D. (2003). Cross-cultural research. In S. Davis (Ed.), The hand-book of research methods in experimental psychology (pp. 189-208). Ox-ford, UK: Blackwell.

Matsumoto, D., Choi, J. W, Hirayama, S., Domae, A., &Yamaguchi, S. (2003). Culture, display rules, emotion regulation, and emotion judgments. Manu­script currently submitted for publication.

Matsumoto, D., Consolacion, T., Yamada, H., Suzuki, R., Franklin, B., Paul, S., Ray, R., & Uchida, H. (2002). American-Japanese cultural differences in judgments of emotional expressions of different intensities. Cognition & Emotion, 16(6), 721-747.

Matsumoto, D., Grissom, R., & Dinnel, D. (2001). Do between-culture differ­ences really mean that people are different? A look at some measures of cul­tural effect size. Journal of Cross-Cultural Psychology, 32(4), 478-490.

Matsumoto, D., LeRoux, J. A., Bernhard, R., & Gray, H. (2001). Personality and behavioral correlates of intercultural adjustment potential. Manu­script submitted for publication.

Matsumoto, D., LeRoux, J. A., Iwamoto, M., Choi, J. W, Rogers, D., Tatani, H., & Uchida, H. (in press). The robustness of the Intercultural Adjustment Po­tential Scale (ICAPS). International Journal of Intercultural Relations.

Matsumoto, D., LeRoux, J. A., Ratzlaff, C., Tatani, H., Uchida, H., Kim, C., & Araki, S. (2001). Development and validation of a measure of intercultural adjustment potential in Japanese sojourners: The Intercultural Adjust­ment Potential Scale (ICAPS). International Journal of Intercultural Rela­tions, 1-28.

Matsumoto, D., LeRoux, J. A., Wilson-Cohn, C., Raroque, J., Kooken, K., Ekman, P, Yrizarry, N., Loewinger, S., Uchida, H., Yee, A., Amo, L., & Goh, A. (2000). A new test to measure emotion recognition ability: Matsumoto and Ekman's Japanese and Caucasian Brief Affect Recognition Test (JACBART). Journal of Nonverbal Behavior, 24(3), 179-209.

Matsumoto, D., Takeuchi, S., Andayani, S., Kouznetsova, N., & Krupp, D. (1998). The contribution of individualism-collectivism to cross-national dif­ferences in display rules. Asian Journal of Social Psychology, 1, 147-165.

Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. (2001). Emotional in­telligence as a standard intelligence. Emotion, 1(3), 232-242.

Miller, J. G. (1984). Culture and the development of everyday social explana­tion. Journal of Personality and Social Psychology, 46, 961-978.

Morris, D., Collett, P, Marsh, P, & O'Shaughnessy, M. (1980). Gestures: Their origins and distribution. New York: Scarborough.

Mumford, D. B. (1998). The measurement of culture shock. Social Psychiatry and Psychiatric Epidemiology, 33, 149-154.

Neuliep, J. W. (1997). A cross-cultural comparison of teacher immediacy in American and Japanese classrooms. Communication Research, 24, 431-451.

Oetzel, J., Ting-Toomey, S., Masumoto, T., Yokochi, Y., Pan, X., Takai, J., & Wilcox, R. (2001). Face and facework in conflict: A cross-cultural compari­son of China, Germany, Japan, and the United States. Communication Monographs, 68(3), 238-253.

Peng, K., Nisbett, R., & Wong, N. Y. C. (1997). Validity problems comparing val­ues across cultures and possible solution. Psychological Methods, 2, 329-344.

Pipher, M. (1998). Another country: Navigating the emotional terrain of our elders. New York: Putnam.

Poortinga, Y. H. (1989). Equivalence of cross-cultural data: An overview of ba­sic issues. International Journal of Psychology, 24, 737-756.

Poortinga, Y. H. (1990). IACCPpresidential address: Towards a conceptual­ization of culture for psychology. Tilburg, The Netherlands.

Rohner, R. P. (1984). Toward a conception of culture for cross-cultural psy­chology. Journal of Cross-Cultural Psychology, 15, 111-138.

Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution process. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 10, pp. 174-221). New York: Academic Press.

Saarni, C. (1979). Children's understanding of display rules for expressive be­havior. Developmental Psychology, 15(4), 424--429.

Seligman, M. E., & Hager, J. (1972). Biological boundaries of learning. New York: Appleton-Century-Crofts.

Shin, H., & Abell, N. (1999). The homesickness and contentment scale: Devel­oping a culturally sensitive measure of adjustment for Asians. Research on Social Work Practice, 9(1), 45-60.

Shweder, R. A., & Bourne, E. J. (1984). Does the concept of the person vary cross-culturally? In R. A. Shweder & R. A. LeVine (Eds.), Culture theory: Essays on mind, sef, and emotion (pp. 158-199). Cambridge, England: Cambridge University Press.

Singelis, T., Bond, M., Sharkey, W. E, & Lai, C. S. Y. (1999). Unpacking cul­ture's influence on self-esteem and embarassability. Journal of Cross-Cul­tural Psychology, 30, 315-341.

Tooby, J., & Cosmides, L. (1992). Psychological foundations of culture. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind (pp. 19-136). New York: Oxford University Press.

Triandis, H. C. (1972). The analysis of subjective culture. New York: Wiley. Triandis, H. C. (1994). Culture and social behavior. New York: McGraw Hill. Triandis, H. C., McCusker, C., & Hui, C. H. (1990). Multimethod probes of in‑

dividualism and collectivism. Journal of Personality and Social Psychol‑

ogy 59(5), 1006-1020.

Von-Raffler Engel, W. (1981). Developmental kinesics: How children acquire communicative and non-communicative nonverbal behavior. Infant Mental Health Journal, 2(2), 84-94.

Wittgenstein, L. (1953). Philosophical investigations. New York: Macmillan.