Psychology of Human Kin Recognition: Heuristic Cues, Erroneous Inferences, and Their Implications

Justin H. Park Mark Schaller University of Groningen University of British Columbia

Mark Van Vugt University of Kent at Canterbury

Humans possess explicit, rule-based, and culturally determined systems for identifying kin, but kinship inferences are also inﬂuenced implicitly by cue-based mechanisms found commonly across the animal kingdom. These mechanisms are fallible. An evolutionarily informed signal-detection analysis suggests that (a) cue-based kin recognition may sometimes be biased in favor of false-positive errors, resulting in implicit kinship inferences even in the presence of nonkin, and (b) the tendency toward this inferential error may vary predictably in response to speciﬁc developmental and contextual circumstances. This analysis has important implications for a wide variety of psychological phenomena (especially in the realms of person perception, interpersonal attraction, and prosocial behavior) and leads to the deduction of many novel hypotheses.

Keywords: incest avoidance, kin recognition, nepotism, person perception, signal detection

How do we know who our genetic relatives broad range of phenomena that, on the surface, are? One answer is that we have been explicitly would seem irrelevant to kinship. taught. But there is much more than that to the psychology of kin recognition. Like many nonhuman species, people implicitly infer kinship Adaptive Context of Kin Recognition from a variety of cues. This cue-based process—which may operate independently of ra- Kin-recognition processes appear to have tional assessments of kinship—has many impli- evolved in many animal species to facilitate cations that have yet to be fully explored in adaptive behavior, and there are at least two psychology. In this article, we review theory distinct functions that may be served by kin- and research on the psychology of cue-based recognition processes: inbreeding avoidance kin recognition and discuss its wide-ranging and nepotism. implications. These implications extend to a First, because inbreeding tends to increase homozygosity of deleterious recessive alleles, inbred offspring are less likely to survive; when they do survive, they show a variety of Justin H. Park, University of Groningen, Groningen, The deﬁcits in physical and psychological func- Netherlands; Mark Schaller, University of British Colum- tioning, such as depressed immunocompe- bia, Vancouver, British Columbia, Canada; Mark Van Vugt, tence and reduced intelligence (Badaruddoza, University of Kent at Canterbury, Canterbury, United King- 2004; Penn & Potts, 1999; Reid, Arcese, & dom. We thank Josh Ackerman, Liz Dunn, Tim Fawcett, David Keller, 2003). Given such ﬁtness costs, be- Herring, Satoshi Kanazawa, Rob Kurzban, Lidewij Niezink, havioral tendencies to avoid sex with close Elske van der Vaart, and several anonymous reviewers for kin would have been adaptive; indeed, avoid- their helpful comments. ance of sex with close kin is observed in a wide Correspondence concerning this article should be ad- dressed to Justin H. Park, Department of Psychology, Uni- variety of animal species and across virtually all versity of Groningen, Grote Kruisstraat 2/1, 9712 TS Gro- human cultures (D. E. Brown, 1991; Waldman, ningen, The Netherlands. E-mail: [email protected] 1988).

215 216 PARK, SCHALLER, AND VAN VUGT

Second, although indiscriminate altruism is have written records and good memories” costly and maladaptive, altruism can evolve if (Dawkins, 1989, p. 99). Moreover, we readily it leads to benefits (measured in terms of the apply relational rules in order to identify reproductive fitness of the altruist) that novel individuals according to some set of outweigh their costs. There are many concep- explicitly labeled kinship categories. (For in- tually distinct means through which this con- stance, following a set of simple rules, one dition can be achieved, and each of these can infer that one’s brother’s newborn daugh- evolutionary mechanisms is the basis for ter is one’s niece.) much theory and research (e.g., Axelrod & Of course, these solutions to the problem of Hamilton, 1981; Tooby & Cosmides, 1996; kin recognition are, from an evolutionary per- Trivers, 1971; Zahavi & Zahavi, 1997). One spective, relatively recent inventions. They of these evolutionary mechanisms is relevant depend on recently evolved—and uniquely here—the mechanism of kin selection (Ham- human—capacities for language, symbolic rep- ilton, 1964). If a set of genes predisposes an resentation, and their complex cultural conse- individual toward assisting a closely related other, there is a high probability that these quences. A more complete, and ultimately more same genes also exist in the bodies of the illuminating, perspective on human kin recog- recipients by virtue of common descent. Such nition requires that we consider the existence of genes—that underlie nepotism—can then in- additional, more evolutionarily ancient mecha- crease in frequency within populations. The nisms as well. normative conditions under which altruistic An extensive body of work in the biological tendencies can evolve via kin selection are sciences, complemented by a rapidly growing typically summarized by the formula rB Ͼ C body of research in the psychological sciences, (C ϭ fitness costs experienced by the altruist, suggests that much of human kin recognition is B ϭ fitness benefits accrued by the recipient, characterized by a highly automated, and some- r ϭ relatedness between the altruist and re- times fallible, cue-based process. Cue-based cipient). The logic of kin selection predicts kin-recognition mechanisms are observed greater altruistic tendencies toward closer rel- widely across the animal kingdom, suggesting atives, and there is substantial evidence for an evolutionary history that far predates the this. Within many animal species, including emergence of Homo sapiens (Porter, 1987; humans, altruistic tendencies are sensi- Rendall, 2004; Waldman, 1987). Given their tive—at least to some degree—to genetic re- cross-species prevalence, it is likely that these latedness (e.g., Burnstein, Crandall, & mechanisms persist and operate within humans Kitayama, 1994; Chapais, Savard, & as well, regardless of the fact that humans now Gauthier, 2001; Sherman, 1977). possess additional, more cognitively sophisti- Both of these behavioral tendencies (in- cated means of assessing kinship. This is be- breeding avoidance and nepotism) are predi- cause the evolution of increasingly complex cated on the ability to discriminate kin from psychological mechanisms has generally oc- nonkin (and closer kin from more distant kin). curred by adding to—rather than replacing— These adaptive behavioral tendencies there- existing mechanisms (Geary, 2005). Thus, fore require some sort of kin-recognition mechanisms. uniquely human inference processes involving recently evolved neocortical structures may be Hidden World of Cue-Based complemented by other, more “primitive” pro- Kin-Recognition Processes cesses that operate independently, often outside of conscious awareness and control. As we re- What psychological processes might solve view below, much research suggests that this is the problem of kin recognition among hu- the case for cue-based kin-recognition mecha- mans? At first glance, this question may ap- nisms. Consequently, the operation of this cue- pear boring and the answer obvious: “We based process is relatively hidden, having many know who our relatives are because we are subtle implications for human social cognition told, because we give them names, because and behavior, most of which have escaped the we have formal marriages, and because we attention of psychologists. KIN RECOGNITION 217

Operation of Cue-Based Kin-Recognition association facilitate kin recognition across sev- Mechanisms eral animal species, ranging from ants to pigs (Puppe, 1998; Singer & Espelie, 1998). Cue-based kin-recognition mechanisms can Location and association are important kin- be conceptualized as simple decision rules con- ship cues for many primate species as well forming to the logic of if–then statements (e.g., (Silk, 2002). On the basis of her work on chim- “If I grew up with this individual, then she is panzees, Goodall (1986) noted the following: kin”). To articulate the operation of these mech- Attitudes toward kin are shaped, to a large extent, by anisms, we consider a series of more specific the degree of familiarity of the individuals concerned, questions. First, what exactly are the stimuli that and depend on close and prolonged association. It is signal kinship, and what are the psychological only logical, therefore, that helping behaviors will on responses to the perception of those stimuli? occasion be extended to familiar individuals even Second, just how reliable are the kinship cues, when they are not very close kin. (p. 380, emphases in the original) and what sorts of kin-recognition errors might occur? Third, to what extent might these stimu- Nonhuman primates also avoid mating with lus–response associations vary across persons each other on the basis of familiarity (Paul & and contexts? We address these questions in Kuester, 2004). After reviewing evidence across turn. many nonhuman primate species, Rendall (2004) concluded that “the available evidence paints a fairly consistent picture—namely that kin are ‘rec- Kinship Cues and Functional Responses ognized’ via the familiarity accruing to them dur- Reviews of the animal literature typically ing development” (p. 302). identify two broad classes of kinship cues, each Humans also rely heavily on early-life spatial associated with a distinct set of mechanisms that proximity (e.g., coresidence) as a kinship cue. appear to facilitate kin recognition. Some of Opposite-sex siblings are more likely to engage these cues pertain to the spatial location of a in incestuous activity if they had been separated target individual and the ensuing familiarity. for significant periods during childhood (Bevc Other cues pertain more directly to phenotypic & Silverman, 1993, 2000). People with a his- features of the target individual. tory of coresidence with opposite sex siblings are more likely to judge others’ incestuous acts as morally wrong (Fessler & Navarette, 2004; Spatial Location and Early Association Lieberman, Tooby, & Cosmides, 2003), and (Familiarity Cues) people who grow up together—even when they are unrelated—tend to find each other unappeal- As studies of imprinting have revealed, new- ing as sexual partners (a phenomenon known as born goslings rely on spatial location to identify the Westermarck effect); this effect has been their parent. The fallible nature of this heuristic is observed across multiple cultures (Shepher, evident in the fact that, simply as a function of 1971; Walter & Buyske, 2003; Wolf, 1970). spatial proximity, goslings sometimes imprint on Coresidence duration influences altruism as entirely unrelated individuals (including, well (Anderson, Kaplan, & Lancaster, 1999). famously, decidedly un-goose-like organisms Recent research indicates that the role of such as Konrad Lorenz, a zoologist who studied coresidence duration in kin detection may de- the phenomenon). Likewise, warblers often treat pend on the availability of other, even more any egg in their nest as their own, even if that egg highly diagnostic kinship cues. Lieberman, belongs to another species (Winfree, 1999). They Tooby, and Cosmides (2007) reasoned that rec- also treat any bird that emerges from those eggs as ognition of younger siblings need not depend their own offspring, even if that emergent bird solely on coresidence duration, as there exist appears un-warbler-like. Cross-fostering experi- other, more reliable cues (e.g., observing one’s ments have revealed that warblers are often unable mother feeding the sibling). Consistent with this to discriminate between their own and foster eggs, reasoning, these researchers found that coresi- or between their own and foster nestlings, indicat- dence duration predicts nepotistic and incest- ing a heavy reliance on spatial location (Komdeur, avoidance responses more strongly in the ab- Richardson, & Burke, 2004). Spatial location and sence of other, more highly diagnostic cues. In 218 PARK, SCHALLER, AND VAN VUGT the presence of such cues, coresidence duration in parent–offspring recognition (Yamazaki, had little impact. Beauchamp, Curran, Bard, & Boyse, 2000). When using phenotypic resemblance as a Phenotypic Resemblance (Similarity Cues) kinship cue, individual animals are not necessarily comparing target features to their own. Among animals with more complex social ar- Because the kin prototype can be based on other rangements (including humans), location- and as- individuals (e.g., littermates, nestmates, and sociation-based mechanisms may be insufficient, coresidents), individuals may be comparing tar- because they do not clearly distinguish kin from get features to those of other individuals nonkin, and because they are susceptible to ex- (Hepper, 1991). In other words, it is not neces- ploitation by nonkin (Silk, 2002). In addition to sary for animals to be aware of their own phe- familiarity, many animals infer kinship on the notypic features in order to infer kinship via a basis of phenotype matching—a process in which phenotype-matching mechanism. Nevertheless, phenotypic features of target individuals are com- there is evidence that some mammals may infer pared against a kin prototype (Hauber & Sherman, kinship directly on the basis of phenotypic re- 2001). Similarity to that prototype is then used as semblance between self and other (e.g., Mateo an indicator of kinship. A variety of similarity & Johnson, 2000; Sun & Mu¨ller-Schwarze, cues serve as indicators of kinship across many 1997). And although there is little direct evi- animal species. dence, it seems likely that nonhuman apes— Among avian species, auditory signals are with a substantial capacity for self-reflection commonly used as kinship cues (Beecher, 1988). and self-knowledge (e.g., Gallup, 1970; In both human and nonhuman animals, more Povinelli et al., 1997)—may use various forms highly related individuals have more similar odors of self—other resemblance as kinship cues. (e.g., Heth, Todrank, & Johnston, 1999; Roberts et Humans appear to use self–other facial sim- al., 2005), and many animals—ranging from ze- ilarity as a cue for kinship. Fathers favor chil- brafish to humans—appear to use odor similarity dren who look more like them (Apicella & to discriminate kin from nonkin (Gerlach & Ly- Marlowe, 2004; Burch & Gallup, 2000), and siak, 2006; Weisfeld, Czilli, Phillips, Gall, & perceived self–infant resemblance predicts de- Lichtman, 2003). sire to adopt unfamiliar infants in hypothetical Although most kin-recognition heuristics scenarios, especially among men (Volk & operate at levels far removed from genes, one Quinsey, 2002, 2007). Furthermore, people circumscribed area of genes appears to be in- report greater willingness to assist unrelated volved in kin recognition: the major histocom- children whose faces have been experimentally patibility complex (MHC). There is evidence manipulated to resemble their own (DeBruine, that individuals of many mammalian species— 2004; Platek, Burch, Panyavin, Wasserman, & including humans—prefer mates with dissimi- Gallup, 2002; Platek et al., 2003, 2004). Such lar MHC genes (detected via odor cues), which effects of facial resemblance are not limited to are more likely to reside within bodies of unre- adults’ responses to children: Adults also related individuals (Penn & Potts, 1999; spond favorably to facially similar adults Wedekind & Fu¨ri, 1997; Wedekind, Seebeck, (DeBruine, 2002). There is also evidence that Bettens, & Pepke, 1995). A study of romantic one particular form of nepotism—the tendency couples found that the proportion of shared to attend more closely to the mating relation- MHC alleles is negatively correlated with wom- ships of closer kin—is partially mediated by en’s sexual responsivity to their partners and perceived physical similarity (Faulkner & positively correlated with women’s number of Schaller, 2007). Results reported by DeBruine extrapair sexual partners (Garver-Apgar, Gang- (2005) reveal that the effects of facial similarity estad, Thornhill, Miller, & Olp, 2006). Cross- cannot be attributed to the kinship-irrelevant fostered mice have been found to favor mates tendency to simply respond more positively to whose MHC differs from their foster parents’ similar others: When presented with the faces of MHC, rather than their own (Yamazaki et al., opposite sex strangers that were either facially 1988), which indicates that association and phe- similar to or different from themselves, students notype matching processes likely interact. rated the similar others as more trustworthy and There is also evidence that the MHC plays a role also less sexually attractive. KIN RECOGNITION 219

Among primates, similarity cues may not be typic similarity automatically activates seman- limited to surface features, but may include tic concepts associated with kinship (Park & inferred cues such as age similarity (e.g., Schaller, 2005). Other kinds of behavior- Alberts, 1999; Widdig, Nu¨rnberg, Krawczak, facilitating cognitions may also be activated by Streich, & Bercovitch, 2001). Research in be- the perception of kinship cues, but the specific havioral genetics indicates that more highly re- nature of these cognitions may vary depending lated people are more similar to each other on a on the specific interpersonal context. In altruis- variety of attitudes, values, and personality tic contexts, the activated cognitions are likely characteristics (e.g., Olson, Vernon, Harris, & to be those consistent with approach behavior Jang, 2001), and people appear to use these (e.g., perceptions of trustworthiness and likabil- similarities as kinship cues. Using a reaction- ity). In sexual contexts, however, activated cog- time measure to assess implicit cognitive asso- nitions may be more negative in evaluative ciations, Park and Schaller (2005) found that, in tone, so as to promote physical avoidance. relation to a target person with dissimilar atti- Activation of context-specific emotions and tudes, a target person with similar attitudes was motivational states. These cognitive more likely to implicitly activate kinship cog- responses are likely to be complemented by nitions. This effect was stronger among individ- specific emotional responses. The arousal of a uals who more readily trust their intuitions, a specific emotion (and the accompanying moti- finding that substantiates the heuristic nature of vational state) is often essential to facilitate an this process. adaptive behavioral response (Keltner, Haidt, & Shiota, 2006). If kin-recognition mechanisms Functional Psychological Responses to evolved to facilitate specific forms of adaptive Kinship Cues behavior, then kinship cues are expected to trigger whatever specific emotional response facil- If cue-based kin-recognition mechanisms itates those behaviors. Different behaviors are evolved in such a way as to serve fitness- adaptive under different circumstances; thus, enhancing behavior (e.g., altruism and sexual contextual cues are expected to modulate the aversion toward kin), then the perception of specific emotional response triggered by the kinship cues would be expected to activate rel- perception of a kinship cue. evant cognitive and emotional responses that Within a mating context, the adaptive re- facilitate specific fitness-enhancing behaviors. sponse to close kin is sexual aversion. There- Kinship index. Lieberman et al. (2007) sug- fore, within such a context, it has been argued— gested that there exists an implicit mechanism and empirically documented—that the percep- that, on the basis of the availability of relevant tion of kinship ought to arouse an emotional cues, computes something like a kinship index response at odds with sexual arousal: disgust for each target individual. This index serves as (Fessler & Navarette, 2004). The capacity for an estimate of self–target relatedness and serves disgust likely emerged to serve other, more as input to psychological responses pertaining primitive functions such as avoidance of toxins to incest avoidance and nepotism. and parasites (Curtis, Aunger, & Rabie, 2004; Activation of categorical and semantic con- Rozin & Fallon, 1987). But once in place, the cepts into working memory. For any kind of evolved mechanisms that produce disgust may inferred estimation of kinship to influence be- have been co-opted to produce this aversive havior, some sort of cognitive response is nec- emotional response in potentially incestuous essary. Most fundamentally, the perception of circumstances. kinship cues should activate cognitive re- Outside of the mating domain, kinship cues sponses that conceptually pertain to kinship are likely to arouse emotions that promote (though not necessarily consciously). There is prosocial (rather than aversive) responses. Kin evidence that nonhuman primates are able to are associated with feelings of emotional close- represent kin as a distinct conceptual category ness or social bonding, and altruistic behavior is (e.g., Cheney & Seyfarth, 1990); studies of mediated in part by these feelings (e.g., young children have yielded similar findings Korchmaros & Kenny, 2001; Neyer & Lang, (e.g., Springer, 1992, 1996). As mentioned in 2003). A subjective sense of closeness is not an the preceding section, the perception of pheno- emotion, per se. Perhaps a purer emotional re- 220 PARK, SCHALLER, AND VAN VUGT sponse to perceived kinship—especially under encountering kin is relatively low (and thus the circumstances when the other person’s welfare likelihood of false-negative errors is also low), a is at stake—is the experience of empathy (i.e., bias in favor of false-negative, underinclusive sympathy or compassion). Empathy is a power- errors is more likely to be adaptive. This bias ful predictor of altruistic behavior (e.g., Batson may manifest in the implicit requirement for a & Shaw, 1991; Eisenberg & Miller, 1987). Ac- relatively high standard of evidence for an in- cordingly, many researchers have speculated ference connoting kinship and thus in the use of that one function of empathy is to serve as an only highly diagnostic kinship cues. The behav- emotional response to perceived kinship (Hoff- ioral implication is that, in these contexts, indi- man, 1981; Krebs, 1987; Schaller, 2003). In- viduals may be more likely to treat kin as non- deed, there is evidence that people feel more kin, rather than the reverse. empathy toward unrelated individuals who are On the other hand, within ecological contexts subjectively familiar or similar (e.g., Krebs, in which the relative frequency of encountering 1975). kin is especially high (and thus the likelihood of false-positive errors is relatively low), a bias in Signal-Detection Problem and favor of false-positive, overinclusive errors may Its Consequences actually be adaptive. This bias may manifest in a relatively low standard of evidence for an The cues that animals use to infer kinship are inference connoting kinship and in the use of correlated to varying degrees with actual ge- more fallible cues to kinship. The correspond- netic relatedness, but the correlations are far ing behavioral implication is that, in these con- from perfect. As a result, cue-based kin recog- texts, individuals may be more likely to treat nition creates a classic signal-detection prob- nonkin as kin, rather than the reverse. lem. At the most simplistic level of analysis, a target individual might be kin or nonkin, and the Relative Costs of False-Positive and cue-based mechanisms might lead to an infer- False-Negative Kin-Recognition Errors ence of kinship or nonkinship. Consequently, two distinct kinds of errors may occur: One may Adaptive solutions to signal-detection prob- infer kinship when, in reality, the target indi- lems are those that yield fitness outcomes that vidual is nonkin (false positive), or the target are greater than those yielded by other plausible individual may actually be kin, but one may fail solutions. Indeed, people tend to draw infer- to make this inference (false negative). ences that are predictably biased in such a way There is, of course, no perfect solution to this as to minimize the most costly form of error, signal-detection problem. Decision rules de- even though this leads to an increase in the less signed to reduce the likelihood of false positives costly form of error (Haselton & Nettle, 2006). increase the likelihood of false negatives, and Thus, to predict how the signal-detection prob- vice versa. With respect to kin recognition, the lem in kin recognition might be resolved, one “best” solution depends importantly on two must consider not only base rates of kin versus considerations: (a) the base-rate likelihood of nonkin but also the cumulative fitness outcomes encountering kin versus nonkin and (b) the rel- that might accrue from any bias toward false- ative costs associated with each kind of error. positive errors or false-negative errors. Consider recognition errors within the con- Relative Likelihood of Kin Versus text of mating. Any single false-positive error Nonkin Interactions has the consequence of inhibiting sexual intercourse with a nonkin member, whereas any From an evolutionary perspective, the rela- single false-negative error has the consequence tive superiority associated with a bias toward of allowing sexual intercourse with kin. The net false-positive or false-negative kin-detection er- costs associated with the two errors may depend rors varies as a function of the relative fre- on many factors, including actual degree of quency of encountering kin and nonkin (Reeve, relatedness. Mating between genetically close 1998). kin is especially likely to incur net fitness costs, Reeve’s (1998) analysis suggests that within whereas mating between more genetically dis- ecological contexts in which the frequency of tant kin may not incur net fitness costs, espe- KIN RECOGNITION 221 cially if there are few alternative mating oppor- thus, an individual’s developmental circum- tunities available. stances may have important consequences on Different kinds of costs must be considered the specific cues that later connote “kin” to that within altruistic contexts as well. Any false- individual. And like most complex psychologi- positive kin-recognition error may precipitate cal processes that operate in the service of social an act of altruism toward nonkin. Any false- interaction, kin-recognition mechanisms are negative kin-recognition error may result in a likely to be influenced by internal and external failure to assist a needy kin member. If the cost regulatory cues that heuristically signal infor- of the former error is greater, a bias in favor of mation about the costs and benefits associated false-negative errors (treating kin as nonkin) is with particular psychological responses likely to occur, and vice versa. Because these (Schaller, Park, & Kenrick, 2007). costs must be considered in terms of fitness implications, these relative costs may vary Learning and Developmental Processes greatly depending on several factors. For instance, if an individual has few fitness-relevant Any mechanism that compels an individual resources available, a bias in favor of false- to treat spatially proximate others as kin re- negative errors may be more adaptive. But if an quires some learning mechanism enabling that individual has abundant resources, a bias in individual to acquire the association between favor of false-positive kin-recognition errors specific proximate individuals (e.g., Konrad may be more adaptive. Lorenz) and kin-relevant responses (e.g., imprinting). Indeed, it is because learning plays Implication such a prominent role that kin recognition sometimes becomes conspicuously derailed These preceding considerations have several (e.g., erroneous imprinting or sexual aversion specific implications, which we discuss more toward unrelated coresidents). fully below. More generally, perhaps the most Learning is also important for mechanisms interesting implication is that people may some- that are responsive to phenotypic similarity times (perhaps even often) psychologically re- (Hepper & Cleland, 1999). Within any particu- spond to nonkin as though they are kin. This lar species, there may be a general tendency for implication resonates with Goodall’s (1986) phenotypic similarity to signal kinship. But be- speculations about the role of familiarity as a cause there is inevitable phenotypic variation kinship cue in chimpanzees: between families (or litters or nests), there will also be individual differences in the specific Patterns of comfort and reassurance, helping and shar- ing, that have emerged over thousands of years in the phenotypic features that reliably distinguish kin context of the mother–child and family relationship from nonkin. For this reason, individuals must and that are firmly embedded in the genetic endow- learn the specific phenotypic features that reli- ment, may be released not only by the distress or pleas ably serve as kin-connoting cues for them. For of biological kin, but by similar appeals from unrelated but highly familiar individuals. (p. 380) instance, in some birds, hatchlings are innately prepared to learn certain “songs,” and they learn And, importantly, this tendency toward over- the specific features of these songs primarily inclusive kin recognition is likely to be flexible, from their kin—or, more accurately, from their varying according to the context within which nestmates (Sharp, McGowan, Wood, & cue-based inferences occur. Hatchwell, 2005). As a result, the songs performed by any single bird more closely resem- Functional Flexibility of ble those performed by their own kin (or nest- Kin-Recognition Mechanisms mates, who are usually kin), which is why the songs are a reliable cue for kinship. As the review of kinship cues makes clear, Furthermore, although both sexual aversion animals are not born with knowledge of which and altruism are adaptive responses to kin, the individuals are kin. Rather, animals are predis- functional utility of each response may differ posed to associate kin-connoting features with across one’s life stage. Nepotistic interactions specific individuals. In some cases, animals are beneficial across the life cycle; thus, it must first learn which features connote kinship; would be beneficial for individual animals to 222 PARK, SCHALLER, AND VAN VUGT respond positively to kin (e.g., maintaining more certain than paternity. For this reason, physical proximity) from the moment of birth. males may be more dependent on heuristic kin- Sexual-aversion mechanisms, on the other ship cues to correctly identify kin. This has hand, become functional only when one reaches implications for the extent to which men and sexual maturity. Therefore, behavioral re- women use superficial cues (e.g., phenotypic sponses relevant to sexual aversion (e.g., main- similarity) to inform decisions to allocate re- taining physical distance) may emerge later in sources to children. Indeed, empirical research life. Studies of kin recognition in zebrafish reveals that, in comparison with women, the found just this pattern: Juveniles preferred to prosocial inclinations of men are more strongly maintain physical proximity to odors of kin; influenced by a child’s facial resemblance among adults, when given the choice between (Platek et al., 2002, 2003, 2004; Volk & odors of opposite sex kin or nonkin, females Quinsey, 2002, 2007). preferred nonkin whereas males did not show a The sex differences emerging from paternity preference (Gerlach & Lysiak, 2006). uncertainty extend to more distant relatives as well. For example, paternal grandfathers face Sex Differences two instances of relatedness uncertainty, paternal grandmothers and maternal grandfathers There are sex differences that can be ex- face one, and maternal grandmothers face none. pected on the basis of evolutionary cost–benefit It follows that the level of grandparental invest- considerations. In humans, as in other mam- ment should be highest for maternal grandmoth- mals, women are required to make a substan- ers and lowest for paternal grandfathers—a tially greater investment than men in the pro- hypothesis that has been supported (Euler & duction of offspring. Women thus incur a Weitzel, 1996; Laham, Gonsalkorale, & von heavier cost to their reproductive fitness when- Hippel, 2005). A further hypothesis is that ever a sexual coupling leads to the conception grandparents may also be differentially depen- of an unfit child. One consequence is that, dent on kinship cues, with paternal grandparents among animals with higher female investment, being the most dependent, which has also re- inbreeding-avoidance responses tend to be ceived some support (Euler & Weitzel, 1996). stronger among females (e.g., Chapais & Mig- The same line of reasoning can be applied to nault, 1991; Gerlach & Lysiak, 2006). Women other kin (e.g., uncles, aunts, and cousins) to are also more likely than men to find the pros- investigate differences in level of investment pect of incest aversive, and they are more sen- (e.g., Jeon & Buss, 2007) and in the impact of sitive than men to the heuristic cue of early life kinship cues. coresidence (Fessler & Navarette, 2004; Lieber- man et al., 2003; Walter & Buyske, 2003). Impact of Family Environment These differential fitness costs associated with incest also imply that the costs of false- For humans, the family serves as an impor- negative kin-recognition errors will be greater tant developmental context. Idiosyncratic as- for women than for men, which further implies pects of an individual’s immediate family may that women may show a stronger bias toward also exert a lasting influence on the tendency to false-positive kin-recognition errors—for ex- use certain kinds of phenotypic similarities, ample, a more pronounced tendency to treat rather than others, as kinship cues. Develop- familiar or similar nonkin as though they are mentally early inferences about kinship may be kin. Recent empirical evidence is consistent based substantially on physical proximity (e.g., with this hypothesized sex difference. In the coresidence), which may then set the stage for domain of both sexual behavior and prosocial the acquisition of additional kinship cues based behavior, women are more likely than men to on their probabilistic association with coresi- respond to close friends as though they were kin dent “kin” (cf. Hepper, 1991). This may be (Ackerman, Kenrick, & Schaller, 2007). relevant especially to the acquisition, and later Additional sex differences in the use of kin- use, of cues based on phenotypic similarity. ship cues, and their behavioral consequences, There are many different kinds of heritable may result from differences in parental cer- phenotypic features, including not only differ- tainty. Among mammals, maternity is generally ent morphological features (e.g., size of nose KIN RECOGNITION 223 and color of hair) but also psychological fea- in assistance given to actual family members tures (e.g., attitudes and personality traits) Be- (Amato, 1993). cause of the probabilistic nature of genetic Other aspects of the broader ecological envi- transmission, close kin within any specific fam- ronment may also influence the nature of kin- ily may be more objectively similar on some of recognition biases. For instance, because offspring these features than on others. Thus, although of incest tend to have depressed immune systems, both facial similarity and attitude similarity may it follows that the costs of incest are especially implicitly connote kinship (DeBruine, 2005; high within ecological contexts in which patho- Park & Schaller, 2005), specific individuals gens are prevalent. There is evidence that patho- may learn to use facial similarities rather than gen prevalence influences mating preferences and attitudinal similarities as kinship heuristics, or behaviors in a variety of ways (Gangestad & Buss, vice versa, depending on which specific kind of 1993; Low, 1990; Penton-Voak, Jacobson, & similarity was objectively more diagnostic of Trivers, 2004). One intriguing hypothesis is that kinship within their specific family environ- individuals in more highly pathogen-prevalent ment. ecologies may show a stronger bias toward over- Early-life family environment might also inclusive kinship inferences and stronger incest- have implications for an individual’s tendency avoidant responses. toward making overinclusive (versus underinclusive) kin-recognition errors. Recall that the functional implications of these biases depend Situation-Specific Variation in the substantially on the base rate of kin in the local Operation of ecology (Reeve, 1998). Therefore, individuals Kin-Recognition Mechanisms who grow up in larger families may be more likely to acquire a persistent tendency toward The heuristic process of kin recognition may overinclusive kin recognition. also be responsive to specific features in the immediate situation. One way in which the immediate situation influences the operation of Impact of the Broader kin-recognition processes is by modulating the Ecological Environment nature of the specific psychological response that is stimulated by the perception of a kinship Any tendency toward an overinclusive kin- cue. In a potential mating situation, the percep- recognition bias may be influenced by the tion of a kinship cue may trigger aversive psy- broader social ecology as well. Base rates for chological responses (e.g., disgust and reduced interactions with kin versus nonkin are influ- sexual attraction), whereas in other situations, enced not only by the number of kin in the local the same kinship cue may instead trigger re- environment but also by the number of nonkin. sponses that motivate more approach-oriented And this may vary according to the local social prosocial behavior (e.g., empathy and percep- density: People who grow up in rural areas tend tions of trustworthiness). This point is illus- to have a higher number of interactions with trated by research on the context-specific effects kin, in comparison with people who grow up in of facial resemblance (DeBruine, 2005). densely populated urban environments (Amato, Even within a particular functional domain, 1993). One consequence is that people raised in additional situation-specific information may rural environments may be more likely to de- influence the extent to which a particular kin- velop a tendency toward an overinclusive kinship cue triggers the associated response. Any recognition bias—they may more readily use a information bearing on the relative costs of greater array of cues as kin-connoting heuristics false-negative versus false-positive kin-recogni- and thus, all else being equal, be more likely to tion errors may influence the extent to which respond to nonkin as kin. This offers one pre- perceivers are likely to be overinclusive or un- viously unidentified reason why people in rural derinclusive in their implicit inferences about areas are more likely than those in urban areas kinship. Any information bearing on the bene- to assist strangers (Steblay, 1987). Moreover, fits and costs of kinship-relevant responses (i.e., this explanation is compatible with the finding sexual aversion or empathy) may influence the that this urban–rural difference does not emerge strength of those psychological responses. 224 PARK, SCHALLER, AND VAN VUGT

Additional Implications for Specific cally explained by drawing on concepts derived Psychological Phenomena from balance theory (Heider, 1958), they are also consistent with the heuristic operation of That people treat kin and nonkin differently is kin-recognition mechanisms. uncontroversial. More intriguing is the implica- In principle, two things can be “similar” in an tion that people may sometimes respond—at infinite number of ways. Consequently, addi- cognitive, emotional, and behavioral levels—to tional factors must constrain the kinds of simi- nonkin as though they were kin. Because larities that are noticed and used as kinship cue-based kin-recognition mechanisms operate cues. One factor may be the degree of diagnos- largely outside of conscious awareness, the re- ticity of the similarity concerned. Indeed, per- sulting responses may not be subjectively expe- ceived attitudinal similarity is more reinforcing, rienced as even pertaining to kinship. Never- and leads to greater interpersonal liking, when theless, these cue-based mechanisms may be the self–other similarity occurs on attitudes that activated more often than commonly acknowl- are more highly heritable (Crelia & Tesser, edged, coloring people’s psychological and be- 1996; Tesser, 1993). More generally, one would havioral responses toward a broad array of other expect similarity to exert a more powerful pos- people. Another implication is that this ten- itive effect on impressions and interpersonal dency is likely to be functionally flexible. There behavior when that specific form of similarity are likely to be predictable individual differ- is, objectively, more diagnostic of actual kin- ences rooted in specific aspects of an individu- ship. Similarity in facial features is more highly al’s developmental environment. Furthermore, diagnostic of actual kinship than, say, similarity any tendency toward overinclusive kinship in- in clothing style. Thus, while sartorial similarity ferences is also likely to be moderated by pre- may have nontrivial effects on impressions and dictable features of the environment. behavior, the effects of facial similarity are These implications may manifest in numer- likely to be stronger and may manifest across a ous specific psychological phenomena that may greater range of situations and individuals. seem, on a more superficial analysis, to have nothing to do with kinship whatsoever. We have Effects of Similarity and Familiarity mentioned some of these phenomena already on “Attraction” (e.g., the tendency to trust facially similar strangers). We now proceed to discuss addi- If perceived similarity serves as a kinship tional implications for specific phenomena per- cue, and if people are averse to sexual relations taining to social cognition and behavior. Our with kin, then doesn’t this analysis contradict goals are threefold: (a) to shed new explanatory the classic similarity–attraction effect? No: A light on these various phenomena, (b) to suggest closer examination reveals that the so-called that these superficially dissimilar phenomena similarity–attraction effect is not about sexual may be linked by a common conceptual frame- attraction—it is about liking (Byrne, 1961; work, and (c) to identify novel hypotheses that Byrne et al., 1971; Byrne, London, & Reeves, still await rigorous empirical testing. 1968; Byrne & Nelson, 1965). As Bem (1996) noted, “similarity may promote friendship, Effects of Similarities on Impressions compatibility, and companionate love, but it is and Behavior dissimilarity that sparks erotic/romantic attraction and passionate love” (p. 323). Thus, our Our impressions about and behaviors toward analysis of kin-recognition heuristics is fully other people are influenced—sometimes power- compatible with the classic similarity–liking ef- fully—by surface cues. Among other things, fect (a label that is more appropriate and less these impressions and behaviors are influenced confusing than the similarity–attraction label). by perceived similarities between self and other. It is also fully compatible with Bem’s (1996) When a person happens to be similar to us, we theory about erotic responses to exotic people are more likely to respond favorably to that and with Sternberg’s (1986) triangular theory of person (e.g., Burger, Messian, Patel, del Prado, love in which intimacy (i.e., liking) and passion & Anderson, 2004; Oates & Wilson, 2002). (i.e., sexual attraction) are presumed to be sep- Although these effects of similarity are typi- arate components. Because women pay higher KIN RECOGNITION 225 costs for any incestuous coupling, the impact of mere exposure effect) may—as with the effects similarity on sexual aversion may be stronger of similarity—be moderated by sex, family among women (e.g., Garver-Apgar et al., 2006). composition, and the broader local ecology. Furthermore, given inherent trade-offs associated with different mating strategies Perceptions of Similarity (Gangestad & Simpson, 2000), the tendency for similarity to dampen sexual attraction (and for Any objective similarity between two indi- dissimilarity to spark sexual attraction) may be viduals is distinct from the perception of simi- especially strong within short-term relationship larity between those individuals. What similar- contexts in which obtaining genetic benefits is ities are perceived? What similarities matter in a more central. Indeed, facial resemblance re- psychological sense? A number of scholars duces attraction specifically within a short-term, have grappled with these kinds of questions but not long-term, relationship context (e.g., Goodman, 1972; Medin, Goldstone, & (DeBruine, 2005; see also Little, Penton-Voak, Gentner, 1993). A consideration of kin- Burt, & Perrett, 2003, and Perrett et al., 2002). recognition processes generates novel hypothe- Within long-term relationship contexts, the ben- ses that may yield additional answers. If people efits of similarity (e.g., compatibility) may out- use similarity as a kinship cue, and if people are weigh the potential costs, which may help to somewhat sensitive to the diagnosticity of spe- explain why similarity predicts marital longev- cific kinds of cues, then people may be espe- ity (e.g., Caspi & Herbener, 1990). Thus, the cially attuned to—and make inferential use of— similarity–attraction hypothesis, as typically similarities along features that are generally construed, may be insufficiently nuanced. One more diagnostic of kinship. I might judge a must consider the type of attraction, relation- stranger to be more similar to me if that stranger ship, and similarity when investigating the psy- shares my last name, rather than my first name. chological consequences of similarity. I might judge two individuals to be more similar This analysis also suggests a conceptual con- if they happen to agree with each other on nection between the similarity–liking effect and attitude dimensions that are highly heritable the mere exposure effect, whereby repeated ex- than if they agree on a less heritable attitude posure to people and other objects leads to dimension. These conjectures remain untested. greater liking for those people or objects (Zajonc, 2001). Because exposure breeds famil- Effect of Empathy on Helping Behavior iarity, if the effect of familiarity on liking is rooted in a cue-based kin-recognition mecha- If empathy is a functional response to the nism of considerable antiquity, then it makes perception of kinship, then—as a result of re- sense that the affective response may occur peated association—empathy may come to even in the absence of any meaningful cognitive serve as a heuristic kinship cue itself. The effect mediation (Zajonc, 1980). This is not to suggest of empathy on helping may therefore represent that the mere exposure effect is due solely to the a manifestation of overinclusive kin recogni- operation of a kin-recognition process (surely tion, which may help resolve some questions other processes play a role as well; Zajonc, concerning the effect of empathy. 2001), but it is intriguing to consider the possi- That empathy facilitates helping behavior is bility that a kin-recognition heuristic does con- not in doubt; what is controversial is why. There tribute. If so, an interesting implication is that has been debate as to whether empathy facili- the effects of repeated exposure on positive tates helping through activation of a genuinely perceptions of others may not extend to percep- altruistic goal (the empathy–altruism hypothe- tions of sexual attraction (which was also noted sis) or through the activation of some other, by Bem, 1996). In addition, just as the effects of more egoistic goal (Batson et al., 1989; Batson similarity on liking may be stronger when sim- & Shaw, 1991; Maner et al., 2002; Schaller & ilarity occurs along a dimension that is more Cialdini, 1988). The focus of debate has been on diagnostic of kinship, the same may be true for the psychologically meaningful contents of ac- the effects of familiarity on liking. Furthermore, tivated goal states. If, however, empathy serves given the flexibility of kin-recognition heuris- as a mediator of nepotism, then the behavioral tics, the effects of familiarity (including the consequences of empathy may often occur 226 PARK, SCHALLER, AND VAN VUGT reflexively, without meaningful deliberation Interpersonal Relationships and without any higher-order goal state activated into working memory (Schaller, 2003). There appear to be a few “core” human rela- This may explain why the effects of empathy tionships that underlie most social interactions are difficult to eliminate through procedures (e.g., Bugental, 2000; Fiske, 1992). One type of that appeal to helpers’ rational appraisal of core relationship that has been recurrently pro- means that might satisfy egocentric goals (Bat- posed is the communal relationship. Such rela- son & Shaw, 1991). tionships are typified by family relations and A kin-recognition perspective on empathic close friendships and are defined in part by a helping implies that the experience of empathy tendency for individuals to provide assistance may be used as a kinship heuristic in some and bestow benefits to each other largely un- circumstances more than others. It is more conditionally. Communal relationships can be likely to be applied overinclusively (and thus to contrasted with other types of relationships, induce helping of nonkin) when the costs of such as those defined by the norms of reciprocal helping are relatively low. Indeed, the usual exchange (e.g., Clark & Mills, 1979). The psy- empathy–helping relationship disappears when chological distinction between communal and the costs of helping are made more substantial exchange relationships resembles the logical (Batson, O’Quin, Fultz, Vanderplass, & Isen, distinction between two evolutionary processes 1983). that can give rise to a capacity for helping Another hypothesis implicit in this analysis is behavior, one based on kin selection (Hamilton, that the empathy–helping effect may be muted 1964) and the other based on reciprocity if there exists diagnostic information that (Trivers, 1971). This may not be coincidental. strongly disqualifies a target person from being Reciprocity is fundamental to the definition of perceived as kin. Several studies show exactly an exchange relationship and governs interac- this pattern of results: When people empathize tions within such a relationship. Similarly, kin- with those who are perceived to be fundamen- ship may be fundamental to the establishment of tally dissimilar, empathy has little effect on a communal relationship (whether or not the helping (Maner et al., 2002; Stu¨rmer, Snyder, communal relationship involves actual kin). Kropp, & Siem, 2006). For instance, Stu¨rmer et The perception of any kin-connoting heuristic al. (2006) found that empathy had little effect cue (e.g., phenotypic similarity) may dispose on helping when the target person was from a individuals toward the development of a com- cultural outgroup, the sort of information that munal rather than an exchange relationship, would quickly disqualify a person from being producing psychological responses that are con- considered kin. sistent with communal relations (e.g., trust) and inhibiting responses that are emblematic of ex- Effects of Kinship Cues on Aggression change relationships. and Antisocial Behavior Various kinds of psychological phenomena (e.g., quick detection of cheaters and desire for Just as people are more giving to kin than to equity) typify exchange relationships (e.g., Cos- nonkin, so too are they less hostile toward kin mides & Tooby, 1992; Walster, Walster, & than toward nonkin (Daly & Wilson, 1988). It is Berscheid, 1978). We know already that some not just actual kinship that inhibits hostility and of the psychological tendencies underlying ex- aggression; heuristic kinship cues appear to change relationships are attenuated within com- have analogous effects. Anger and aggression munal relationships (e.g., Clark, 1984). These are less likely to be directed toward others— tendencies may also be attenuated simply by the even strangers—who are perceived to be more presence of heuristic kinship cues such as fa- similar or with whom we empathize (e.g., miliarity and empathy. There is some evidence Harmon-Jones, Vaughn-Scott, Mohr, Sigelman, consistent with this conjecture. In a study & Harmon-Jones, 2004; Miller & Eisenberg, involving chimpanzees (who, like humans, are 1988). Thus, some of the variability in antiso- sensitive to equity and other elements of social cial reactions toward strangers might be under- exchange), individual animals were offered a stood as a manifestation of the heuristic mech- low-value reward after observing a partner anisms of kin recognition. chimpanzee receiving a high-value reward. KIN RECOGNITION 227

When their partner was relatively unfamiliar, kin (Van Vugt & Van Lange, 2006). Moreover, the chimpanzees displayed displeasure by refus- research has found that perceived similarity ing the low-value reward; but when their partner boosts social identity: People identify more was highly familiar (though not more geneti- strongly with their group when they believe that cally related), the chimpanzees tolerated the in- the group members share their attitudes (Van equity (Brosnan, Schiff, & de Waal, 2005). Vugt & Hart, 2004). Among humans, there is evidence that the ex- Especially noteworthy is the fact that group perience of empathy alters people’s responses in members deliberately exploit kinship senti- Prisoner’s Dilemma games, which are normally ments to enhance group solidarity. Words such dictated by rules of reciprocity. For instance, as fraternity, brotherhood, and soul sisters are individuals experiencing empathy are more often used to arouse emotions normally re- likely to cooperate, even after defection by the served for kin—and political speech is espe- partner (Batson & Ahmad, 2001). cially evocative when it uses such terms (e.g., Because close friends present many cues that Salmon, 1998). heuristically imply kinship (e.g., similarity and familiarity), people may sometimes treat close Intergroup Prejudice and Discrimination friends, at a psychological level, as though they were kin (Ackerman et al., 2007). Of course, Intergroup prejudice manifests in a variety of this does not limit the importance of psycholog- familiar guises. People are more likely to help ical processes that are specific to friendships ingroup members and aggress against outgroup (e.g., Tooby & Cosmides, 1996). It suggests members (Donnerstein & Donnerstein, 1973; simply that friendship may provide an espe- Gaertner & Dovidio, 1977); when given the cially common context for overinclusive kin opportunity to allocate resources to others, peo- recognition. ple selectively favor ingroup members over outgroup members (Brewer, 1979). These forms of Intragroup Processes behavioral discrimination parallel the many ways in which people favor kin over nonkin. Of Humans tend to form and sustain cooperative course, multiple psychological processes con- groups comprising genetically unrelated indi- tribute to intergroup prejudice and discrimina- viduals, and the psychology of kinship may tion, many of which have nothing to do with have played an important role in the kinship. Still, in addition to those processes, it is development of such groups. Because the costs possible that the psychology of kin recognition of unreciprocated investment are lower if the plays some role. recipient is kin, it has been theorized that the Historically, kin tended to live in a single tendency toward reciprocal altruism, and of co- clan, tribe, or other form of coalitional ingroup. operative behavior more generally, emerged ini- Thus, although a person’s ingroup membership tially in interactions among kin (Alexander, may not by itself be strongly diagnostic of kin- 1987). In hunter–gatherer societies, which re- ship, a person’s outgroup membership can be semble the social structure of ancestral human strongly diagnostic of nonkinship. It follows groups, there is a high degree of genetic relat- that negative responses toward outgroup mem- edness among the residents (Chagnon, 1997). bers may be muted by interventions that exploit Therefore, kin-recognition processes may offer the processes of kin recognition. Many pieces of new insights for some group-level phenomena. existing evidence can be interpreted in this way. One specific intragroup phenomenon that Intergroup contact can reduce prejudice espe- might be seen as a manifestation of the heuristic cially when it results in true familiarity with psychology of kinship is social identity—the outgroup members (Wright, Brody, & Aron, tendency for people to incorporate group mem- 2005). Perceived similarity is associated with bership into their own self-concept (Ellemers, lower levels of prejudice and discrimination Spears, & Doosje, 2002). When people identify (Brewer & Campbell, 1976; R. Brown & with their group, their self-concept is defined at Abrams, 1986); empathy, too, can have this the group level, and they merge their own and effect (Stephan & Finlay, 1999). other members’ interests—that is, they treat If indeed prejudice-reduction interventions their fellow group members as they would treat based on contact, similarity, and empathy are 228 PARK, SCHALLER, AND VAN VUGT tapping into the heuristic psychology of kin people may also judge that there is a stronger recognition, there are implications for the likely moral obligation to help others who happen to outcomes of those interventions. One implica- share some sort of phenotypic similarity. These tion is that the effects may be mediated more by moral judgments may also be sensitive to lin- affect than by reason and thus may reduce prej- guistic cues that tap into the heuristic psychol- udices even if stereotypes remain largely un- ogy of kinship. People may judge that there is a changed. Consistent with this line of reasoning, stronger moral obligation to help others who are research has found that prejudice reduction at identified metaphorically as brothers or sisters explicit and implicit levels occurs somewhat in- or otherwise part of some symbolic family. dependently; moreover, implicit prejudice reduction tends to be more strongly associated with affective processes than with purely cognitive pro- Additional Issues and Directions for cesses (e.g., Rudman, Ashmore, & Gary, 2001). Future Research The psychology of kin recognition warrants Morality more attention than it has received in the study It has been argued that human conceptions of of human social cognition and behavior. Such morality are rooted in psychological processes attention will be productive in many ways. Most that evolved in response to persistent problems obviously, because kin-recognition mechanisms of survival and reproduction (Krebs & Janicki, are an important component of the human 2004). Consequently, moral judgments are of- psyche, the delineation of their nature contrib- ten not so much the product of reasoned con- utes to a more complete understanding of how sideration but are instead triggered, more or less humans navigate their social world. In addition, automatically, by crude heuristics, such as the the psychology of kin recognition provides a emotional experiences of disgust and empathy deep conceptual link between many different (Haidt, 2001; Pizarro, 2000). Not surprisingly, kinds of psychological phenomena that on the people have powerful, and not necessarily ra- surface may seem to have little in common. It tional, moral intuitions about kin relations. Peo- implies many new hypotheses bearing on these ple judge incestuous acts to be morally wrong many different phenomena. In this final section, even when they cannot articulate any logical we briefly discuss a few additional issues, each justification for that judgment (Haidt, 2001). of which indicates potentially fruitful directions People also feel a stronger moral obligation to for future research. individuals who are more closely genetically related (Kruger, 2001). Kin Recognition and Human Prosociality If indeed these sorts of moral judgments are intuitive rather than rational, then these same As we have seen, a signal-detection analysis sorts of moral judgments may be triggered by of cue-based kin recognition implies that people heuristic kinship cues and thus apply even out- sometimes make false-positive errors; accord- side of actual kin relations. Just as people may ingly, the implicit psychology of kin recogni- be morally repulsed by actual incest, they may tion may help explain many acts of altruistic also be morally repulsed by sexual relations behavior that are directed toward strangers. between truly unrelated individuals with a his- Does this mean that overinclusive kin recogni- tory of childhood coresidence (e.g., adoptive tion provides the exclusive answer to the ques- siblings) or some apparent familial connection tion of human altruism? No, of course not. (e.g., Woody Allen and and his adopted daugh- There is no shortage of answers to the question ter, Soon-Yi Previn). Also, just as negative re- of how altruism evolved in humans. It is clear actions to actual incest are stronger among that many processes conceptually independent women than among men (Fessler & Navarrete, of kin recognition—such as social exchange, 2004; Lieberman et al., 2003), negative reac- interdependence, coalitional alliances, and tions to these particular kinds of nonincestuous costly signaling—have contributed to the hu- relations may also be stronger among women. man capacity for altruism and cooperative Similarly, just as people judge that there is a behavior (e.g., Fehr & Ga¨chter, 2002; McAn- stronger moral obligation to help actual kin, drew, 2002; Richerson & Boyd, 2005; Tooby & KIN RECOGNITION 229

Cosmides, 1996). While overinclusive kin- Solomon, Johnson, Zaitchik, & Carey, 1996]). recognition processes do not provide the only This suggests that even in the absence of any explanation, it appears increasingly clear that formal knowledge of genetics, there exist intu- they must be part of the conversation. itive beliefs about some sort of physical essence Furthermore, although the different processes shared by genetic kin (Springer, 1992, 1996). If underlying human altruism may be conceptual indeed young children intuitively believe that independent, they may sometimes be triggered kinship manifests especially in physical proper- simultaneously. Just as kinship may be implic- ties, then there may be an innate tendency to- itly inferred from available cues, expectations ward inferring kinship from facial resemblance about reciprocity may also be inferred from and other manifestations of physical similarity, available cues. Similar cues may also indicate in which case the use of these particular cues whether an individual is a member of a coali- may be universal and robust. In contrast, learn- tional alliance. Some of the characteristics that ing processes may play a larger role in the serve as kinship cues may also serve as cues for tendency to use less intuitively appealing man- potential reciprocity or for membership in a ifestations of kinship (e.g., attitude similarity coalitional alliance. Thus, while our review has and emotional experiences) as kin-recognition focused on mechanisms through which these heuristics; the use of these less obvious cues cues heuristically trigger kin-relevant re- may be more dependent on additional variables. sponses, some of the same cues may also trigger an additional set of responses rooted in the Nature of Underlying Mechanisms implicit psychology of social exchange and coalitional alliances. This does not mean, As a means of generating predictions about however, that these other processes offer viable interpersonal behavior, it makes sense to treat alternative explanations for the entire set of kin recognition as an inferential process, and it specific cognitive, emotional, and behavioral can be useful to treat measurable outcome vari- phenomena we have reviewed here; they do not. ables (e.g., sexual attraction, perceptions of trustworthiness and helping behavior) as indic- Intuitive Theories About Kinship and ative of those underlying inferences. But these Kin Recognition indirect indicators do not easily reveal the actual nature of the underlying mechanisms. People depend on theories to understand the It remains unclear, for instance, whether the world. The theories held by young children can perception of kin-connoting cues actually re- be especially informative about the nature of sults in some cognitive representation of kin- intuitive theories—sets of beliefs that result ship, or whether these cues facilitate affective from innate predispositions rather than from and behavioral responses even in the absence of formal education. There is now abundant evi- any cognitive representation connoting kinship dence bearing on children’s intuitions about per se. If these cue-based mechanisms are as specific kinds of ontological categories, includ- evolutionarily ancient as we suspect, actual cog- ing physical objects, biological organisms, and nitive representations of kinship may not be a mental states (e.g., Gelman, 2004; Gergely & necessary precondition for kin-connoting be- Csibra, 2003). Do children have intuitive theo- havioral responses. Resolution of this question ries about kinship? If so, how do they relate to may require methods that attempt to tap more our conceptual analysis of kin-recognition directly into the semantic contents of cognitive mechanisms? representations (e.g., Park & Schaller, 2005). Some evidence suggests that certain kinds of An even more fundamental issue pertains to heuristic kinship cues may be more intuitive the impact of moderating variables (e.g., than others. Young children make a distinction cost–benefit ratio and frequency of actual kin between physical phenotypic properties (e.g., a in the local ecology): Do these moderating person’s height) and other personal characteris- variables have an impact on an implicit com- tics that are less immediately anchored in the putation of kinship (e.g., a “kinship index” of physical body (e.g., a person’s belief, and they the sort articulated by Lieberman et al., believe that kin are more likely to share the 2007), or on the extent to which that kinship former than the latter [Springer, 1996; but see computation triggers specific motivational 230 PARK, SCHALLER, AND VAN VUGT systems guiding behavior, or both? The an- and kin recognition within the biological sci- swer has important implications. If the effects ences. We have attempted to highlight the many are specific to the implicit kinship computa- ways in which an evolutionary cost–benefit tion itself, then these moderating variables are analysis can yield novel hypotheses at a psy- likely to have similar effects across various chological level of analysis. As this analysis different domains in which kinship matters makes clear, an evolutionary approach to kin- (e.g., helping behavior and sexual behavior). recognition suggests predictable ways in which If the effects occur on the implementation of the actual operation of these mechanisms is motivational systems, then these moderating likely to vary across persons and situations; this effects may differ across domains. Of course, approach also accords an important role to it is also possible that the implicit computa- learning. tion of kinship differs across different do- Given that many specific kinship cues are mains. There is evidence that some kin- learned, and that the associative mechanisms recognition cues exert effects on both nepo- through which they are learned are also used to tistic and incest-avoidant responses serve a wide variety of functions that have noth- (Lieberman et al., 2007), which is consistent ing to do with kinship, we must entertain some with a single computational mechanism, but it questions about human evolution that go well remains possible that there may be some dif- beyond a psychological level of analysis. Just ferences in implicit kinship inferences across what exactly did evolve? Did these particular functionally distinct domains. If so, then there elements evolve specifically to serve the func- may be differences in the sets of cues used in tion of kin recognition? Or did they evolve in mating versus resource-allocation contexts. response to other adaptive problems and then One additional consideration suggests the were co-opted in the service of kin recognition? possibility of additional complexities in the These are big-picture questions, extending nature of the underlying mechanisms. Kinship well beyond a psychological level of analysis. can be defined quantitatively, indexed by the Our ability to answer these questions will de- probable degree of genetic overlap between pend on advances in other scientific disciplines. two individuals, but it can also be defined By understanding the evolutionary roots of hu- categorically, according to the specific kind man kin-recognition mechanisms, we will be in of relation that exists between two persons a better position to predict how these mecha- (e.g., one’s expected genetic overlap with nisms actually operate in contemporary con- parents, siblings, and children is identical, but texts. We will more fully understand the many parents, siblings, and children represent qual- subtle ways in which the heuristic processes of itatively distinct categories of kin). These cat- kin recognition influence human cognition and egorical differences matter from a functional behavior. perspective because—for reasons distinct from genetic relatedness—behaviors directed toward different kinds of kin can have differ- References ent consequences for one’s reproductive fitness. Not surprisingly, therefore, these cate- Ackerman, J. M., Kenrick, D. T., & Schaller, M. gorical differences also influence behavioral (2007). Is friendship akin to kinship? Evolution responses (e.g., Burnstein et al., 1994). Yet, and Human Behavior, 28, 365–374. again, it remains unclear whether these differ- Alberts, S. C. (1999). Paternal kin discrimination in ences might occur on actual implicit inferences wild baboons. Proceedings of the Royal Society of about kinship, or on the motivational systems that London B, 266, 1501–1506. respond to these inferences, or both. Alexander, R. D. (1987). The biology of moral systems. New York: Aldine. Amato, P. R. (1993). Urban–rural differences in help- If Kin-Recognition Processes Evolved, ing friends and family members. Social Psychol- Exactly What Evolved, and Why? ogy Quarterly, 56, 249–262. Anderson, K. G., Kaplan, H., & Lancaster, J. (1999). In this article, we have located our discussion Paternal care by genetic fathers and stepfathers I: within an evolutionary framework. This seems Reports from Albuquerque men. Evolution and sensible given the abundant research on kinship Human Behavior, 20, 405–431. KIN RECOGNITION 231

Apicella, C. L., & Marlowe, F. W. (2004). Perceived Bugental, D. B. (2000). Acquisition of the algorithms mate fidelity and paternal resemblance predict of social life: A domain-based approach. Psycho- men’s investment in children. Evolution and Hu- logical Bulletin, 126, 187–219. man Behavior, 25, 371–378. Burch, R. L., & Gallup, G. G., Jr. (2000). Perceptions Axelrod, R., & Hamilton, W. D. (1981). The evolu- of paternal resemblance predict family violence. tion of cooperation. Science, 211, 1390–1396. Evolution and Human Behavior, 21, 429–435. Badaruddoza. (2004). Effect of inbreeding on Wech- Burger, J. M., Messian, N., Patel, S., del Prado, A., & sler intelligence test scores among North Indian Anderson, C. (2004). What a coincidence! The ef- children. Asia-Pacific Journal of Public fects of incidental similarity on compliance. Person- Health, 16, 99–103. ality and Social Psychology Bulletin, 30, 35–43. Batson, C. D., & Ahmad, N. (2001). Empathy- Burnstein, E., Crandall, C., & Kitayama, S. (1994). induced altruism in a prisoner’s dilemma II: What Some neo-Darwinian decision rules for altruism: if the target of empathy has defected? European Weighing cues for inclusive fitness as a function of the biological importance of the decision. Journal of Journal of Social Psychology, 31, 25–36. Personality and Social Psychology, 67, 773–789. Batson, C. D., Batson, J. G., Griffitt, C. A., Barrien- Byrne, D. (1961). Interpersonal attraction and atti- tos, S., Brandt, J. R., Sprengelmeyer, P., et al. tude similarity. Journal of Abnormal and Social (1989). Negative-state relief and the empathy– Psychology, 62, 713–715. altruism hypothesis. Journal of Personality and Byrne, D., Gouaux, C., Griffitt, W., Lamberth, J., Social Psychology, 56, 922–933. Murakawa, N., Prasad, M. B., et al. (1971). The Batson, C. D., O’Quin, K., Fultz, J., Vanderplass, M., ubiquitous relationship: Attitude similarity and at- & Isen, A. M. (1983). Influence of self-reported traction: A cross-cultural study. Human Rela- distress and empathy on egoistic versus altruistic tions, 24, 201–207. motivation to help. Journal of Personality and Byrne, D., London, O., & Reeves, K. (1968). The Social Psychology, 45, 706–718. effects of physical attractiveness, sex, and attitude Batson, C. D., & Shaw, L. L. (1991). Evidence for similarity on interpersonal attraction. Journal of altruism: Toward a pluralism of prosocial motives. Personality, 36, 259–271. Psychological Inquiry, 2, 107–122. Byrne, D., & Nelson, D. (1965). Attraction as a linear Beecher, M. D. (1988). Kin recognition in birds. function of proportion of positive reinforcements, Behavior Genetics, 18, 465–482. Journal of Personality and Social Psychology, 1, Bem, D. J. (1996). Exotic becomes erotic: A devel- 659–663. opmental theory of sexual orientation. Psycholog- Caspi, A., & Herbener, E. S. (1990). Continuity and ical Review, 103, 320–335. change: Assortative marriage and the consistency Bevc, I., & Silverman, I. (1993). Early proximity and of personality in adulthood. Journal of Personality intimacy between siblings and incestuous behav- and Social Psychology, 58, 250–258. ior: A test of the Westermarck hypothesis. Ethol- Chagnon, N. (1997). Yanomamo. New York: Holt, ogy and Sociobiology, 14, 171–181. Rinehart, & Winston. Bevc, I., & Silverman, I. (2000). Early separation and Chapais, B., & Mignault, C. (1991). Homosexual sibling incest: A test of the revised Westermarck incest avoidance among females in captive Japa- theory. Evolution and Human Behavior, 21, 151– nese macaques. American Journal of Primatol- 161. ogy, 23, 171–183. Chapais, B., Savard, L., & Gauthier, C. (2001). Kin Brewer, M. B. (1979). In-group bias in the minimal selection and the distribution of altruism in relation intergroup situation: A cognitive–motivational to degree of kinship in Japanese macaques (Ma- analysis. Psychological Bulletin, 86, 307–324. caca fuscata). Behavioral Ecology and Sociobiol- Brewer, M. B., & Campbell, D. T. (1976). Ethnocen- ogy, 49, 493–502. trism and intergroup attitudes: East African evi- Cheney, D. L., & Seyfarth, R. M. (1990). The rep- dence. New York: Sage. resentation of social relations by monkeys. Cogni- Brosnan, S. F., Schiff, H. C., & de Waal, F. B. M. tion, 37, 167–196. (2005). Tolerance for inequity may increase with Clark, M. S. (1984). Record keeping in two types of social closeness in chimpanzees. Proceedings of relationships. Journal of Personality and Social the Royal Society B, 272, 253–258. Psychology, 47, 549–557. Brown, D. E. (1991). Human universals. Philadel- Clark, M. S., & Mills, J. (1979). Interpersonal attrac- phia: Temple University Press. tion in exchange and communal relationships. Brown, R., & Abrams, D. (1986). The effects of Journal of Personality and Social Psychology, 37, intergroup similarity and goal interdependence on 12–24. intergroup attitudes and task performance. Journal Cosmides, L., & Tooby, J. (1992). Cognitive adap- of Experimental Social Psychology, 22, 78–92. tations for social exchange. In J. H. Barkow, L. 232 PARK, SCHALLER, AND VAN VUGT

Cosmides, & J. Tooby (Eds.), The adapted mind: Gangestad, S. W., & Buss, D. M. (1993). Pathogen Evolutionary psychology and the generation of prevalence and human mate preferences. Ethology culture (pp. 163–228). New York: Oxford Univer- and Sociobiology, 14, 89–96. sity Press. Gangestad, S. W., & Simpson, J. A. (2000). The Crelia, R. A., & Tesser, A. (1996). Attitude heritability evolution of human mating: Trade-offs and strate- and attitude reinforcement: A replication. Personality gic pluralism. Behavioral and Brain Sciences, 23, and Individual Differences, 21, 803–808. 573–644. Curtis, V., Aunger, R., & Rabie, T. (2004). Evidence Garver-Apgar, C. E., Gangestad, S. W., Thornhill, R., that disgust evolved to protect from risk of disease. Miller, R. D., & Olp, J. J. (2006). Major histocom- Proceedings of the Royal Society of London B, patibility alleles, sexual responsivity, and unfaith- 271, S131–S133. fulness in romantic couples. Psychological Sci- Daly, M., & Wilson, M. (1988). Homicide. Haw- ence, 17, 830–835. thorne NY: De Gruyter. Geary, D. C. (2005). The origin of mind: Evolution of Dawkins, R. (1989). The selfish gene (new ed.). Ox- brain, cognition, and general intelligence. Wash- ford, UK: Oxford University Press. ington, DC: American Psychological Association. DeBruine, L. M. (2002). Facial resemblance en- Gelman, S. A. (2004). Psychological essentialism in hances trust. Proceedings of the Royal Society of children. Trends in Cognitive Sciences, 8, 404–409. London B, 269, 1307–1312. Gergely, G., & Csibra, G. (2003). Teleological rea- DeBruine, L. M. (2004). Resemblance to self increases soning in infancy: The naı¨ve theory of rational the appeal of child faces to both men and women. action. Trends in Cognitive Sciences, 7, 287–292. Evolution and Human Behavior, 25, 142–154. Gerlach, G., & Lysiak, N. (2006). Kin recognition DeBruine, L. M. (2005). Trustworthy but not lust- and inbreeding avoidance in zebrafish, Danio re- worthy: Context-specific effects of facial resem- rio, is based on phenotype matching. Animal Be- blance. Proceedings of the Royal Society B, 272, haviour, 71, 1371–1377. 919–922. Goodall, J. (1986). The chimpanzees of Gombe: Pat- terns of behavior. Cambridge, MA: Harvard Uni- Donnerstein, E., & Donnerstein, M. (1973). Vari- versity Press. ables in interracial aggression: Potential ingroup Goodman, N. (1972). Seven strictures on similarity. censure. Journal of Personality and Social Psy- In N. Goodman (Ed.), Problems and projects (pp. chology, 27, 143–150. 437–447). New York: Bobbs-Merrill. Eisenberg, N., & Miller, P. A. (1987). The relation of Haidt, J. (2001). The emotional dog and its rational empathy to prosocial and related behaviors. Psy- tail: A social intuitionist approach to moral judg- chological Bulletin, 101, 91–119. ment. Psychological Review, 108, 814–834. Ellemers, N., Spears, R., & Doosje, B. (2002). Self Hamilton, W. D. (1964). The genetical evolution of and social identity. Annual Review of Psychol- social behaviour, I, II. Journal of Theoretical Bi- ogy, 53, 161–186. ology, 7, 1–52. Euler, H. A., & Weitzel, B. (1996). Discriminative Harmon-Jones, E., Vaughn-Scott, K., Mohr, S., grandparental solicitude as reproductive strategy. Sigelman, J., & Harmon-Jones, C. (2004). The effect Human Nature, 7, 39–59. of manipulated sympathy and anger on left and right Faulkner, J., & Schaller, M. (2007). Nepotistic nos- frontal cortical activity. Emotion, 4, 95–101. iness: Inclusive fitness and vigilance of kin mem- Haselton, M. G., & Nettle, D. (2006). The paranoid bers’ romantic relationships. Evolution and Hu- optimist: An integrative evolutionary model of man Behavior, 28, 430–438. cognitive biases. Personality and Social Psychol- Fehr, E., & Ga¨chter, S. (2002). Altruistic punishment ogy Review, 10, 47–66. in humans. Nature, 415, 137–140. Hauber, M. E., & Sherman, P. W. (2001). Self- Fessler, D. M. T., & Navarrete, C. D. (2004). Third- referent phenotype matching: Theoretical consid- party attitudes toward sibling incest: Evidence for erations and empirical evidence. Trends in Neuro- Westermarck’s hypothesis. Evolution and Human science, 24, 609–616. Behavior, 25, 277–294. Heider, F. (1958). The psychology of interpersonal Fiske, A. P. (1992). The four elementary forms of relations. New York: Wiley. sociality: Framework for a unified theory of social Hepper, P. G. (Ed.). (1991). Kin recognition. New relations. Psychological Review, 99, 689–723. York: Cambridge University Press. Gaertner, S. L., & Dovidio, J. F. (1977). The subtlety Hepper, P. G., & Cleland, J. (1999). Developmental of White racism, arousal, and helping behavior. aspects of kin recognition. Genetica, 104, 199–205. Journal of Personality and Social Psychology, 35, Heth, G., Todrank, J., & Johnston, R. E. (1999). 691–707. Similarity in the qualities of individual odors Gallup, G. G., Jr. (1970). Chimpanzees: Self- among kin and species in Turkish (Mesocricetus recognition. Science, 167, 86–87. brandti) and golden (Mesocricetus auratus) ham- KIN RECOGNITION 233

sters. Journal of Comparative Psychology, 113, effects of perspective taking on motivations for 321–326. helping: Still no evidence for altruism. Personality Hoffman, M. L. (1981). Is altruism a part of human and Social Psychology Bulletin, 28, 1601–1610. nature? Journal of Personality and Social Psychol- Mateo, J. M., & Johnston, R. E. (2000). Kin recog- ogy, 40, 121–137. nition and the “armpit effect”: Evidence of self- Jeon, J., & Buss, D. M. (2007). Altruism towards referent phenotype matching. Proceedings of the cousins. Proceedings of the Royal Society B, 274, Royal Society of London B, 267, 695–700. 1181–1187. McAndrew, F. T. (2002). New evolutionary perspec- Keltner, D., Haidt, J., & Shiota, M. N. (2006). Social tives on altruism: Multilevel-selection and costly- functionalism and the evolution of emotions. In M. signaling theories. Current Directions in Psycho- Schaller, J. A. Simpson, & D. T. Kenrick (Eds.), logical Science, 11, 79–82. Evolution and social psychology (pp. 115–142). Medin, D. L., Goldstone, R. L., & Gentner, D. New York: Psychology Press. (1993). Respects for similarity. Psychological Re- Komdeur, J., Richardson, D. S., & Burke, T. (2004). view, 100, 254–278. Experimental evidence that kin discrimination in Miller, P. A., & Eisenberg, N. (1988). The relation of the Seychelles warbler is based on association and empathy to aggressive and externalizing/antisocial not on genetic relatedness. Proceedings of the behavior. Psychological Bulletin, 103, 324–344. Royal Society of London B, 271, 963–969. Neyer, F. J., & Lang, F. R. (2003). Blood is thicker Korchmaros, J. D., & Kenny, D. A. (2001). Emo- than water: Kinship orientation across adulthood. tional closeness as a mediator of the effect of Journal of Personality and Social Psychology, 84, genetic relatedness on altruism. Psychological Sci- 310–321. ence, 12, 262–265. Oates, K., & Wilson, M. (2002). Nominal kinship Krebs, D. (1975). Empathy and altruism. Journal of cues facilitate altruism. Proceedings of the Royal Personality and Social Psychology, 32, 1134–1146. Society of London B, 269, 105–109. Krebs, D. (1987). The challenge of altruism in biol- Olson, J. M., Vernon, P. A., Harris, J. A., & Jang, ogy and psychology. In C. Crawford, M. Smith, & K. L. (2001). The heritability of attitudes: A study D. Krebs (Eds.), Sociobiology and psychology: of twins. Journal of Personality and Social Psy- Ideas, issues, and applications (pp. 81–118). chology, 80, 845–860. Hillsdale, NJ: Erlbaum. Park, J. H., & Schaller, M. (2005). Does attitude Krebs, D., & Janicki, M. (2004). Biological founda- similarity serve as a heuristic cue for kinship? tions of moral norms. In M. Schaller & C. S. Evidence of an implicit cognitive association. Evo- Crandall (Eds.), The psychological foundations of lution and Human Behavior, 26, culture (pp. 125–148). Mahwah, NJ: Erlbaum. 158–170. Kruger, D. J. (2001). Psychological aspects of adap- Paul, A., & Kuester, J. (2004). The impact of kinship tations for kin directed altruistic helping behaviors. on mating and reproduction. In B. Chapais & Social Behavior and Personality, 29, 323–330. C. M. Berman (Eds.), Kinship and behavior in Laham, S. M., Gonsalkorale, K., & von Hippel, W. primates (pp. 271–291). Oxford, UK: Oxford Uni- (2005). Darwinian grandparenting: Preferential in- versity Press. vestment in more certain kin. Personality and So- Penn, D. J., & Potts, W. K. (1999). The evolution of cial Psychology Bulletin, 31, 63–72. mating preferences and major histocompatibility Lieberman, D., Tooby, J., & Cosmides, L. (2003). complex genes. The American Naturalist, 153, Does morality have a biological basis? An empir- 145–164. ical test of the factors governing moral sentiments Penton-Voak, I. S., Jacobson, A., & Trivers, R. regarding incest. Proceedings of the Royal Society (2004). Population differences in attractiveness of London B, 270, 819–826. judgments of male and female faces: Comparing Lieberman, D., Tooby, J., & Cosmides, L. (2007). British and Jamaican samples. Evolution and Hu- The architecture of human kin detection. Nature, man Behavior, 25, 355–370. 445, 727–731. Perrett, D. I., Penton-Voak, I. S., Little, A. C., Tidde- Little, A. C., Penton-Voak, I. S., Burt, D. M., & man, B. P., Burt, D. M., Schmidt, N., et al. (2002). Perrett, D. I. (2003). Investigating an imprinting- Facial attractiveness judgements reﬂect learning of like phenomenon in humans: Partners and oppo- parental age characteristics. Proceedings of the Royal site-sex parents have similar hair and eye colour. Society of London B, 269, 873–880. Evolution and Human Behavior, 24, 43–51. Pizarro, D. (2000). Nothing more than feelings? The Low, B. S. (1990). Marriage systems and pathogen role of emotions in moral judgment. Journal for stress in human societies. American Zoologist, 30, the Theory of Social Behaviour, 30, 355–375. 325–340. Platek, S. M., Burch, R. L., Panyavin, I. S., Wasser- Maner, J. K., Luce, C. L., Neuberg, S. L., Cialdini, man, B. H., & Gallup, G. G., Jr. (2002). Reactions R. B., Brown, S., & Sagarin, B. J. (2002). The to children’s faces: Resemblance affects males 234 PARK, SCHALLER, AND VAN VUGT

more than females. Evolution and Human Behav- Salmon, C. A. (1998). The evocative nature of kin ior, 23, 159–166. terminology in political rhetoric. Politics and the Platek, S. M., Critton, S. R., Burch, R. L., Frederick, Life Sciences, 17, 51–57. D. A., Myers, T. E., & Gallup, G. G., Jr. (2003). Schaller, M. (2003). Ancestral environments and mo- How much paternal resemblance is enough? Sex tivated social perception: Goal-like blasts from the differences in hypothetical investment decisions evolutionary past. In S. J. Spencer, S. Fein, M. P. but not in the detection of resemblance. Evolution Zanna, & J. M. Olson (Eds.), Motivated social and Human Behavior, 24, 81–87. perception: The Ontario Symposium (pp. 215– Platek, S. M., Raines, D. M., Gallup, G. G., Jr., 231). Mahwah, NJ: Erlbaum. Mohamed, F. B., Thomson, J. W., Myers, T. E., et Schaller, M., & Cialdini, R. B. (1988). The econom- al. (2004). Reactions to children’s faces: Males are ics of empathic helping: Support for a mood man- more affected by resemblance than females are, agement motive. Journal of Experimental Social and so are their brains. Evolution and Human Psychology, 24, 163–181. Behavior, 25, 394–405. Schaller, M., Park, J. H., & Kenrick, D. T. (2007). Porter, R. H. (1987). Kin recognition: Functions and Human evolution and social cognition. In R. I. M. mediating mechanisms. In C. Crawford, M. Smith, Dunbar & L. Barrett (Eds.), Oxford handbook of & D. Krebs (Eds.), Sociobiology and psychology: evolutionary psychology (pp. 491–504). Oxford, Ideas, issues, and applications (pp. 175–203). UK: Oxford University Press. Hillsdale, NJ: Erlbaum. Sharp, S. P., McGowan, A., Wood, M. J., & Hatch- Povinelli, D. J., Gallup, G. G., Jr., Eddy, T. J., Bier- well, B. J. (2005). Learned kin recognition cues in schwale, D. T., Engstrom, M. C., Perilloux, H. K., a social bird. Nature, 434, 1127–1130. et al. (1997). Chimpanzees recognize themselves Shepher, J. (1971). Mate selection among second in mirrors. Animal Behaviour, 53, 1083–1088. generation kibbutz adolescents and adults: Incest Puppe, B. (1998). Effects of familiarity and related- avoidance and negative imprinting. Archives of ness on agonistic pair relationships in newly mixed Sexual Behavior, 1, 293–307. domestic pigs. Applied Animal Behaviour Sci- Sherman, P. W. (1977). Nepotism and the evolution ence, 58, 233–239. of alarm calls. Science, 197, 1246–1253. Reeve, H. K. (1998). Acting for the good of others: Silk, J. B. (2002). Kin selection in primate groups. Kinship and reciprocity with some new twists. In International Journal of Primatology, 23, 849–875. C. Crawford & D. L. Krebs (Eds.), Handbook of Singer, T. L., & Espelie, K. E. (1998). Nest and evolutionary psychology: Ideas, issues, and appli- nestmate recognition by a fungus-growing ant, cations (pp. 43–85). Mahwah, NJ: Erlbaum. Apterostigma collare Emery (Hymenoptera: For- Reid, J. M., Arcese, P., & Keller, L. F. (2003). micidae). Ethology, 104, 929–939. Inbreeding depresses immune response in song Solomon, G. E. A., Johnson, S. C., Zaitchik, D., & sparrows (Melospiza melodia): Direct and inter- Carey, S. (1996). Like father, like son: Young generational effects. Proceedings of the Royal So- children’s understanding of how and why off- ciety of London B, 270, 2151–2157. spring resemble their parents. Child Develop- Rendall, D. (2004). “Recognizing” kin: Mechanisms, ment, 67, 151–171. media, minds, modules, and muddles. In B. Springer, K. (1992). Children’s awareness of the Chapais & C. M. Berman (Eds.), Kinship and biological implications of kinship. Child Develop- behavior in primates (pp. 295–316). Oxford, UK: ment, 63, 950–959. Oxford University Press. Springer, K. (1996). Young children’s understanding Richerson, P. J., & Boyd, R. (2005). Not by genes of a biological basis for parent–offspring relations. alone: How culture transformed human evolution. Child Development, 67, 2841–2856. Chicago: University of Chicago Press. Steblay, N. M. (1987). Helping behavior in rural and Roberts, S. C., Gosling, L. M., Spector, T. D., Miller, urban environments: A meta-analysis. Psycholog- P., Penn, D. J., & Petrie, M. (2005). Body odor ical Bulletin, 102, 346–356. similarity in noncohabiting twins. Chemical Stephan, W. G., & Finlay, K. (1999). The role of Senses, 30, 1–6. empathy in improving intergroup relations. Jour- Rozin, P., & Fallon, A. E. (1987). A perspective on nal of Social Issues, 55, 729–743. disgust. Psychological Review, 94, 23–41. Sternberg, R. J. (1986). A triangular theory of love. Rudman, L. A., Ashmore, R. D., & Gary, M. L. Psychological Review, 93, 119–135. (2001). “Unlearning” automatic biases: The mal- Stu¨rmer, S., Snyder, M., Kropp, A., & Siem, B. leability of implicit prejudice and stereotypes. (2006). Empathy-motivated helping: The moderat- Journal of Personality and Social Psychology, 81, ing role of group membership. Personality and 856–868. Social Psychology Bulletin, 32, 943–956. KIN RECOGNITION 235

Sun, L., & Mu¨ller-Schwarze, D. (1997). Sibling rec- Wedekind, C., Seebeck, T., Bettens, F., & Paepke, ognition in the beaver: A field test for phenotype A. J. (1995). MHC-dependent mate preferences in matching. Animal Behaviour, 54, 493–502. humans. Proceedings of the Royal Society of Lon- Tesser, A. (1993). The importance of heritability in don B, 260, 245–249. psychological research: The case of attitudes. Psy- Weisfeld, G. E., Czilli, T., Phillips, K. A., Gall, J. A., chological Review, 100, 129–142. & Lichtman, C. M. (2003). Possible olfaction- Tooby, J., & Cosmides, L. (1996). Friendship and the based mechanisms in human kin recognition and banker’s paradox: Other pathways to the evolution inbreeding avoidance. Journal of Experimental of adaptation for altruism. In W. G. Runciman, J. Child Psychology, 85, 279–295. Maynard Smith, & R. I. M. Dunbar (Eds.), Evolu- Widdig, A., Nu¨rnberg, P., Krawczak, M., Streich, tion of social behaviour patterns in primates and W. J., & Bercovitch, F. B. (2001). Paternal relat- man (pp. 119–143). New York: Oxford University edness and age proximity regulate social relation- Press. ships among adult female rhesus macaques. Pro- Trivers, R. L. (1971). The evolution of reciprocal ceedings of the National Academy of Sciences altruism. Quarterly Review of Biology, 46, 35–57. (USA), 100, 13,769–13,773. Van Vugt, M., & Hart, C. M. (2004). Social identity as Winfree, R. (1999). Cuckoos, cowbirds and the per- social glue: The origins of group loyalty. Journal of sistence of brood parasitism. Trends in Ecology Personality and Social Psychology, 86, 585–598. and Evolution, 14, 338–343. Van Vugt, M., & Van Lange, P. A. M. (2006). Wolf, A. P. (1970). Childhood association and sexual attraction: A further test of the Westermarck hy- Psychological adaptations for prosocial behavior: pothesis. American Anthropologist, 72, 503–515. The altruism puzzle. In M. Schaller, J. A. Simpson, Wright, S. C., Brody, S. M., & Aron, A. (2005). & D. T. Kenrick (Eds.), Evolution and social psy- Intergroup contact: Still our best hope for improv- chology (pp. 237–261). New York: Psychology ing intergroup relations. In C. S. Crandall & M. Press. Schaller (Eds.), Social psychology of prejudice: Volk, A., & Quinsey, V. L. (2002). The influence of Historical and contemporary issues (pp. 115–142). infant facial cues on adoption preferences. Human Lawrence, KS: Lewinian Press. Nature, 13, 437–455. Yamazaki, K., Beauchamp, G. K., Curran, M., Bard, Volk, A. A., & Quinsey, V. L. (2007). Parental J., & Boyse, E. A. (2000). Parent–progeny recog- investment and resemblance: Replications, refine- nition as a function of MHC odortype identity. ments, and revisions. Evolutionary Psychology, 5, Proceedings of the National Academy of Sciences 1–14. (USA), 97, 10,500–10,502. Waldman, B. (1987). Mechanisms of kin recognition. Yamazaki, K., Beauchamp, G. K., Kupniewski, D., Journal of Theoretical Biology, 128, 159–185. Bard, J., Thomas, L., & Boyse, E. A. (1988). Waldman, B. (1988). The ecology of kin recognition. Familial imprinting determines H-2 selective mat- Annual Review of Ecology and Systematics, 19, ing preferences. Science, 240, 1331–1332. 543–571. Zahavi, A., & Zahavi, A. (1997). The handicap prin- Walster, E., Walster, G. W., & Berscheid, E. (1978). ciple: A missing piece of Darwin’s puzzle. Oxford, Equity: Theory and research. Boston: Allyn & UK: Oxford University Press. Bacon. Zajonc, R. B. (1980). Feeling and thinking: Prefer- Walter, A., & Buyske, S. (2003). The Westermarck ences need no inferences. American Psycholo- effect and early childhood co-socialization: Sex gist, 35, 151–175. differences in inbreeding avoidance. British Jour- Zajonc, R. B. (2001). Mere exposure: A gateway to nal of Developmental Psychology, 21, 353–365. the subliminal. Current Directions in Psychologi- Wedekind, C., & Fu¨ri, S. (1997). Body odour pref- cal Science, 10, 224–228. erences in men and women: Do they aim for specific MHC combinations or simply heterozygos- Received September 11, 2007 ity? Proceedings of the Royal Society of London B, Revision received December 10, 2007 264, 1471–1479. Accepted December 28, 2007 Ⅲ