Hong

Hong, Culture and Communication Processes

s [ Pobierz całość w formacie PDF ]
//-->Multicultural MindsA Dynamic Constructivist Approach to Culture and CognitionYing-yi HongMichael W. MorrisChi-yue ChiuVer6nica Benet-MartfnezHong Kong University of Science and TechnologyStanford UniversityUniversity of Hong KongUniversity of MichiganThe authors present a new approach to culture and cog-nition, which focuses on the dynamics through which spe-cific pieces of cultural knowledge (implicit theories) be-come operative in guiding the construction of meaningfrom a stimulus. Whether a construct comes to the fore ina perceiver’s mind depends on the extent to which theconstruct is highly accessible (because of recentexposure).In a series of cognitive priming experiments, the authorssimulated the experience of bicultural individuals (peoplewho have internalized two cultures) of switching betweendifferent cultural frames in response to culturally ladensymbols. The authors discuss how this dynamic. construc-tivist approach illuminates (a) when cultural constructs arepotent drivers of behavior and (b) how bicultural individ-uals may control the cognitive effects of culture.lthough the multiplicity of cultural identities andinfluencesis hardly a new phenomenon, it is oneincreasingly discussed. In contemporary populardiscourse, it is becoming increasingly rare to hear the wordculturalwithout the prefixmulti-.Multicultural experience,however, has been underinvestigated in psychological re-search on culture, particularly within the most prominentresearch paradigm of cross-cultural psychology (see Segall,Lonner, & Berry, 1998). There are several reasons for this.First, somewhat obviously, methodological orientations in-fluence a researcher’s choice of topics, and culture has beenassessed primarily as an individual difference, with themethods for its evaluation developed by clinical and per-sonality researchers to distinguish types of persons. Insofaras the cross-cultural method relies on uncovering differ-ences across cultural groups (usually indexed by national-ity), the influence of multiple cultures on an individualmerely creates error variance. Second, on a more subtlelevel, the theoretical assumptions predominant in cross-cultural scholarship have impeded an analysis of the dy-namics of multiple cultures in the same mind. The effort toidentify the knowledge that varies between but not withinlarge cultural groups has led to the conceptualization ofcultural knowledge in terms of very general constructs,Asuch as individualistic as opposed to collectivist valueorientations, which apply to all aspects of life (Segall et al.,1998). With the emphasis on domain-general constructshas come the assumption that the influence of culture oncognition is continual and constant. Cultural knowledge isconceptualized to be like a contact lens that affects theindividual’s perceptions of visual stimuli all of the time.This conception unfortunately leaves little room for a sec-ond internalized culture within an individual’s psychology.In sum, the methods and assumptions of cross-culturalpsychology have not fostered the analysis of how individ-uals incorporate more than one culture.Our introduction of an alternative approach to culturetakes as a point of departure a commonly reported experi-ence, which we callframe switching,among biculturalindividuals. While frame switching, the individual shiftsbetween interpretive frames rooted in different cultures inresponse to cues in the social environment (LaFromboise,Coleman, & Gerton, 1993). To capture how biculturalindividuals switch between cultural lenses, we adopt aconceptualization of internalized culture as a network ofdiscrete, specificconstructs that guide cognition only whenthey come to the fore in an individual’s mind. Fortunately,theories and methods have been developed in cognitive andsocial psychology, such as the technique of cognitive prim-Editor’s note.Denise C. Park served as action editor for this article.Authors note.Ying-yi Hong, Division of Social Science, Hong KongUniversity of Science and Technology. Hong Kong; Michael W. Morris,Graduate School of Business, Stanford University; Chi-yue Chiu, Depart-ment of Psychology. University of Hong Kong, Hong Kong; VerdnicaBenet-Martfnez, Department of Psychology, University of Michigan.Preparation of this article was supported by Research GrantsHKU7045/99H and HKUST6lR2/98H from the Research Grants Councilof the Hong Kong Special Administrative Region, China. We thankRobertGore,Steven Heine, Emiko Kashima, Hazel Markus, RichardNisbert, Barry Sautman. Virgina Unkefer, and Robert Wyer for theircomments on an earlier version of this article.Correspondence conceming this article should be addressed toYing-yi Hong, Division of Social Science, Hong Kong University ofScience and Technology, Clear Water Bay, Hong Kong. Electronic mailmay be sent to sohong@ust.hk.July 2000•American PsychologistCopyright 2000 bytheAmerican Poychologiool Arsoniation. joe. 5503-5hhX150/5550Vol.55.No.7.709—720001l0l537/10003-O6hX.55.7759709ferent cultures. Consider, for example, the following expe-rience of a Mexican American individual:At home with my parents and grandparents the only acceptablelanguage was Spanish; actually that’s all they really understood.Everything was really Mexican, but at the same time they wantedme to speak good English....But at school, I felt really differentbecause everyone was American, including me. Then I would gohomein the afternoon and be Mexican again. (quoted in Padilla,1994, p. 30)This example illustrates that frame switching may occur inresponse to cues such as contexts (home or school) andsymbols (language) that are psychologically associatedwith one culture or the other. Reports of frame switching atwork are common in the literature on minority or expatriateemployees (e.g., Bell, 1991). Similar experiences are re-ported by ethnographers during fieldwork:The longer I lived inSamoa, the more I was able to use the Samoans’ cultural resourcest found myself constantly flip-flopping....Ying-yi HongPhoto by Imun Poonthe flow of my everyday experiences was increasingly filteredthrough Samoan models. (Shore, 1996, p. 6)A Dynamic Constructivist AnalysisTo understand frame switching in bicultural individuals,we have adopted an approach influenced by constructivistapproaches to culture in several disciplines and by contem-porary social psychological research on the dynamics ofknowledge activation. A first premise is that a culture is notinternalized in the form of an integrated and highly generalstructure, such as an overall mentality, worldview, or valueorientation. Rather, culture is internalized in the form of aloose network of domain-specific knowledge structures,such as categories and implicit theories (Bruner, 1990;D’Andrade, 1984; Shore, 1996; Strauss, 1992).Asecondpremise is that individuals can acquire more than one suchcultural meaning system, even if these systems containconflicting theories. That is, contradictory or conflictingconstructs can be simultaneously possessed by an individ-ual; they simply cannot simultaneously guide cognition.The key to this distinction is that possessing a particularconstruct does not entail relying on it continuously; only asmall subset of an individual’s knowledge comes to thefore and guides the interpretation of a stimulus. This dy-namic constructivist approach differs in its conception ofculture from cross-cultural psychology, yet it is a comple-mentary rather than a rival approach in that it builds onprevious insights and draws attention to novel researchquestions and novel accounts of phenomena, such as frameswitching.A basic research question relevant to frame switchingis how particular pieces of cultural knowledge becomeoperative in particular interpretive tasks. To investigate thisquestion, we have drawn concepts and methods from socialpsychological research on how stereotypes, schemas, andother constructs move in and out of operation (Fiske,1998). A key concept is that the pieces of an individual’sknowledge vary inaccessibility(Higgins, 1996; Wyer &Srull, 1986). The more accessible a construct, the moreing, to manipulate through experiment which of the con-structs in an individual~s mind comes to the fore (for areview, see Higgins, 1996). We illustrate in this article howthis conceptualization creates a set of new methods thatinvolves bicultural participants testing the consequences ofculture. These methods offer greater internal validity thando the quasi-experimental comparisons typically relied onin cross-cultural research. After reviewing studies of cul-tural frame switching, we then discuss how this approachelucidates other topics, such as the relation between cul-tural beliefs and action, the role of culture in emotions andmotivations, and the process of acculturation. This ap-proach illuminates not only the experiences of biculturalindividuals but also the more general roles that cultureplays in mental and emotional life.Frame SwitchingBicultural individuals are typically described as peoplewho have internalized two cultures to the extent that bothcultures are alive inside of them. Many bicultural individ-uals report that the two internalized cultures take turns inguiding their thoughts and feelings (LaFromboise et al.,1993; Phinney & Devich-Navarro, 1997). This is interest-ing because it suggests that (a) internalized cultures are notnecessarily blended and (b) absorbing a second culturedoes not always involve replacing the original culture withthe new one. Classical scholarship on African Americans,for instance, describes movement back and forth between“two souls, two thoughts, two unreconciled strivings, twowarring ideals” (DuBois, 1903/1989, p.5).Ethnographiesof Asian Americans and Hispanic Americans, among othergroups, describe switches between mindsets rooted in dif-710July 2000•American PsychologistMichael W.MorrisPhoto by S. G!odfelterlikely it is to come to the fore in the individual’s mind andguide interpretation.But what determines whether a piece of knowledge ishighly accessible? A long-standing hypothesis in cognitiveand social psychology holds that a construct, such as acategory, is accessible to the extent that it has been acti-vated by recent use (Bruner, 1957). Abundant evidence forthis comes from experiments in which researchers manip-ulate whether participants are exposed to a word or imagerelated to a construct (aprime)and then measure the extentto which the participants’ subsequent interpretations of astimulus are influenced by the primed construct (for areview, see Higgins, 1996). For example, in one experi-ment (Chiu et al., 1998), participants were primed eitherwith pictures of a masculine man and a feminine woman orwith gender-unrelated (control) pictures. Later, in a pur-portedly unrelated task, they were asked to interpret anambiguous behavior (e.g., “Donna’s friend ordered a cof-fee, and so did Donna”). Participants primed with gender-related pictures constructed interpretations that showed aninfluence of gender stereotypes: For example, they judgedDonna to be dependent on others in making decisions.Participants in the control condition did not make suchinterpretations. In this experiment, gender-related picturesactivated stereotypes in the minds of participants, whichthen made it more likely that these stereotypes becameoperative and guided inferences when participants soughtto make sense of the behavioral stimulus.An important design feature in many priming studiesis that the priming is presented to participants as part of anunrelated experiment, and participants are not aware of itsinfluence in the interpretive task. Some studies have primedconstructs that are one step removed from the construct thatapplies to the interpretive task. For example, priming withwords related to African Americans led White participantsto interpret hostility in stimulus behavior by race-unspeci-fied actors (Gaertner & McLaughlin, 1983); priming withcues with positive affective valence led participants tosubsequently rely on person categories having the sameaffective valence (Niedenthal & Cantor, 1986). Thesepriming effects rely on the spillover or spread of activationfrom one construct to other linked constructs within anetwork of constructs that are psychologically associatedfor participants (see Anderson, 1976).In our research on frame switching, we used theconcept of accessibility and the technique of priming tomodel the phenomenon experimentally. We posited thatbicultural individuals who have been socialized into twocultures, A and B, have, as a result, two cultural meaningsystems or networks of cultural constructs, which can bereferred to as A’ and B’. Accordingly, priming biculturalindividuals with images from Culture A would spreadactivation through Network A’, elevating the accessibilityof the network’s categories and the implicit theories thenetwork comprises. Likewise, priming with images fromCulture B would spread activation through Network B’,elevating the accessibility of the constructs that networkcomprises. In looking for the ideal primes to test thisaccount, we searched for symbols that would activate con-structs central to specific cultural networks yet not sodirectly related to the interpretive task. Thus, participantscould not consciously connect the prime with the stimulus.We turned to iconic cultural symbols.Icons: Triggers of Cultural KnowledgeIcons have been called “magnets of meaning” in that theyconnect many diverse elements of cultural knowledge(Betsky, 1997). Like religious icons, cultural icons areimages created or selected for their power to evoke inobservers a particular frame of mind in a “powerful andrelatively undifferentiated way” (Ortner, 1973, p. 1339).The potency and distinctiveness of icons make them idealcandidates for primes that would spread activation in anetwork of cultural constructs. Some examples of centralicons in the mainstream American and Chinese culturaltraditions are shown in Figure 1. Exposing Chinese Amer-ican bicultural individuals to American icons should acti-vate interpretive constructs in their American culturalknowledge network; exposing the same individuals to Chi-nese icons instead should activate constructs in their Chi-nese cultural knowledge network.Interpreting Behavior of Individual andGroup Actors: A Litmus TestOur research also required an interpretive task that is in-fluenced by cultural knowledge in a well-understood man-ner. Here the legacy of cross-cultural psychology is invalu-able in that we can seek to replicate, by priming differentcultures within the minds of bicultural individuals, thepatterns of differences that have been discovered in previ-ous cross-national comparative studies. Many such patternsJuly 2000•American Psychologist711Chi-yue ChiuPhoto by man PoonAexist. For example, in self-description tasks, North Amer-icans are consistently more likely than Japanese to makeself-enhancing statements (Kitayama & Markus, 1994). Animportant consideration, however, is that many JapaneseAmerican biculturals are, no doubt, aware of this differ-ence. Hence, exposing bicultural individuals to culturalicons could affect this difference either through unobtru-sive priming of knowledge structures or through demandcharacteristics. We needed a stimulus task that participantswould not consciously connect to cultural icons. In short,the task could not be transparently related to culture.To develop a test for cultural priming that would benontransparent to participants, we turned to interpretationsof social behavior. Social psychologists have long studiedhow perceivers attribute the behavior of others to causes,noting systematic biases, such as tracing an individual’sactions to personality dispositions rather than other plausi-ble factors such as social context (Heider, 1958; Ross,1977). Perhaps the most famous evidence for this biascame from studies conducted by Heider and Simmel (1944)in which participants were presented with animated films ofgeometric shapes, such as triangles and circles, that weremoving in patterns suggestive of social interactions. Par-ticipants tended to interpret the films by ascribing motivesand personalities to an individual shape. Heider (1958)concluded that social information is interpreted by formingunits, primarily the unit of an individual person. The personunit then tends to attract most of the perceiver’s attention,resulting incausal attributions that overweigh internal per-sonal factors and underweigh factors in the surroundingsocial situation. Other researchers have studied everydayinteractions in whichthis bias of tracing an individual’sbehavior to dispositions leads to incorrect interpretations ofthe individual’s behavior and suboptimal ways of interact-ing with him or her (Jones & Harris, 1967; Morris, Larrick,& Su, 1999). Because of its pervasiveness and consequen-tiality, this dispositionist bias has been called thefunda-mental attribution error(Ross, 1977).Recent research has allowed psychologists to identifythe role that culture plays in shaping the dispositionist biasin social perception. Prompted by ethnographic accounts ofChinese social understanding (Hsu, 1953), Moms and Peng(1994) investigated the hypothesis that the tendency ofperceivers to focus on individuals and interpret behavior interms of their internal dispositions may be more marked inNorth America than in China. They reasoned that an im-plicit theory that individuals are autonomous relative to thepressures of the group is central to American culture,whereas in Chinese culture a more salient implicit theoryemphasizes that individuals accommodate the greater au-tonomy of groups (So et al., 1999). In studies in which theyused several methods, Morris and Peng showed that Amer-ican participants accorded more weight to an individual’spersonal dispositions, whereas Chinese participants ac-corded more weight to an individual’s social context. Fur-ther evidence for the difference in implicit theoriesemerged fromstudies directly measuring generalized be-liefs about individuals versus social groups and institutions(Chiu, Dweck, Tong, & Fu, 1997). In a recent review ofstudies comparing North American and East Asian perceiv-ers, researchers concluded that the sharpest differences inFigure 1Exampies of Icanic Images in American andChinese CulturesAmerican PrimesChinese Primes~**********~******rr*t*-*U4712July2000•American PsychologistCultural Priming StudiesIn a series of studies, we experimentallycreated frameswitching among bicultural individuals.Next, we reviewthree of the studies. The first two studies used the primingmethod to replicate in bicultural individuals the cross-national attribution differences revealed by Morris andPeng (1994). The third study is a conceptual replication ofthe first two studies, but the dependent measures wereattributions for a social event.Bkultural ParticipantsWho were the bicultural individuals we recruited in thestudies? Our initial studies involved Westernized Chinesestudents in Hong Kong. Although traditional Chinese val-ues are emphasized in the socialization processes in HongKong (Ho, 1986), contemporary university students inHong Kong are acculturated with Western social beliefsand values (Bond, 1993). This is related to the fact thatHong Kong was a British-administrated territory for morethan a century. Before 1997, English, not Chinese, was theofficial language of instruction in about 80% of the sec-ondary schools (Young, Giles, & Pierson, 1986). Further-more, large British and American expatriate communitiesand the salient presence of English-language television,films, and so forth means that Hong Kong Chinese studentsVeronicaBenet-Martinezattributions for the cause of an individual’s behavior lie inthe weight accorded to the contexts of constraints andpressures imposed by social groups (Choi, Nisbett, &Norenzayan, 1999). Consistent with this indication thatEast Asians accord causal potency to social collectives, instudiesof how perceivers attribute actions by groups re-searchers have found that East Asians make attributions tothe dispositions of groups more than Americans do (Me-non, Morris, Chiu, & Hong, 1999). In sum, cultural differ-ences in the attributional weight accorded to the disposi-tions of individuals versus groups are well documented.An important feature of attribution differences is thatthey can be studied with nontransparent methods. One ofthe methods used by Morris and Peng (1994) adaptedHeider’ s strategy of presenting animated films that partic-ipants do not consciously associate with social or culturaltopics. Morris and Peng designed animated films of fishfeaturing an individual and a group in which it was am-biguous whether the individual’s differing trajectory re-flected internal dispositions or the influence of the group. Inone type of display, the individual fish swam outside of thegroup, leaving ambiguous whether the individual’s separa-tion reflected an internal disposition (a leader leading otherfish) or pressure from the group (an outcast being chased byother fish). In explaining the individual fish’s behavior,Chinese participants attributed less to internal dispositionFigure 2A Display Adapted From Morris, NisbeWand Peng (1995)—HfII2INTERNAL FORCEl~I3EXTERNAL FORCE4S~C23—Sof the fish in front but more to the external (group) factorsthan did American participants (see Figure 2). This methodof measuring cultural differences through the ways socialperceptions are anthropomorphically projected onto ani-mals has the advantage that participants are unaware cul-ture is relevant to the task.Note.Paints A and C mark the mean American and Chinese ratings, respec-tively, on the internal and external attribution scales. From ‘Causal AttributionAcross Domains and Cultures,” byM. W. Morris, R. E. Nisbett, and K. Peng,1995,in D. Sperber, D. Premack, and A. J. Premock, Cau.sa/ Cognition:AMultidisciplinary Debate (pp. 577—612), Oxford, England: Clarendon Press.Copyright 1995 by Clarendon Press. Adapted with permission.July2000•AmericanPsychologist713 [ Pobierz całość w formacie PDF ]