Cognition 124 (2012) 227–233

Contents lists available at SciVerse ScienceDirect

Cognition

journal homepage: www.elsevier .com/locate /COGNIT

Brief article

Origins of ‘‘Us’’ versus ‘‘Them’’: Prelinguistic infants prefer similar others

Neha Mahajan a,⇑, Karen Wynn b

a Department of Psychology, Temple University, 580 Meetinghouse Rd., Ambler, PA 19002, USAb Department of Psychology, Yale University, 2 Hillhouse Ave., New Haven, CT 06520, USA

a r t i c l e i n f o a b s t r a c t

Article history:Received 7 February 2011Revised 6 April 2012Accepted 10 May 2012Available online 4 June 2012

Keywords:Social categorizationInfant cognitionCognitive developmentIntergroup biasSocial cognition

0010-0277/$ - see front matter � 2012 Elsevier B.Vhttp://dx.doi.org/10.1016/j.cognition.2012.05.003

⇑ Corresponding author. Tel.: +1 203 436 1415; faE-mail addresses: neha.mahajan@aya.yale.edu

wynn@yale.edu (K. Wynn).

A central feature of human psychology is our pervasive tendency to divide the social worldinto ‘‘us’’ and ‘‘them’’. We prefer to associate with those who are similar to us over thosewho are different, preferentially allocate resources to similar others, and hold more posi-tive beliefs about similar others. Here we investigate the developmental origins of thesebiases, asking if preference for similar others occurs prior to language and extensive expo-sure to cultural norms. We demonstrate that, like adults, prelinguistic infants prefer thosewho share even trivial similarities with themselves, and these preferences appear to reflecta cognitive comparison process (‘‘like me’’/‘‘not like me’’). However, unlike adults, infantsdo not appear to prefer others with an utterly arbitrary similarity to themselves. Together,these findings suggest that the phenomena of ingroup bias, and enhanced interpersonalattraction toward those who resemble ourselves, may be rooted in an inherent preferencefor similarity to self, which itself may be enhanced during development by the influence ofcultural values.

� 2012 Elsevier B.V. All rights reserved.

1. Introduction 1992; Stephenson, Skinner, & Brotherton, 1976), suggesting

When we look at other people, we think of them assimilar to us or different from us based on salient common-alities such as race, ethnicity, class, and religion. Such simi-larities (or lack thereof) influence our preferences for andtreatment of others in striking ways. We engage in acts ofgreat cooperation and support toward those whom we con-sider to be like us, while relations to those we consider to beother, or unlike us are often characterized by hostility andconflict, including prejudice, political unrest, and evengenocide.

Preference for similar others exists across cultures(Allport, 1954/1979; Brewer, 1979). Importantly, itincreases with the salience or cognitive accessibility of a gi-ven dimension of similarity or group distinction (e.g.,Bettencourt, Miller, & Hume, 1999; Hong & Harrod, 1988;Leonardelli & Brewer, 2001; Mullen, Brown, & Smith,

. All rights reserved.

x: +1 203 432 7172.(N. Mahajan), karen.

that the preference for similar others rests on a cognitivecomparison of self to other (Abrams & Hogg, 1990; Mullen,1983; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; Wil-der, 1986). This preference manifests itself in a variety ofways, with consequences for how we perceive and treatothers. We are relatively more attracted to those who aresimilar to us in physical appearance (Berscheid & Walster,1969; Walster, Walster, Berscheid, & Dion, 1971), in atti-tudes and beliefs (Byrne, 1971; Miller, 1972; Clore, 1976),in social and cultural background (Buss & Barnes, 1986; By-rne, Clore, & Worchel, 1966), and in personality (Botwin,Buss, & Shackelford, 1997; Goldman, Rosenzweig, & Lutter,1980; Klohnen & Luo, 2003). Conversely, we tend to evalu-ate dissimilar others more critically, assuming that they arelikely to possess a host of negative qualities, such as unkind-ness, unfairness, and unintelligence (Brewer, 1979; Doiseet al., 1972). We treat them differently as well, avoidingand disliking those who have different attitudes or interests(Singh & Ho, 2000; Tan & Singh, 1995). This relative prefer-ence for similar others is readily observed, even in cases ofrandom assignment to impermanent groups distinguishedby trifling characteristics (e.g., Tajfel, 1970; 1982).

228 N. Mahajan, K. Wynn / Cognition 124 (2012) 227–233

While the nature and formation of social preferences forsimilarity have been extensively studied in adults, less isknown about their developmental origins. Biases for one’sown race and gender appear in childhood using both expli-cit (Aboud, 1988; Huston, 1983) and implicit measures(Baron & Banaji, 2006; Dunham, Baron, & Banaji, 2008;Skowronski & Lawrence, 2001). Moreover, like adults, chil-dren as young as 3 years of age display attitudes biased to-wards others who share with them trivial attributes, suchas a randomly assigned T-shirt color (Bigler, Brown, &Markell, 2001; Bigler, Jones, & Lobliner, 1997; Patterson& Bigler, 2006) when this perceptually salient attribute islabeled and used functionally. In some cases, children haveeven demonstrated both implicit and explicit bias immedi-ately after random assignment (Dunham, Baron, & Carey,2011). These responses may reflect early-learned attitudes,or inborn preferences for similarity to self. Distinguishingbetween these two alternatives is important for elucidat-ing the underlying mechanisms governing these biases.Further, members of our social groups tend to be similarto ourselves on at least one dimension, that of belongingto the specified group, while members of other groups dif-fer on at least this dimension. Thus, understanding how so-cial preference for similarity develops may help elucidatethe development of one key component of ingroup bias.

We address this issue by exploring social preferences inprelinguistic human infants. Previous studies have foundthat infants prefer speakers of their native language(Kinzler, Dupoux, & Spelke, 2007), and, when raised in a ra-cially homogeneous environment, prefer to look at faces oftheir own race (Bar-Haim, Ziv, Lamy, & Hodes, 2006). Thesefindings reflect the existence, early in human life, of a biasfor the familiar over the exotic.

Here, we investigate whether infants also manifest so-cial preferences based on similarity to self, independent offamiliarity. Our experiments ask whether prelinguistic in-fants, like adults, generate social preferences based onminor and even arbitrary similarities to self, and whethersalience of similarity influences preference. Specifically,we explore whether 11-month-old infants prefer otherswho express tastes similar to their own. We chose sharedfood preferences as our first measure of similarity, asshared eating and similar eating habits tends to be animportant marker of different social groups; for this rea-son, we thought that food preferences might be an espe-cially salient domain for infants’ social preferences.

Fig. 1a. Puppets used in Experiments 1a and 1b (photo shows different-foods condition).

2. Experiment 1a

In this experiment, infants were first given a choice be-tween two foods, and asked to choose between a puppetwho shared their tastes (i.e. who expressed a liking forthe infant’s preferred food and a dislike of the other) andone who did not (who expressed the opposite attitudes).Infants were placed into two groups: those in the‘‘Different Foods’’ condition chose between graham crack-ers and green beans, while those in the ‘‘Similar Foods’’condition chose between graham crackers and Cheerios.These two conditions helped to assess the basis for apossible preference for similar others. It is possible that

in the case of highly distinct foods, an infant might holdstrongly valenced attitudes towards each food and viewsomeone with the opposite preferences as objectively mis-taken about the world (and therefore less likeable than onewho is knowledgeable). In the Similar Foods condition, asinfants are typically enthusiastic about both Cheerios andgraham crackers, there should be less tendency to viewan individual who chooses differently as gravely wrongor aberrant. Thus, this manipulation addressed whether apreference for similar others, if observed, were based pri-marily on a judgment about others’ knowledge of the world,or about their similarity to self.

2.1. Methods

2.1.1. Participantswere 32 11.5-month-old infants (mean: 11;15, range

10;27–12;4). Four additional infants were tested but ex-cluded for refusal to choose a puppet.

2.1.2. ProcedureThe experiment included three phases: The

Infant-Chooses-Food task, the Puppets-Choose-Foodpresentation, and the Infant-Chooses-Puppet task. The In-fant-Chooses-Food task began in a reception room. Anexperimenter (E1) held two small bowls equidistant fromthe infant (seated in a parent’s lap) filled with equalamounts of graham cracker bits and either cut green beans(Different Foods condition) or Cheerios (Similar Foods condi-tion). Choice was coded as the first snack an infant pickedup. Puppets were not present during this phase.

Each infant was then brought into a testing room for thePuppets-Choose-Food presentation. The infant sat in a par-ent’s lap across a table from E1, who manipulated two pup-pets distinguished only by their T-shirts (one orange, oneyellow). Before the puppets were two bowls filled, respec-tively, with graham crackers, and either green beans orCheerios (Fig. 1a). E1 directed the infant’s attention tothe two puppets. The first puppet approached a bowl andmade eating noises, followed by either a positive(‘‘Mmmm, yum, I like that!’’) or a negative (‘‘Ewww, yuck,

N. Mahajan, K. Wynn / Cognition 124 (2012) 227–233 229

I don’t like that!’’) verbal/emotional reaction. The samepuppet then went to the other bowl, made eating noisesagain, gave the opposite reaction, and returned to its origi-nal position. The second puppet repeated the actions/vocalizations of the first puppet, starting with the oppositebowl (thus indicating opposite preferences). The puppetexpressing the same taste preferences as the baby wetermed the similar puppet; the other we termed the dissim-ilar puppet.

After the puppet returned to its original position, E1 re-moved the food bowls and turned away. A second experi-menter (E2), blind to both infant’s and puppets’ foodpreferences, offered the puppets to the infant, conductingthe Infant-Chooses-Puppet measure. Choice was coded asthe first puppet the infant touched while looking at thatpuppet.

The following were counterbalanced across subjects:side, t-shirt color, and order for similar and different pup-pets, and order of expression (positive/negative first).

2.2. Results and discussion

Infants reliably chose the similar puppet (14 of 16 in-fants in the Different-foods condition, p = .0021 by a bino-mial test; 13 of 16 infants in the Similar Foods condition,p = .0106; infants’ performance in these two conditionsdid not differ, Pearson’s Chi-Square Test, v2

1 ¼ 0:237,p = 0.626) (Fig. 1b). These results indicate that infants at-tended to the puppets’ food choices, and that this relativelytrivial basis for comparison to self was sufficient to engen-der a disparity in infants’ preferences. Infants showed thispattern regardless of which food they themselves pre-ferred. In the Different Foods condition, 10 of 12 grahamcracker-choosers and 4 of 4 green bean-choosers preferredthe similar puppet; the pattern of responses for these twogroups did not differ (p = 1.000 by a Fisher’s Exact Test). In

Fig. 1b. Percentage of infants’ choices directed at the simi

the Similar Foods condition, similarly, 9 of 10 grahamcracker-choosers and 4 of 6 Cheerios choosers preferredthe similar puppet; again these patterns did not differ(p = 0.518). A second coder coded all babies and was 96%reliable with E2; using the secondary coder’s coding didnot alter the pattern or significance of results. As therewas no difference in the performance of infants who choseone snack versus another in this experiment or others, wehave collapsed these groups for the remainder of thispaper.

3. Experiment 1b: Salience manipulation

Adults’ preference for similar others appears to rest atleast in part on a cognitive comparison to self. Findingsshow that increasing the cognitive salience of subjects’similarity to another individual along a given dimension(for example, common membership in a group) enhancessubjects’ preference for similar others (e.g. Hong & Harrod,1988; Stephenson et al., 1976). To examine the issue ofdevelopmental continuity and address whether this holdsfor infants’ preference for similar others as well, we modi-fied the procedures of Experiment 1a so as to reduce thesalience to subjects of their own preferences. In this‘‘Low-Salience’’ Experiment 1b, subjects selected betweenthe two snacks only after having already seen the puppets’snack preferences and chosen between the two puppets.Thus, their own snack preference was far less salient atthe time they chose between the two puppets than itwas in Experiment 1, in which subjects were ‘‘reminded’’of their own snack preference by choosing between thetwo snacks before choosing between the puppets. If infants’preference for similar over different others, like that ofadults, is influenced by the salience of a distinction, thentheir preference for similarity should be significantly re-duced by this manipulation.

lar and different puppets in Experiments 1a and 1b.

230 N. Mahajan, K. Wynn / Cognition 124 (2012) 227–233

3.1. Methods

3.1.1. Participantswere sixteen 11.5-month-old infants (mean: 11;17,

range 11;7–11;27); half were placed into a Different Foodscondition, half into a Similar Foods condition. An addi-tional three infants were excluded due to experimenter er-ror (2) or refusal to choose (1).

Fig. 2a. Puppets and mittens used in Experiment 2.

3.1.2. ProcedureThe procedure was identical to Experiment 1 except in

the ordering of the three phases. E1 conducted thePuppets-Choose-Food task followed by the Infant-Chooses-Puppet task in the testing room, after which E2(blind to infants’ puppet choice) conducted the Infant-Chooses-Food task in the reception room.

3.2. Results and Discussion

Infants did not show the same pattern as in Experiment1a. Only 7 of 16 infants (p = .8036 by a binomial test) pre-ferred the similar puppet (3 of 8 infants in the Similar-foods condition, ns; 4 of 8 infants in the Different-foodscondition, ns; with no difference in performance acrossthe two conditions). This differed significantly from in-fants’ performance in Experiment 1a (Pearson’s Chi-SquareTest, v2

1 ¼ 8:521, p = 0.0018, one-tailed). A second codercoded all babies and was 93% reliable with E2; again, usingthe secondary coder’s coding did not alter the pattern orsignificance of results.

In Experiments 1a and 1b, we used modest, even trivialdifferences in food preference as a marker of similarity anddifference. While these can be considered minimal dimen-sions upon which to base social preferences, they are nottruly arbitrary dimensions. Older children and adults pre-fer not only those who share slight but genuine intrinsicsimilarities to themselves (such as a preference for Kleeover Kandinsky paintings; Tajfel, Billing, Brundy, &Flament, 1971), but also prefer individuals with a randomlyassigned similarity (e.g. divided into groups by a coin toss:Tajfel et al., 1971), and will even treat harmfully those inanother such-defined group (Sherif, Harvey, White, Hood,& Sherif, 1954/1961).

To test whether infants hold similarly extreme biases, inExperiment 2 we compared a condition in which infantsand puppets actively chose between two items with a con-dition in which an experimenter randomly assigned itemsto infants and puppets.

4. Experiment 2

4.1. Methods

4.1.1. Participantswere 32 11.5-month-old infants (range 10;26–11;29).

Five additional infants were excluded due to fussiness(1), experimenter error (3), or refusal to choose (1).

4.1.2. ProcedureThe procedure for the Choice Condition was identical

to that of Experiment 1a, except that instead of choosingbetween two foods, infants chose between orange and yel-low mittens resting in the palm of E1’s hand (the Infant-Chooses-Mittens task). After moving to the testing room,E1 put the chosen mittens on the infant’s hands, wherethey remained for the rest of the procedure. Next, in thePuppets-Choose-Mittens presentation, each puppet ap-proached piles of orange and yellow mittens on the tablein turn, expressing a positive attitude towards one pairand a negative attitude towards the other, and was thendressed in its preferred-color mittens (Fig. 2a).

In the Assignment Condition, in the Infant-Receives-Mittens task, E1 randomly selected a pair from the twopiles of mittens and put them on the baby’s hands, saying,‘‘I’m going to give you yellow/orange mittens.’’ The infantkept the mittens on for the rest of the study. E1 then placeda pair of mittens on one puppet, saying, ‘‘I’m going to givethis puppet yellow/orange mittens.’’ E1 next did the samewith the second puppet, with the other color mittens.

Infants in both conditions were then encouraged to se-lect the puppet of their choice. In both conditions, thedimension of comparison – mitten color – was visuallypresent and highly salient, as infants and puppets stillwore their mittens during infants’ choice. Thus, this exper-iment asked whether a salient and obvious, but utterlyarbitrary, dimension of contrast was sufficient to generatesocial preference.

4.2. Results and discussion

Infants in the Choice Condition reliably chose the simi-lar puppet (13 of 16 infants, p = 0.0106 by a binomial prob-ability test; Fig. 2b). In contrast, infants in the AssignmentCondition showed no preference (8 of 16 infants chose thelike-mittened puppet, ns; significantly different from theChoice infants (Pearson’s Chi-Square Test, v2

1 ¼ 3:463,p = 0.0314, one-tailed). A second coder was 96% reliablewith E2, and using this secondary coding did not alterthe pattern or significance of results. Thus, while infantsgenerate social preferences on the basis of trivial

Fig. 2b. Percentage of infants’ choices directed at the similar and different puppets in Experiments 2.

N. Mahajan, K. Wynn / Cognition 124 (2012) 227–233 231

similarities to self, they – unlike older humans – do not ap-pear to do so on the basis of truly arbitrary similarities gen-erated through random assignment, suggesting that thebias for similar others expands with development.

5. General discussion

Examining the origins and development of social prefer-ences for similarity is an important step in understandinghow they are formed, the factors that support and main-tain them, and, ultimately, how the negative consequencesof these social preferences, such as intergroup bias andprejudice, can be mitigated. Our findings demonstrate thatinfants in their first year assess others on the basis of sim-ilarity to self, and prefer more-similar others to those lesssimilar. These studies provide, to our knowledge, the firstevidence for these tendencies in preverbal infants.

Like older children and adults, even a minuscule non-arbitrary difference is sufficient to trigger a similarity bias(e.g. infants’ preference for the more-similar other did notvary with degree of difference; it was equally strong in thesimilar-foods and the different-foods conditions). More-over, infants’ preference for the similar appears to entaila comparison of other to self, in which individuals are iden-tified as ‘‘like me’’ or ‘‘not like me’’. These findings suggestthat intergroup attitudes generally, and ingroup biasesspecifically, are not entirely the product of cultural trans-mission and experience, but are rooted in early-emergingtendencies to identify others as similar to or different fromoneself, and to prefer the former to the latter. Importantly,our results from Experiment 2 are consistent with otherfindings indicating that social experiences do play a rolein the specific biases older children (e.g. McGlothlin &Killen, 2010) and adults (Pettigrew & Tropp, 2000) display;this development may occur with increasing exposure tocultural norms and values, as children observe patternsof social preference in adults and older children in theirculture.

What is the psychological nature of infants’ preferencefor the similar? The preferences observed in our studiescannot be based on a simple associationist account inwhich infants associate puppets with favored or disfavoreditems. In our tasks, both puppets interacted to the same ex-tent and duration with both choice items; what differedwas solely their expressed attitudes toward each item.Thus, our results are best explained in terms of a prefer-ence in infants for similarity to self.

The precise nature of the cognitive processes driving in-fants’ preference for the similar remains an open question.First, consider the results in our High- and our Low-sal-ience Experiments 1a and 1b, in which reducing the cogni-tive salience of infants’ own food preference diminishedtheir bias for similar others (as has been found in adults;Hong & Harrod, 1988; Stephenson et al., 1976). One possi-ble explanation for this differential performance is that in-fants evaluate others in relation to themselves, preferring‘‘like me’’ over ‘‘not like me.’’ Alternatively, infants may ex-tend their own attitude for a given item to other individu-als who express an affinity for – an emotional ‘‘closenessto’’ – that item, identifying, for example, an individualwho likes a ‘‘yucky’’ item as ‘‘yucky.’’ On the first account,lessening the cognitive availability of infants’ own atti-tudes in Experiment 1b correspondingly lessened theirability to compare the puppets’ attitudes to their own; onthe second, it lessened their ability to extend their attitudefor each item to the individual who positively affiliatedwith that item.

Second, there are different possible explanations for thecontrast observed in our Choice and Random-Assignmentconditions in Experiment 2. Infants’ lack of preference inthe Random-Assignment condition could reflect a prefer-ence for those who possess an enduring trait, such as a per-sonal preference, in common with the infant, but not forthose who share a transitory trait: the trait of wearing mit-tens of a given color stops as soon as the mittens are re-moved. Alternatively, infants’ lack of preference for the

232 N. Mahajan, K. Wynn / Cognition 124 (2012) 227–233

similar puppet in our Random-Assignment condition couldindicate that for infants, similarities in attitude are moresignificant bases on which to generate social preferencesthan are mere perceptual similarities. Indeed, recent find-ings suggest this is the case for preschool-aged children,who prefer a playmate with similar tastes to their own,but not one wearing a sticker matching their own (Fawcett& Markson, 2010).

Third, it is important to note that in our experiments,participants preferred an individual who made the samechoice they themselves made. The pattern of performanceobserved in Experiments 1 and 2 could indicate that in-fants prefer others who hold similar tastes to themselves,but they could also reflect that infants prefer others whoexpress similar tastes – that is, that infants prefer individu-als who publicly endorse their own views over those whopublicly contradict them. We favor the first explanationfor two reasons. First, in our experiments, infants selectedtheir preferred snack or mitten color in our reception areaand the puppets were not present during this phase, so didnot ‘witness’ infants’ choices. Infants only met the puppetsupon subsequently being brought into the testing room;thus the puppets’ expressed opinions could not easily beseen as public endorsement or correction of the infant’stastes.1 Second, using a similar methodology to the studiesreported here, Mahajan and Wynn (2010) found that infantsof this same age maintained a preference for the similar overthe dissimilar puppet even when the puppets made theirfood choices prior to the infant making his or her own foodchoice; thus it was the infant who expressed, respectively,public agreement and disagreement with the views of thetwo puppets, rather than the other way around. These factssuggest that it was the holding of an opinion, rather than thepublic expression of that opinion, that drove infants’ prefer-ence for similar over dissimilar others in our studies.

Finally, it is important to ask whether infants’ prefer-ence for the more-similar is driven primarily by an attrac-tion towards similarity to self, by an aversion to differencefrom self, or whether both inclinations exert equal influ-ence. Studies investigating intergroup cognition in adultsubjects suggest that intergroup attitudes often rest dis-proportionately upon heightened valuation of ingroupmembers, rather than derogation of outgroup members(e.g. Brewer, 1979, 1999). It will be important for future re-search to examine how closely the early preference forsimilar others we have documented here compares to thatof adults in this respect, so as to gain a more completeunderstanding of both developmental continuity anddevelopmental change in intergroup attitudes. Beyondthe dyadic similarity we have explored here, shared traitsand attributes are key markers that separate different so-cial groups from one another; additional research could ex-plore whether similar processes underlie biased attitudesin larger groups of individuals.

1 Several recent findings indicate that infants of this age and younger areacutely sensitive to the presence or absence of an individual in a scene, andthat they make accurate inferences about the beliefs and knowledge of anindividual based on what that individual has or has not had perceptualaccess to (e.g. Kovacs, Teglas, & Endress, 2010; Luo, 2011).

Acknowledgments

The authors would like to thank Kiley Hamlin, LizzyMadva, Alicia Fujii, Annie Spokes, and Rita McNamara fortheir help running and coding these studies. This workwas supported by NSF Grant #BCS-0715557 and NIH Grant#R01 MH 081877 to the second author, and an NSF Grad-uate Fellowship to the first author. The contents of thismanuscript are solely the responsibility of the authorsand do not necessarily represent the official views of NSFor NIH. The authors declare no competing financialinterests.

References

Aboud, F. E. (1988). Children and prejudice. New York: Blackwell.Abrams, D., & Hogg, M. A. (1990). Social identification, self-categorization,

and social influence. European Review of Social Psychology, 1, 195–228.Allport, G. W. (1954/1979). The nature of prejudice (25th anniversary ed.).

Cambridge, MA: Perseus Books.Bar-Haim, Y., Ziv, T., Lamy, D., & Hodes, R. M. (2006). Nature and nurture

in own-race face processing. Psychological Science, 17, 159–163.Baron, A. S., & Banaji, M. R. (2006). The development of implicit attitudes:

Evidence of race evaluations from ages 6, 10 & adulthood.Psychological Science, 17(1), 53–58.

Berscheid, E., & Walster, E. H. (1969). Interpersonal attraction. Addison-Wesley Publishing.

Bettencourt, B. A., Miller, N., & Hume, D. L. (1999). Effects of numericalrepresentation within cooperative settings: Examining the role ofsalience in in-group favouritism. British Journal of Social Psychology,38, 265–287.

Bigler, R. S., Brown, C. S., & Markell, M. (2001). When groups are notcreated equal: Effects of group status on the formation of intergroupattitudes in children. Child Development, 72(4), 1151–1162.

Bigler, R. S., Jones, L. C., & Lobliner, D. B. (1997). Social categorization andthe formation of intergroup attitudes in children. Child Development,68(3), 530–543.

Botwin, M. D., Buss, D. M., & Shackelford, T. K. (1997). Personality andmate preferences: Five factors in mate selection and maritalsatisfaction. Journal of Personality, 65(1), 107–136.

Brewer, M. B. (1979). Ingroup bias in the minimal intergroup situation: Acognitive motivational analysis. Psychological Bulletin, 86, 307–324.

Brewer, M. B. (1999). The psychology of prejudice. Ingroup love oroutgroup hate? Journal of Social Issues, 55(3), 429–444.

Buss, D. M., & Barnes, M. (1986). Preferences in human mate selection.Journal of Personality and Social Psychology, 50(3), 559–570.

Byrne, D. (1971). The attraction paradigm. New York: Academic Press.Byrne, D., Clore, G. L. J., & Worchel, P. (1966). Effect of economic

similarity-dissimilarity on interpersonal attraction. Journal ofPersonality and Social Psychology, 4(2), 220–224.

Clore, G. (1976). Interpersonal attraction: An overview. In Thibaut,Spence, & Carson (Eds.), Contemporary topics in social psychology.Morristown, NJ: General Learning Press.

Doise, W., Csepeli, G., Dann, H. D., Gouge, C., Larsen, K., & Ostell, A. (1972).An experimental investigation into the formation of intergrouprepresentations. European Journal of Social Psychology, 2, 202–204.

Dunham, Y., Baron, A. S., & Banaji, M. R. (2008). The development ofimplicit intergroup cognition. Trends in Cognitive Sciences, 12(7),248–253.

Dunham, Y., Baron, A. S., & Carey, S. (2011). Consequences of ‘minimal’group affiliations in children. Child Development, 82(3), 793–811.

Fawcett, C. A., & Markson, L. (2010). Similarity predicts liking in 3-year-old children. Journal of Experimental Child Psychology, 105, 345–358.

Goldman, J. A., Rosenzweig, C. M., & Lutter, A. D. (1980). Effect ofsimilarity of ego identity status on interpersonal attraction. Journal ofYouth and Adolescence, 9(2), 153–162.

Hong, O., & Harrod, W. (1988). The role of reasons in the ingroup biasphenomenon. European Journal of Social Psychology, 18, 537–545.

Huston, A. C. (1983). Sex-typing. In E. M. Hetherington (Ed.) & P. H.Mussen (Series Ed.), Handbook of child psychology: Vol. 4.Socialization, personality, and social development, (pp. 387-486).New York: Wiley.

Kinzler, K. D., Dupoux, E., & Spelke, E. S. (2007). The native language ofsocial cognition. Proceedings of the National Academy of Sciences, 104,12577–12590.

N. Mahajan, K. Wynn / Cognition 124 (2012) 227–233 233

Klohnen, E. C., & Luo, S. (2003). Interpersonal attraction and personality:What is attractive – self similarity, ideal similarity, complementarity,or attachment security? Journal of Personality and Social Psychology,85(4), 709–722.

Kovacs, A., Teglas, E., & Endress, A. (2010). The social sense: Susceptibilityto others’ beliefs in human infants and adults. Science, 330,1830–1834.

Leonardelli, G. J., & Brewer, M. B. (2001). Minority and majoritydiscrimination: When and why. Journal of Experimental SocialPsychology, 37, 468–485.

Luo, Y. (2011). Do 10-month-old infants understand others’ false beliefs?Cognition, 121, 289–298.

Mahajan, N., & Wynn, K. (2010). Developmental Origins of Ingroup Bias:Early Attitudes Towards Similar and Dissimilar Others. Talk presentedat the XVIIth Biennial International Conference on Infant StudiesSymposium: ‘‘Infants’ and Toddlers’ Expectations About SocialGroups,’’ Baltimore, MD.

McGlothlin, H., & Killen, M. (2010). How social experience is related tochildren’s intergroup attitudes. European Journal of SocialPsychology: Special Issue: Children’s Intergroup Attitudes (Guesteditors: J. Degner & Y. Dunham), 40, 625–634.

Miller, A. G. (1972). Effect of attitude similarity-dissimilarity on theutilization of additional stimulus inputs in judgments of interpersonalattraction. Psychonomic Science, 26(4), 199–203.

Mullen, B. (1983). Operationalizing the effect of the group on theindividual: A self-attention perspective. Journal of ExperimentalSocial Psychology, 19, 295–322.

Mullen, B., Brown, R., & Smith, C. (1992). Ingroup bias as a function ofsalience, relevance and status: An integration. European Journal ofSocial Psychology, 22, 103–122.

Patterson, M. M., & Bigler, R. S. (2006). Preschool children’s attention toenvironmental messages about groups: Social categorization and theorigins of intergroup bias. Child Development, 77, 847–860.

Pettigrew, T. F., & Tropp, L. R. (2000). Does intergroup contact reduceprejudice? Recent meta-analytic findings. In S. Oskamp (Ed.),

Reducing prejudice and discrimination: The Claremont Symposiumon Applied Social Psychology (pp. 93–114). Mahwah, NJ: Erlbaum.

Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. (1954/1961). Intergroup conflict and cooperation: The robbers cave experiment.Norman, OK: University Book Exchange.

Singh, R., & Ho, S. Y. (2000). Attitudes and attraction: A new test of theattraction, repulsion and similarity-dissimilarity asymmetryhypotheses. British Journal of Social Psychology, 39(2), 197–211.

Skowronski, J. J., & Lawrence, M. A. (2001). A comparative study of theimplicit and explicit gender attitudes of children and college students.Psychology of Women Quarterly, 25, 155–165.

Stephenson, G. M., Skinner, M., & Brotherton, C. J. (1976). Groupparticipation and intergroup relations: An experimental study ofnegotiation groups. European Journal of Social Psychology, 6, 51–70.

Tajfel, H. (1970). Experiments in intergroup discrimination. ScientificAmerican, 223, 96–102.

Tajfel, H. (1982). Social psychology of intergroup relations. Annual Reviewof Psychology, 33, 1–39.

Tajfel, H., Billing, M., Brundy, R., & Flament, C. (1971). Socialcategorization and intergroup behavior. European Journal of SocialPsychology, 1, 149–178.

Tan, D. T. Y., & Singh, R. (1995). Attitudes and attraction: A developmentalstudy of the similarity-attraction and dissimilarity-repulsionhypotheses. Personality and Social Psychology Bulletin, 21, 975–986.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. (1987).Rediscovering the social group: A self-categorization theory. Oxford:Blackwell.

Walster, E. G., Walster, W., Berscheid, E., & Dion, K. (1971). Physicalattractiveness and dating choice. A test of the matching hypothesis.Journal of Experimental Social Psychology, 7, 173.

Wilder, D. A. (1986). Cognitive factors affecting the success of intergroupcontact. In S. Worchel & W. G. Austin (Eds.), Intergroup relations(pp. 49–66). Chicago: Nelson-Hall.