Intergroup Cognition - Squarespace · Intergroup Cognition YARROW DUNHAM Yale University, United States Thestudyofgroupshasalonghistorywithinthe psychological sciences, reflecting

Trim Size: 170mm x 244mm Callan wbiea1767.tex V1 - 04/05/2017 7:56 P.M. Page 1�

� �

�

Intergroup CognitionYARROW DUNHAMYale University, United States

The study of groups has a long history within thepsychological sciences, reflecting the intuitionthat the natural ecologies within which humansexist are in large part social and cultural. Indeed,the variety of social distinctions our species usesto divide the social world far outstrips what isseen in any other species, with each individualbelonging at once to dozens of social groups.Due to the large number of influential earlyresearchers who came of age in the aftermath ofWorld War II, group polarization, prejudice, andintergroup hostility have been major foci. Otherwell-traveled topics include the ways in whichindividuals make sense of themselves via theirmembership in social groups and how groupsbecome the locus of cooperation and other formsof normative behavior. The breadth and depthof research in this area precludes easy character-ization and the present review will be thematicrather than thorough, using the lens of intergroupbias (i.e., the tendency to positively or negativelyevaluate social groups) as an organizing tool.For a more comprehensive treatment, access theoriginal sources cited throughout this entry andsee the list of other related entries in this volume.

A categorization perspective

Intergroup cognition only arises when we beginto sort individuals into kinds. It follows that thepsychology of intergroup cognition is parasitic onthe psychology of categorization, a psychology wedeploy with respect to nearly everything in ourworld, social and nonsocial alike. Categories canbe thought of as primary constituents of men-tal life, atoms from which communication andculture are composed. Their power lies in down-playing superficial variation while emphasizingthe causal or functional roles we deem vital. Thecategory chair, for example, is useful precisely

The International Encyclopedia of Anthropology. Edited by Hilary Callan.© 2018 John Wiley & Sons, Ltd. Published 2018 by John Wiley & Sons, Ltd.

because it is silent about many details—plasticor wood, three legs or four—but loud aboutfunction—useful for sitting on. Understandingcategorization requires focusing on two com-plementary questions: First, what features areused to pick out category members; that is, whatproperties must an entity have for me to concludethat it is a chair? Second, once I know that theobject in the next room falls under a particularcategory, what do I think this entails; that is, whatfeatures do I expect it to have? These questionsare not as symmetrical as they first appear: wehave no trouble identifying a mid-century mod-ern clear plastic wave as a chair but would beunlikely to expect a chair we have not yet seento possess those idiosyncratic features. Thesesame considerations arise for social categories.Consider an ethnic category widely identified inmy society. How do I decide if someone belongsto this group? Do I attend to dress, adornment,accent? And if I know that someone belongs tothis group, what else do I think is true of them?

Focusing on categorization reminds us thatsome notable features of social categories arederived from the fact that they are categories,not from the fact that they are social. A centralexample is the tendency to underestimate varietywithin social groups (“they are all the same!”)and the parallel tendency to overestimate thedifferences between social groups (“they are verydifferent from us!”). While real and robust, thesephenomena appear just as reliably in completelynonsocial contexts, when the entities being cat-egorized are not people at all, but rather stimulisuch as lengths of lines arbitrarily divided into aset of longer and shorter lengths. In such cases,perceivers learning the category boundary beginto recall lines belonging to different sets to bemore different in length than they are and linesbelonging to the same set to be more similarthan they are. Thus, some of the most importantconsequences of carving the world into kinds arenot social in origin, though of course they mightplay out in particularly problematic ways in thesocial domain.

Thinking in terms of categories also remindsus that things could have been different: forevery seemingly meaningful feature we use to


� �

�

2 INTERGROUP COGNIT ION

categorize others, there are numerous equallymeaningful features we do not use. For example,skin color is frequently meaningful as a marker ofsocial difference in the United States but height orhair color is not. It is not obvious that skin color isin fact or intrinsically more diagnostic of impor-tant differences than these other dimensions andthe genetics of racial variation show that contem-porary racial categories are more cultural thanbiological. Further complicating matters, no indi-vidual belongs to just one social category; rather,they belong to many, dozens even.This multiplic-ity implicates another level of choice: which ofthose myriad identities will I focus on in a givenmoment? The answer to this question will haveimportant consequences for how I see the personin question because there is abundant evidencethat the categories I deploy affect how I interpretevents in the world. For example, perceiversfrequently see the same actions in different wayswhen performed by bodies that are gendered orracialized in particular ways. Certain gestures areperceived by many Americans as more furtive,suspicious, or hostile when performed by bodiesclassified as black and male; certain behaviorsappear more vigorous versus graceful when per-formed by bodies classified as male versus female.These effects can be particularly perniciousbecause the process of seeing others as membersof social categories frequently occurs outside ofconscious awareness, subtly structuring how wemake sense of the social world.

Social identity and the roots of intergroupbias

Within the universe of categories, why are weso drawn to categories of people? The obviousif initially unsatisfying answer is that we carebecause they are about us. From this simpleobservation influential theories of social identityhave arisen, in which group memberships areor can be means of defining and bolstering theself. In describing group membership in this way,an account of ingroup bias (i.e., the tendencyto prefer the group to which one belongs) alsobegins to emerge.

Any consideration of our general orienta-tion toward groups must begin with a hugely

influential body of work suggesting that thetendency to prefer groups to which we belongsprings into existence from the most humbleof origins. In what has come to be called the“minimal group” effect, researchers have shownthat ingroup preference emerges even whenindividuals are placed into previously unfamil-iar and arbitrary social groups, such as thosebased on preference in art, tendency to over-or underestimate dot arrays or even randomassignment (Mullen, Brown, and Smith 1992).This implies that many factors thought criticalto the emergence of intergroup bias, such ascompetition, unequal status, or specific culturalinput, are strictly speaking not necessary forits formation. Thus, the striking tendency foryoung children to manifest ingroup bias withrespect to race, nationality, and so on (reviewedin Raabe and Beelmann 2011) does not nec-essarily require a rich learning-based account.Indeed, the tendency to prefer arbitrarily assignedgroups is robustly in place by the age of five andchildren in minimal groups also tend to remem-ber more ingroup-favoring information andto construe ambiguous situations in ways thatfavor the ingroup. Taken together, a pictureemerges in which children seize on whateversocial distinctions they come across and begin togenerate ingroup bias in earnest, even absent anyspecifically biasing input.

The minimal group effect provides a com-pelling origin story for intergroup bias, suggestingthat the role of social learning is limited. Promi-nent theories seeking to clarify our proclivitytoward ingroup bias argue that individuals arehighly motivated to enhance and positively con-strue the self and they see membership in groupsas a means by which to do so (most prominently,social identity theory: Ellemers and Haslam 2011).This leads individuals to seek out or emphasizemembership in socially favored groups but alsoto focus attention on or even fabricate positiveaspects of their existing group memberships(including experimenter-imposed minimalgroups). In short, inflating the worth of ingroupsbolsters the self. Another class of theories runsthe causal arrow in the other direction. Ratherthan suggesting that positivity associated with thegroup gets transferred to the self, these theoriesbegin with the observation that most individ-uals have a highly, even unrealistically, positive


� �

�

INTERGROUP COGNIT ION 3

view of themselves, revealed by (for example)overestimating their skills or preferring the lettersin their own names (reviewed in Greenwald andBanaji 1995).When an individual learns that theybelong to a particular social group, self-relatedpositivity might spread quite naturally to thatgroup. In short, an already inflated self bolstersthe worth of ingroups. In fact, there is evidenceconsistent with both of these processes.

The pervasive nature of ingroup bias canalso be placed in the context of a major shiftin social–psychological research over recentdecades. Prior developments in the science ofmemory revealed that previously encounteredinformation could have an impact on cognitionand behavior even when the information couldnot be consciously recalled. In the social domain,researchers wondered if self-reports of groupattitudes might be missing an important if sub-tle dimension of intergroup cognition. Mightrespondents be self-censoring or, more pro-foundly, be unaware of a dimension of their owngroup attitudes (Greenwald and Banaji 1995)?If so, cultural portrayals of groups might leavea subtle but important trace on how groups areconceived, of which the respondents themselvesmight be unaware. Researchers have developeda number of tasks to explore this possibility.For example, priming tasks ask if exposure tostimuli representing ingroup or outgroup mem-bers facilitate faster responding to positive andnegative words, respectively. Other tasks usecategorization speed to measure subtle evaluativeassociations between concepts. In one widelyused measure (the implicit association test [IAT];demonstration site: https//implicit.harvard.edu),participants rapidly categorize four kinds ofstimuli using only two response keys. In oneblock, ingroup faces and positive words share oneresponse key while outgroup faces and negativewords share the other. In a second block, thepairings are reversed such that ingroup facesshare a response key with negative words andoutgroup faces with positive words. The logic ofthe task is that responses will generally be fasterwhen a positively evaluated group (usually aningroup) shares a response with positive words.The upshot of research in this area is that askingparticipants to report their intergroup attitudesor stereotypes cannot be taken to exhaust theattitudes and stereotypes they actually hold and

that a large number of individuals who claimnot to hold prejudice show evidence of this moreimplicit form of intergroup bias (Greenwald andBanaji 1995).These implicit forms of ingroup biasalso appear in minimal group settings, suggestingthat it too can be considered a basic consequenceof social identity.

Dimensions of variability

While ingroup bias might have a claim to being astatistical universal true of most humans in mostsettings, it is also widely variable across many lev-els of analysis. Some individuals are more biasedthan others or biased toward some groups but notothers; some entire groups manifest more or lessbias toward some other groups; and some culturescarve up the space of groups, and the evaluationsassociated with them, in different ways. Account-ing for this variability remains a major focus ofmuch research in intergroup cognition.

Individual variability

The perspectives already discussed begin to offeran explanation for variation at the level of theindividual. For example, individual differences inthe need to self-enhance, or in the extent to whichthe self is positively evaluated, do predict levelsof intergroup bias. But another class of theoriesplaces individual variation front and center, con-sidering bias an outgrowth of personality. Perhapsthe first modern elaboration of this idea is theauthoritarian personality. Originally framed inpsychodynamic terms, allegiance to strict grouphierarchies and derogation of other groups wasconsidered a means of resolving intrapsychicconflict by imposing a rigid stability on the exter-nal world. While this interpretation has wanedin popularity along with its Freudian foundation,the broader notion of personality types given togreater prejudice remains current. For example,social dominance orientation (SDO: Sidanius andPratto 2011) refers to an individual’s tendencyto prefer hierarchical systems and to seek outpower and domination over others; individualshigh in SDO generally express more prejudice.Relatedly, individuals may vary in their need tobelieve the world orderly and just. Preserving


� �

�


the belief that the world is just requires believingthat social inequities reflect just desserts; thusindividuals high in this tendency might be moreinclined to internalize intergroup status hierar-chies. In a more existential spin on this generalidea, another view suggests that fear of deathrepresents the ultimate uncertainty, which beliefin a just and well-ordered society offers a hedgeagainst (Greenberg and Arndt 2011). These viewsrepresent forms of motivated social cognition inwhich individuals seek to defend a view of theworld that they find psychologically palliative.

Variability between groups and contexts

By emphasizing social inequities and how they areunderstood or defended, some theories of indi-vidual variation also provide a critical reminderthat intergroup phenomena do not exclusivelyreside “in the head.” Groups and group-basedinequities have material reality and their hier-archical arrangements structure many aspectsof social life. But the preceding accounts stillsay little about how that external reality affectsthe emergence and unfolding of intergroup bias.Realistic conflict theories (e.g., Stephan andMealy 2011), as their name implies, focus on thereal or perceived presence of material conflict,generally over scarce resources. Competition, forexample, generally increases intergroup bias andcan transform outgroup neutrality to outgroupnegativity.Theoretical extensions have focused onincorporating not just material but also symbolicthreats, usually challenges to a group’s core values,or have sought to place conflict in the context ofintergroup hierarchies (Stephan andMealy 2011).

Social and material inequality also raises thequestion of between-group variability in inter-group cognition and in particular how it varies asone travels up or down a status hierarchy. In thislight the critical observation is that among thegroup attitudes of individuals who occupy lowerstatus social positions, such as racial or sexualminorities in the United States, black or coloredindividuals in South Africa, and lower casteindividuals in India, intergroup bias frequentlydoes not emerge or in some cases emerges in theform of preference for the higher-status outgroup.In general, as status disparities grow stronger,members of disadvantaged groups grow less likely

to manifest ingroup positivity. A straightforwardinterpretation of this pattern is that negativecultural portrayals of their group reduce thepositivity they would otherwise associate withtheir ingroup and work with children suggeststhat this occurs quite early, perhaps as young asthe age of three (Dunham and Olson 2008). Acritical question emerging in the wake of thesefindings is what cues children are drawing on toform representations of their own group’s socialstatus. One emerging suggestion is that childrenuse cues denoting economic status and positivelyevaluate groups associated with greater wealth.It would be fascinating to determine whetherstatus cues of other sorts, such as those that aremore common outside contemporary Westerncultures, also exert these early effects.

Cross-cultural and developmentalperspectives

A final key element of variation occurs acrosscultures, which construct the social order outof different social categories entirely and mayalso reason about groups in different ways. Asalready seen, work on the relationship betweensocial status and intergroup attitudes has incor-porated data from multiple cultural settings andrepresents one attempt to capture cultural varia-tion. But perhaps the more pressing question iswhether theories that were generally developedto account for race attitudes in North Americacan reliably be extended to other cultural set-tings. While space precludes detailed treatmenthere, one prominent debate has centered onwhether self-enhancement can be a viable generalorigin of intergroup bias given evidence that“interdependent” or “collectivist” individualssuch as East Asians self-enhance considerablyless but show equal or perhaps even greaterdegrees of ingroup favoritism. Other notablework has sought macrolevel cultural correlatesof intergroup bias such as the strength of socialinstitutions (which could reduce the need to relyon group affiliation). Despite these few areas ofwork, however, investigation of cultural variationin intergroup cognition is still at an early stage.

A quite different approach to cultural variation,or at least to enculturation, is provided by devel-opmental research. How does the cultural raw


� �

�

INTERGROUP COGNIT ION 5

material a child encounters interact with basiclearning processes to produce adult-like inter-group cognition? For many developmentalists,the question is how constrained the learning pro-cess is: Do children simply absorb their culture’svalues in a sponge-like manner? Or do theytransform the input in particular ways, leadingparticular forms of group cognition to emergewith greater regularity than they might other-wise (Dunham and Olson 2008)? Supportingthe latter view, children appear to hold somebasic assumptions about how groups function.For example, they expect members of the samegroups to preferentially interact and expect indi-viduals to avoid directing negative behavior atingroup members (Rhodes 2012). Further, theyalso expect intergroup interactions to deteriorateunder conditions of resource scarcity or competi-tion (Rhodes 2012). Another oft-discussed aspectof early intergroup cognition is essentialism—thatis, the belief that group differences are under-written by an invisible but causally powerfulessence that predicts numerous observed andunobserved similarities among category mem-bers. Such psychological essentialism is usuallyexamined by assessing whether children expectcategory membership to be enduring, unchange-able, and hereditary. Interestingly, children donot essentialize all social categories but, rather,are sensitive to features of the groups themselvesas well as patterns of adult input that signalspecial status. For example, Western childrenessentialize gender more than race and race morethan occupation, which some have interpreted asevidence that children reason about social kindsby analogy to biological kinds (Hirschfeld 1998).Further, there is evidence that children essen-tialize at least some social groups even if theircultural elders do not, suggesting they are notmerely imitating adult conceptions, though somespecific forms of linguistic input appear to triggeressentialist construals of groups. Taken together,these tendencies can be interpreted as a naivesociology that characterizes early social cognitionand serves as the child’s attempt to identify themost meaningful or “deepest” social distinctionsin their cultural milieu (Hirschfeld 1998).

Lurking in the background of this discussionis the question of whether social categories areintrinsically equal prior to being emphasized orprivileged in a given culture or if some kinds

of social categories are special—for example,because they are supported by specialized, pre-sumably evolved, cognitive machinery. Thissuggestion has most frequently been made aboutbiological sex because of its role in sexual repro-duction (Sidanius and Pratto 2011). Others havesuggested a privileged role for kinship on similargrounds and for age as well as language or accentbecause those distinctions were plausibly presentacross human history and so could have servedas stable factors over which selection could haveoperated. While there is no clear consensus onthese questions, the framing of the problem asan effort to distinguish “arbitrary” from “natural”social categories parallels other work in the cog-nitive sciences which seeks to uncover naturalstructure in the child’s attempts to generate ratio-nal models of a complex external world. In anycase it is clear that, given their variability acrosshistory and culture, the vast majority of socialcategories are cultural constructions.

From groups to norms

As life unfolds in small collectives, numerousbehavioral routines will emerge and develop,constituting the conventional aspects of everydaylife. How is food prepared? How does one dress?How do men and women interact? An individualgrowing up in a given culture tacitly internalizes ahost of such cultural norms and they become thefamiliar and comfortable backdrop against whichlife exists. Further, young children frequentlyconsider cultural norms and rituals not merelydescriptive but actively proscriptive; that is, ashaving normative force as what one ought to doand merely engaging in shared novel rituals canenhance ingroup preference. These tendenciesplausibly support intergroup bias by paintingconventional ingroup actions as laudable and,at least where those conventions diverge, paint-ing outgroup actions as blameworthy. Thus,considering groups as containers of normativebehavior offers insight into how familiarity andproximity produce similarity, which can in turndrive preferences.

A norm-focused perspective also offers awindow into a related topic of cooperation.Individuals will typically be more comfortable


� �

�


interacting with those with whom they sharenorms and perhaps also be better able to iden-tify reliable partners by accurately interpretingpast patterns of behavior and other reputationalcues. Thus, shared group membership mightcome to serve as a heuristic cue to a cooperativepartner, shifting the calculus of cooperation anddefection in risky economic games such as thosemodeled by prisoners’ dilemmas or public goodsgames in which long-term payoffs are height-ened by cooperation (sometimes referred to as“tag-based” cooperation). On this picture, groupsmove from containers for norms to containers forgeneralized reciprocity (Yamagishi and Kiyonari2000). This special role for shared norms andcooperation within the group also dovetails withtheories that assume the human ancestral pastto be warlike and therefore that group cohesionwould be a critical element of survival.

Conclusion

Research on intergroup cognition is broad anddeep enough to resist easy characterization. Thisentry has used the tendency to prefer ingroupsas an organizing framework because our speciesseems inexorably given to not just form socialcategories but to render them deeply evaluative.This “first fact” of human intergroup cognition isa pillar that supports much of intergroup cogni-tion, from our tendency to preferentially interactwith, learn from, and cooperate with those weconsider an “us” over those we consider a “them.”

SEE ALSO: <DRAFT: Social Categories> ;Race and Racisms; Categories and Taxonomies;wbiea1700

wbiea1702 Genocide; Interethnic Friction; Ethnicity,wbiea1770wbiea1825wbiea1924

Multiculturalism, and Transnationalism;Transnationalism; Purity, Pollution, and Systemswbiea1980

wbiea2003 of Classification; Intergroup Conflict; Gender,wbiea2219wbiea2279 Nationalism, and Ethnicity

REFERENCES AND FURTHER READING

Dunham, Yarrow, and Olson, Kristina R. 2008. “TheImportance of Origins:WhyCognitive DevelopmentIs Central to a Mature Understanding of Social Cog-nition.” Open Psychology Journal 1: 59–65.

Ellemers, Naomi, and S. Alexander Haslam. 2011. “So-cial Identity Theory.” Handbook of Theories of SocialPsychology 2: 379–98.

Greenberg, Jeff, and Jamie Arndt. 2011. “Terror Man-agementTheory.”Handbook of Theories of Social Psy-chology 1: 398–415.

Greenwald, Anthony G., andMahzarin R. Banaji. 1995.“Implicit Social Cognition: Attitudes, Self-esteem,and Stereotypes.” Psychological Review 102 (1): 4.

Hewstone, Miles, Mark Rubin, and Hazel Willis. 2002.“Intergroup Bias.” Annual Review of Psychology53: 575–604. doi:10.1146/annurev.psych.53.100901.135109.

Hirschfeld, Lawrence A. 1998. Race in the Making:Cognition, Culture, and the Child’s Construction ofHuman Kinds. Cambridge, MA: MIT Press.

Mullen, Brian, Rupert Brown, and Colleen. 1992. “In-group Bias as a Function of Salience, Relevance, andStatus: An Integration.” European Journal of SocialPsychology 22 (2): 103–22.

Raabe, Tobias, and Andreas Beelmann. 2011. “De-velopment of Ethnic, Racial, and National Preju-dice in Childhood and Adolescence: AMultinationalMeta-analysis of Age Differences.” Child develop-ment 82 (6): 1715–37.

Rhodes, Marjorie. 2012. “Naïve Theories of SocialGroups.” Child Development 83 (6): 1900–1916.doi:10.1111/j.1467-8624.2012.01835.x.

Sidanius, Jim, and Felicia Pratto. 2011. “Social Domi-nance Theory.” Handbook of Theories of Social Psy-chology 2: 418.

Stephan, Walter G., and Marisa D. Mealy. 2011. “In-tergroup Threat Theory.” The Encyclopedia of PeacePsychology. Oxford: Wiley Blackwell.

Yamagishi, Toshio, and Toko Kiyonari. 2000. “TheGroup as the Container of Generalized Reciprocity.”Social Psychology Quarterly 63 (2): 116–32.


� �

�

Please note that the abstract and keywords will not be included in the printed book, but are requiredfor the online presentation of this book which will be published on Wiley’s own online publishingplatform.

If the abstract and keywords are not present below, please take this opportunity to add them now.The abstract should be a short paragraph of between 50 and 150 words in length and there shouldbe at least 3 keywords.

ABSTRACTIntergroup cognition has been among the central topics of social psychology since its inceptionand research has recently surged in allied fields such as developmental and cognitive psy-chology. Cognitive anthropology has also offered unique contributions, especially by drawingattention to commonalities as well as differences across cultural settings. This review placesthe study of intergroup cognition within a broader categorization framework, asking how cat-egorical thinking about people affects social judgment and interaction. The ease with whichintergroup bias emerges, its close connection to intergroup status hierarchies, themany dimen-sions along which intergroup bias varies (between individuals, groups, and cultures), and theconnection between groups and norms are highlighted.

KEYWORDScognition; cognitive anthropology; cognitive science; intergroup cognition

Documents

Intergroup Cognition - Squarespace · Intergroup Cognition YARROW DUNHAM Yale University, United States Thestudyofgroupshasalonghistorywithinthe psychological sciences, reflecting