Psychology and Neuroscience of Person...

Trim Size: 7in x 10in Wixted-Vol4 c13.tex V1 - 10/11/2017 7:50 A.M. Page 429�

� �

�

CHAPTER 13

Psychology and Neuroscience of PersonPerception

JEFFREY A. BROOKS AND JONATHAN B. FREEMAN

INTRODUCTION

Understanding other people amounts to oneof the most important and complex challengeshumans regularly encounter. And yet, despitethe enormously variable and noisy nature ofsocial stimuli (e.g., faces and bodies, bothin static and in dynamic configurations),humans reliably extract information aboutone another in a process that appears auto-matic and obligatory. Among the informationspontaneously extracted from viewing even astatic image of another’s face are social cate-gories, such as sex, race, and age; personalitytraits, such as trustworthiness and compe-tence; emotion; identity; and intentions.The initial perceptual operations giving riseto such impressively efficient informationprocessing are undoubtedly important andconsequential for social interaction. How-ever, most research on person perceptioncarried out by social psychologists during the20th century was not focused on people asperceptual targets. Instead of focusing on therole of external appearance and perceptualinput (visual, auditory, etc.) that contributesto perceiving and understanding others,research typically focused on the internalstates (intentions, beliefs, traits) that can beinferred about a person, the internal cogni-tive mechanisms responsible for processing

such social knowledge, and the downstreamconsequences these inferences might havefor interpersonal interaction (e.g., Brewer,1988; Hassin, Bargh, & Uleman, 2002;Kunda & Thagard, 1996; for a review, seeGilbert, 1998).

Indeed, early research typically inves-tigated person perception by presentingresearch participants with written descrip-tions of a person’s behavior or personalitycharacteristics or by directly presentingparticipants with social category labels andobserving the kinds of assumptions andinferences made about a person based solelyon category membership. Thus, within socialpsychology, early research focused on pri-marily postperceptual processes downstreamof visually based perceptions, investigatingthe effects of placing someone in a socialcategory on memory for and behaviorsenacted toward that individual. Indeed, socialcategorization in particular has a long historyin social psychology of being considered aprecursor to the more consequential act ofstereotyping (Allport, 1954; Bargh et al.,1996; Fiske & Taylor, 1991).

Although social psychologists amassedan extensive literature on the downstreamconsequences of person perception and socialcategorization, a great deal of research in cog-nitive psychology and neuroscience began

429


� �

�

430 Psychology and Neuroscience of Person Perception

to unravel the specific perceptual operationsperformed on social stimuli. such as faces.Researchers in these fields were not neces-sarily interested in the social consequencesof person perception, but understandingthe process of perceiving complex stimuli,such as faces, provided an important toolfor probing visual perception in general.This work largely focused on the perceptualmechanisms responsible for successful facerecognition. Prominent theoretical modelsemerging from this research focused on thedistinction between static facial cues anddynamic facial movements and the distinctperceptual processes these two types of cueselicit in the perceiver (Haxby, Hoffman, &Gobbini, 2000, 2002). These models werebuilt off of early observations that visualprocessing of faces diverged between rec-ognizing the identity of faces (invariant orstatic qualities of a face) and understand-ing dynamic qualities of the face, such asthose occurring during speech productionand transient displays of emotion (Bruce &Young, 1986).

Such research demonstrated the utility ofintegrating vision and face perception intoperson perception research. As research onthe mechanisms of face processing devel-oped, the field began to appreciate what aprivileged status faces and bodies have inthe perceptual system. Humans dispropor-tionately deploy attention toward faces inparticular, a tendency that emerges earlyin development. From birth, infants preferlooking at faces and facelike stimuli, andthe ability to distinguish faces by sex andovert emotional expression begins to developsoon after (Nelson, 2001), suggesting at thevery least an innate predisposition to attendto faces as motivationally (i.e., adaptively)relevant stimuli in the environment, althoughsome work suggests that these early effectsare due more to the perceptual features offaces rather than their functional relevance

(e.g., Macchi, Turati, & Simion, 2004;Simion & Di Giorgio, 2015). Seminal workin face recognition also documented the faceinversion effect (Yin, 1969), the phenomenonof extremely poor face recognition whenfaces are presented upside-down, which doesnot happen nearly as severely for nonfaceobjects. Further research demonstrated thatthis unique effect occurs because faces areprocessed configurally, with recognitionand understanding of face stimuli depend-ing on successful encoding of the spatiallayout of a face’s individual features. Inter-estingly, bodies also exhibit an inversioneffect to a similar degree as faces (Reed,Stone, Bozova, & Tanaka, 2003) and alsohave a similarly privileged role in sensoryprocessing, with considerably more attentiondeployed to bodies than to objects in theenvironment (Downing, Bray, Rogers, &Childs, 2004). These results for faces andbodies were interpreted as clear evidence thatthey are an extremely special class of stimulithat convey a wealth of relevant informationabout people, and the human perceptual sys-tem is differentially tuned to their detectionand understanding as a result.

Contemporary work by social psychol-ogists and interdisciplinary researchersworking from social neuroscience and“social vision” approaches have begun tointegrate these insights by shifting personperception research to focus more on theearly perceptual processes that underpinmore sophisticated understanding of otherpeople. Of importance to this chapter, thisincreased interest in the insights offered fromstudying visual perceptual processes wasconcurrent with the introduction of cognitiveneuroscience techniques (particularly func-tional magnetic resonance imaging [fMRI])into social psychology. The combinationof new methods and theoretical approachesallowed researchers to integrate insightsfrom vision science as well as cognitive


� �

�

Introduction 431

neuroscience into their theories of personperception. This shift has depended largelyon researchers integrating visual perceptualstimuli, such as faces and bodies, into theirstudies, allowing research on person per-ception to reflect the kinds of implicit andspontaneous perceptual inferences occurringin everyday social interaction.

Early work from this approach yieldedthe surprising insights that humans can bequite accurate in their ability to glean socialinformation from brief glimpses of faces andbodies. Much of this work used researchparadigms such as presenting participantswith silent “thin slices” of a stranger’s behav-ior, usually video clips as short as a fewseconds long, and observing the inferencesparticipants were able to make about peoplefrom these brief behavioral displays. Stun-ningly, participants were able to confidentlyrate the individuals in the video clips alonga number of dimensions, such as personal-ity (e.g., extraversion, warmth; Gangestad,Simpson, DiGeronimo, & Biek, 1992), traitanxiety (Harrigan, Harrigan, Sale, & Rosen-thal, 1996), sexual orientation (Ambady,Hallahan, & Conner, 1999), and even racialbias (Richeson & Shelton, 2005). Moreover,multiple perceivers exhibit a high agreementin their judgments of a target’s personalitytraits (termed “consensus at zero acquain-tance”; Kenny, 1991), and such reliable andconsensual inferences about other peopleoften can accurately predict real-world out-comes. For example, early work on thinslices showed that judgments of college pro-fessors’ nonverbal behavior from 30-secondsilent video clips accurately predicted theprofessors’ end-of-semester evaluations bystudents (Ambady & Rosenthal, 1993; for areview, see Weisbuch & Ambady, 2011).

Inspired by the observation that humansare rapid, consistent, and often accurate inthe impressions they draw from the externalappearance of another person, much research

has focused on the facial features that aremost informative to perceivers. This work hasbenefited from preexisting theoretical move-ments in vision research known as ecologicalor Gibsonian approaches (Zebrowitz &Montepare, 2006). The Gibsonian approachto visual perception (Gibson, 1979) stressesthe importance of directly perceptiblebottom-up cues of objects in the envi-ronment and how those cues are inherentlyinformative about the “affordances” of thoseobjects—their function, capacity to be actedon, and possible benefits or dangers to theperceiver. A basic type of affordance wouldbe the graspability of an object, informationthat is readily present in the object’s stimulusfeatures. The Gibsonian approach also allowsfor the influence of the perceiver in the formof “attunements”—individual differences inthe sensitivity to particular affordances in theenvironment. The information in consciousperception thus reflects a direct perceptionof the world that is weighted by the adaptivefunction of objects in the environment as wellas by an individual perceiver’s attunementsto which affordances are most important topick up on. An ecological approach to personperception (McArthur & Baron, 1983) thusemphasizes the importance of bottom-upfacial cues and the adaptive informationthey directly convey to the perceiver, suchas the potential personality characteristicsof an individual; whether an individualshould be approached or avoided; and relatedopportunities for behavioral response by theperceiver (e.g., caregiving, mating). A greatdeal of work from this perspective has out-lined the types of facial cues humans are mostattuned to and how the social affordancesgleaned from another person’s face (e.g.,whether someone looks young or old due tothe invariant structure of their face) can biasimpressions of their personality and behav-iors enacted toward them; in this view, evenbottom-up input can serve as a biasing factor

Trim Size: 7in x 10in Wixted-Vol4 c13.tex V1 - 10/11/2017 7:50 A.M. 32�

� �

�


in perceptions of other people (discussedlater in the subsection titled “Facial Cues”).

Despite how basic and obligatory theseprocesses seem to be, they are nonethelessmalleable and susceptible to influence froma variety of factors that reside outside of sen-sory input. Vision science is replete with workinvestigating such “top-down” influences onperception, with a particular focus on the sys-tematic biases and errors that can be causedfrom motivations or expectations harboredin the perceiver. This approach to vision canbe traced back to early observations in psy-chology (e.g., Helmholtz, 1867) that sensoryinput often is impoverished, lacking, or briefbut nonetheless must be rapidly understoodand acted on by the perceiver. Thus, percep-tion consists of the concurrent and interactiveprocessing of ongoing sensory input withthe memories, thoughts, emotions, and con-ceptual knowledge of the observer, togetherleading to a probabilistic interpretation of agiven stimulus.

An influential movement in cognitivepsychology beginning in the 1950s (calledthe New Look; Erdelyi, 1974) was the firstresearch program to investigate these per-ceiver effects experimentally. For instance,researchers found that children of lowersocioeconomic status tend to overestimatethe size of coins in their hands (Bruner &Goodman, 1947) and individuals are moreperceptually sensitive to faces when theyare highly motivated to affiliate with others(Atkinson & Walker, 1956). Contemporaryresearch also has examined the influenceof transient motivational states on visualperception, showing that participants aremore likely to interpret an ambiguous visualstimulus (such as two adjacent lines thatcould either be perceived as the letter Bor the number 13) in a certain way whenthey anticipate a positive outcome from thatinterpretation (Balcetis & Dunning, 2006),and ambiguous blends of two colors are more

likely to be interpreted as one color insteadof the other when that color is associatedwith financial gain for the participant (Voss,Rothermund, & Brandtstadter, 2008).

Indeed, such top-down influences onvisual perception tend to have a pronouncedimpact when the target of perception isambiguous (Bruner & Goodman, 1947;Pauker, Rule, & Ambady, 2010; Trope,1986). Within the realm of person percep-tion, this finding has important consequencesfor the social categorization of individualsof ambiguous social category membership(e.g., biracial individuals) or members of aperceptually ambiguous social category (e.g.,gay individuals). Indeed, if top-down factorsexert more of an influence when sensoryinput is ambiguous, then their importance toperson perception cannot be understated, associal stimuli such as faces and bodies typ-ically provide sensory input that is variableand ambiguous.

Recent research integrating perspectivesfrom vision science, cognitive neuroscience,and social psychology has permitted inves-tigation into the basic processes throughwhich social information is extracted fromsocial perceptual cues, such as faces, bodies,and voices. In this chapter, we describe thewealth of bottom-up cues and top-downfactors that contribute to the process ofperceiving another person. For descriptivepurposes, we discuss bottom-up cues andtop-down influences in separate sections, butwe emphasize that there is a rich interplaybetween bottom-up and top-down processesduring person perception, and the two maybe difficult to tease apart in practice. Afterdiscussing the particular bottom-up percep-tual cues and top-down factors that interact toproduce stable perceptions and impressionsof other people, we turn to a discussion of theneural mechanisms that have been observedto contribute to the person perception processand computational models that have been


� �

�

Bottom-Up Perceptual Cues and the Target of Social Perception 433

proposed to account for the complex inter-play of environmental cues and top-downconstraints in person perception. We end thechapter with a discussion of the downstreamconsequences in social interaction that canresult from subtle components of these earlyperceptual processes.

BOTTOM-UP PERCEPTUAL CUESAND THE TARGET OF SOCIALPERCEPTION

Facial Cues

An extensive body of research has docu-mented the kinds of social information thatcan be gleaned from the human face and whatspecific facial cues and characteristics drivethese perceptions. Indeed, person perceptionmay be prone to systematic biases due tothe underlying characteristics of face per-ception. Facial cues convey variant qualitiesof a person, such as their current emotionalstate and health, but they also can conveyrelatively invariant qualities of a person, suchas sex, race, age, and personality charac-teristics. Social categories, particularly the“Big Three” (sex, race, and age; Brewer,1988), have received a great deal of attentionin the person perception literature becausethey can be gleaned reliably and efficientlyfrom specific salient biological cues, andconsequently they come to dominate ourperceptions of other people.

Following advances in statistical facemodeling, researchers became capable ofexperimentally manipulating the features offace stimuli in order to assess the differentialrole of certain cues in determining an indi-vidual’s social category membership. Earlywork considered the differential impact ofspecific facial cues, such as the shape and sizeof the jaw, brows, and chin, on determiningsocial category memberships in the domainof gender (Brown & Parrett, 1993) and age

(Berry & McArthur, 1986). Specific facialcues also have been proposed to give rise tocategorization along perceptually ambiguousdimensions, such as sexual orientation. Sem-inal work on “thin slices” of behavior foundthat sexual orientation could be classifiedreliably with above-chance accuracy frombrief presentations of dynamic nonverbalcues (Ambady et al., 1999), and succeedingstudies found that extremely brief exposureto static faces (e.g., 50 ms) is sufficient foraccurate categorization of sexual orientation(Rule & Ambady, 2008a). These researchersadditionally found that specific features oftarget faces—hair, eyes, and mouth—wereable to yield, as a whole, relatively accuratecategorizations of sexual orientation evenwhen these cues were presented indepen-dently of any other facial features (Rule,Ambady, Adams, & Macrae, 2008). Inter-estingly, participants underestimated theirown accuracy on these tasks, suggestingan implicit fluency for the diagnosticity ofcertain facial cues on seemingly ambiguousdimensions, such as sexual orientation.

Although less is known about the relativeimportance of specific facial cues to differ-ent social category dimensions, researchersagree that several broader qualities of theface contribute to social categorization.These patterns of features include the shapeof the face (encompassing broad patterns ofstructural variation) as well as face pigmen-tation, alternately referred to as color andtexture in the literature. Early studies focusedon the role of face shape, and researcherstheorized a primacy for face shape in driv-ing evaluations and categorizations of faces(Biederman, 1987). Later work showed animportant additional role for pigmentationcues in determining social categories such asage, race, and gender (Price & Humphreys,1989; Hill, Bruce, & Akamatsu, 1995), butmany researchers still emphasized the impor-tance of shape, assuming that pigmentation


� �

�


cues (to the extent that they are informativeat all) are integrated later in the perceptualprocess. However, recent work has shownthat both patterns of shape and pigmentationcues are integrated into the social catego-rization process in a temporally dynamicfashion, with parallel processing of shapeand pigmentation giving rise to coherentperceptions (although pigmentation actuallyappears to have primacy in the perception ofgender, showing an influence on the percep-tual process at earlier time points; Freeman &Ambady, 2011b).

Researchers also have considered theimpact of a face’s width-to-height ratio(fWHR), a useful measure of face shapeand structure with particular importance forthe perception of male faces. The fWHR isdefined as the distance between the cheek-bones divided by the distance between theupper lip and mid-brow, and the magnitudeof the fWHR is driven by pubertal testos-terone in men (Lefevre, Lewis, Perrett, &Penke, 2013). As predicted by an ecologicalapproach to person perception (McArthur &Baron, 1983), perceivers seem to use thefWHR as a relatively accurate index ofbehavior. Larger fWHR is correlated withdeceptive (Haselhuhn & Wong, 2012) andaggressive behaviors (Carré & McCormick,2008), and perceivers readily evaluate indi-viduals with high fWHR as less friendly(Hehman, Carpinella, Johnson, Leitner, &Freeman, 2013), less trustworthy (Stirrat &Perrett, 2010), and more aggressive (Carré,McCormick, & Mondloch, 2009).

With its strong suggestion of an indi-vidual’s traits and behaviors, the fWHRis a prominent candidate for the type offacial cues that drive overgeneralizationeffects (Zebrowitz, Fellous, Mignault, &Andreoletti, 2003), prominent biases intrait attribution occurring when cues on theface tied to specific social affordances biasimpressions of another person. For example,

older adults (who display lower perceivedfWHR due to changes in the skin) are per-ceived as less aggressive, less physicallycapable, less socially competent, less sociallydominant, and friendlier as a function ofdecreases in fWHR (Hehman, Leitner, &Freeman, 2014b). On the other end of theage spectrum, there is a well-documented,age-related overgeneralization effect suchthat adults with babyish features are in turnperceived to be more childlike (e.g., weak,submissive, vulnerable, submissive, and hon-est; Montepare & Zebrowitz, 1998). Centralto researchers’ interpretation of these effectsis the Gibsonian assumption that perceiversare specifically attuned to social affordancesthat require rapid orienting and appropri-ate behavioral responses. In this view, ageovergeneralization effects occur becausehumans are attuned to rapidly detect and acton age-related cues in the environment toprovide necessary care to vulnerable infants.Similar overgeneralization effects have beenobserved in trait impressions from emo-tional expressions, such that individuals withpermanent resemblance to certain canoni-cal emotional expressions are perceived tohave invariant personalities related to thosetraits. For example, individuals with facesresembling neutral or angry expressionsusually are judged to be low in affiliativetraits, while individuals with faces resem-bling happy expressions usually are judgedto be high in affiliative traits (Montepare &Dobish, 2003).

The relationship between specific facialcues and social categories and trait attri-butions is further complicated by the factthat specific facial cues may signal morethan one social category at a time, at timessubjecting social categorization to systematicbiases. For example, facial cues signalingsex-category membership overlap with cuessignaling emotional state, resulting in consis-tent biases of impressions of trait dominance


� �

�


and affiliation in men and women (Hess,Adams, & Kleck, 2004, 2005). Addition-ally, race and sex-category membershipintersect such that the African Americanrace category shares overlapping phenotypiccues with the male sex category (Johnson,Freeman, & Pauker, 2012), biasing stereo-typic expectations for individuals who do notsatisfy this expected congruence (e.g., Blackwomen; Johnson et al., 2012). However, theexact degree to which these effects are dueto overlapping facial cues is under debate(Johnson et al., 2012). Another strong influ-ence on intersectionality effects is overlapin stereotype content between categorieson orthogonal dimensions (e.g., the Blackrace category and male sex category sharingstereotypes for aggression and athleticism).We discuss such influences on category inter-sectionality effects further in the subsectiontitled “Stereotypes.”

As evident in overgeneralization effects,humans infer a wealth of trait informationfrom features of the face. Although theseinferences occur impressively rapidly andwith surprising consensus in the popula-tion, they also may be particularly proneto error. A prominent model of personalitycharacteristics is the Big Five factor model(Goldberg, 1990), which describes humanpersonality characteristics in terms of thefive factors of openness, conscientiousness,extraversion, agreeableness, and neuroti-cism. Early research examining the abilityof perceivers to glean these traits from facesshowed that they were rapid, consistent, andoften accurate (e.g., Watson, 1989).

However, research on the extraction ofspecific personality characteristics (like theBig Five) suffers from the fact that severalof these judgments are intercorrelated anddifficult to differentiate in terms of theirassociated facial cues (Sutherland et al.,2015). Thus, a large portion of the work onpersonality judgments of faces has focused

on broader impressions, such as trustworthi-ness, a personality trait which some researchposits as a broader personality dimensionthat accounts for many of the intercorrelatedjudgments of more granular personalitycharacteristics, such as those in the Big Five(Oosterhof & Todorov, 2008). Researchersstudying impression formation and trait attri-bution from faces have offered many differentpossibilities for a parsimonious encoding oftrait information along universal dimensions,such as warmth and competence (Fiske,Cuddy, & Glick, 2007), trustworthiness anddominance (Oosterhof & Todorov, 2008),and valence and dominance (Todorov, 2011).

These theoretical approaches are united intheir attempt to account for the remarkablespeed and consensus with which such judg-ments are made. Such theoretical accountsform a Gibsonian approach that assumes thatthe fundamental information extracted fromfaces is that which is adaptively relevantto the perceiver: In the case of warmth andcompetence, for example, the “warmth”dimension reflects whether a novel individualis antagonistic (approachability), and the“competence” dimension reflects whether anindividual can cause harm to the perceiver(Fiske et al., 2007). Several insights have beenmade regarding the specific facial featuresthat give rise to these social judgments. Moti-vated by work on overgeneralization effects,researchers in this dimensional approachhave observed that an approachability dimen-sion like “trustworthiness” is most closelyrelated to a face’s general resemblance toan emotional facial expression, while harmcapability dimensions such as “competence”or “dominance” are most closely relatedto cues signifying strength and maturity(Oosterhof & Todorov, 2008). Additionally,recent work has shown that trustworthinessjudgments in particular are driven by a face’saverageness, such that average faces (interms of proximity to the physical average


� �

�


of faces in the population) consistently arerated as more trustworthy (Sofer, Dotsch,Wigboldus, & Todorov, 2015).

Averageness is also a potent contribu-tor to perceptions of facial attractiveness(Langlois & Roggman, 1990). Early workused face-morphing techniques to show thatratings of attractiveness for a compositeface are consistently higher than ratings ofattractiveness for any of the individual facesused to make the composite (Langlois &Roggman, 1990). Some researchers haveproposed that this tendency is due to innatedrives to pursue partners with a high degreeof genetic diversity (Thornhill & Gangestad,1993). Indeed, like personality traits, attrac-tiveness judgments for novel faces show ahigh level of consensus across participants,even cross-culturally (Langlois et al., 2000).In addition to facial averageness, facial sym-metry also contributes to judgments of aface’s attractiveness (Grammer & Thornhill,1994). However, other research shows thatthe overall symmetry of a face neverthelesscorrelates with attractiveness judgmentswhen only half of a face is presented to par-ticipants (Scheib, Gangestad, & Thornhill,1999), suggesting that symmetry may covarywith other featural aspects of the face thatconfer attractiveness. Researchers also haveexamined the role of sexually dimorphicfacial cues, although the emerging pictureis complicated: Highly feminine cues con-sistently increase ratings of attractivenessfor female faces (M. R. Cunningham, 1986),but male faces also are rated more attrac-tive when some feminine cues are presenton the face (Little, Burt, Penton-Voak, &Parrett, 2001).

Throughout this section, we have doc-umented facial features that give rise toremarkably consensual judgments amongperceivers in the population, includinginferences about personality traits. Clearlythis consistency in judgments shows that

perceivers draw on some perceptual heuristicthat they find to be reliable, but the issue ofwhether face-based trait inferences truly canbe accurate is complicated. One componentof the problem is the difficulty of definingperceiver accuracy for personality traits inthe first place: whether to assess the corre-spondence between perceiver inferences andthe target’s self-reported personality traits,consensus of the target’s peers, or real-worldbehavioral outcomes. For example, someresearch has assessed the ability of perceiversto accurately predict real-world outcomes,such as a corporate firm’s success based onthe facial appearance of chief executives(Re & Rule, 2016), but it remains unclear towhat degree findings like this reflect accu-racy per se. The complexities of this issueare covered in much greater detail elsewhere(e.g., Alaei & Rule, 2016).

In general, for accurate trait inferences toarise from static facial photos in any consis-tent fashion across contexts, these inferenceswould have to rely on cues that are relativelyfixed in the target. Indeed, some work hasshown that the most accurate and consistentimpressions are drawn from skeletal cues,such as the fWHR (discussed previously),which drive impressions of dominance andphysical ability. However, perception of othertraits (such as trustworthiness) relies on astatic face’s resemblance to a more dynamicfacial expression (such as an emotionalexpression), and thus these inferences arerelatively less stable across multiple imagesof a target individual, leading to less oppor-tunity for these trait inferences to be accurate(Hehman, Flake, & Freeman, 2015). Thatsaid, there are baseline, resting levels of suchresemblances, and these potentially couldbe able to produce accurate judgments in acontext-free fashion.

Relatedly, dynamic facial cues can conveyconsiderably more information than staticimages, but research on person perception,


� �

�


social categorization, and face processingmore broadly has focused primarily on staticimages of faces. There is a great deal ofwork on dynamic facial cues in the contextof emotion perception, for which dynamicfacial cues appear to convey more informa-tion about emotional state than static facialcues (Ambadar, Schooler, & Cohn, 2005). Inaddition, there is also a large body of work oneye gaze, a dynamic facial cue that can signalsocial motives and intent more generally(Frischen, Bayliss, & Tipper, 2007). We omita deeper discussion of dynamic facial cuessince they have not been studied nearly tothe same extent in the context of extractingsocial information, such as category mem-bership and personality traits, from faces.However, as discussed earlier in the chapter,there is a long history of research lookingat dynamic bodily displays and nonverbalbehavior (such as thin slices), and this bodyof work finds that even brief displays ofnonverbal bodily behavior are extremelyinformative to perceivers and frequently giverise to consistent impressions that can predictreal-world outcomes.

Bodily Cues

In everyday interaction, human faces are, ofcourse, rarely perceived in isolation from ahuman body. As such, the body can providea powerful source of visual context for faceperception as well as an ample source ofsocial knowledge about an individual in itsown right. In the case of emotion perception,cues from the body even appear to dominateinput from the face, and when body posture isincongruent with facial expressions, the ulti-mate emotion categorization often can beconsistent with the body posture rather thanthe facial cues (Van den Stock, Righart, &de Gelder, 2007). As with faces, the per-ception of human bodies in both static anddynamic configurations is greatly privileged

by the perceptual system, with perceptualattunement to bodies and “biological motion”perception (i.e., the perception of bodilymovement) subject to similar privilegedconfigural visual processes as faces (Reedet al., 2003), although these processes seemto emerge more slowly and later in devel-opment than those for faces (Freire, Lewis,Maurer, & Blake, 2006).

The perception of static bodily cues hasbeen studied mostly in the context of emotionperception. Bodily cues are strongly sugges-tive of the emotional state of an individualand provide such a potent source of visualinformation about emotional states that theycan disambiguate facial displays of emotionand also influence or override initial percep-tions of facial emotion (Van den Stock et al.,2007; for a review, see de Gelder, 2005).However, the majority of research on theperception of the body more generally hasexamined biological motion, which primarilyrefers to naturalistic human movement, suchas walking. The study of bodily movementwas propelled by the psychophysicist GunnarJohansson, who developed a novel techniquefor isolating displays of human movementfrom their visual context (Johansson, 1973).The stimuli created using this techniquegenerally are referred to as point-light dis-plays, and researchers have carried out agreat deal of work examining the surpris-ing amount of information that perceiversreadily extract from these perceptually min-imal stimuli. To create point-light displays,researchers attach reflective or infrared mark-ers to an individual’s major joints and headand record videos of the person in displaysof naturalistic movement. When participantssee these videos, only the points of lightare visible.

Despite the highly impoverished andcontext-free nature of these stimuli, per-ceivers display a readiness and sensitivity todetect the information present in point-light


� �

�


displays. Early work showed that partici-pants can readily and accurately identify thespecific actions performed by individuals inpoint-light displays, such as swinging a ham-mer or knocking on a door (Johansson, 1973,1975). In the case of walking, point-lightdisplays of gait contain cues to the identityof the walker, and studies have shown thatparticipants can reliably identify themselvesand known others in point-light displays(Cutting & Kozlowski, 1977; Richardson &Johnston, 2005). Early work also showedthat sex-category membership is categorizedaccurately in biological motion paradigms(Kozlowski & Cutting, 1977). Indeed, laterwork established that biological motion is areliable and determinant cue to sex-categorymembership because of specific variationsin male and female bodies that drive stablebiomechanical variants in gait, such as the“center of moment” (Cutting, Proffitt, &Kozlowski, 1978). Fascinating work alsohas examined the interplay between bodyshape and bodily motion in driving per-ceptions of sexual orientation (Johnson,Gill, Reichman, & Tassinary, 2007). Theseresearchers found that gender-atypical com-binations of body shape and gait (e.g., amale body exhibiting the typical “sway”gait pattern of female bodies or a femalebody exhibiting the typical “swagger” gaitpattern of male bodies) were consistentlymore likely to be categorized as homosexual.Other social categories have been studiedfar less in the context of biological motion,but some research suggests that age canbe determined reliably from point-lightdisplays (Montepare & Zebrowitz, 1993)as well.

Emerging work also has examinedperceivers’ ability to make attractivenessjudgments based on point-light displays.Interestingly, these studies have largelyconverged with the literature on facial attrac-tiveness, showing that cues to biological

fitness (e.g., symmetry and “internal consis-tency”; Kluver, Hecht, & Troje, 2016) as wellas the presence of sexually dimorphic cues(Troje, 2003) both contribute to judgmentsof attractiveness from point-light displays. Inaddition, studies have found that perceiversare able to extract variant psychological statesfrom point-light displays, such as discreteemotions (Atkinson, Dittrich, Gemmell, &Young, 2004) and intent (such as whether amovement was natural or purposely exagger-ated; Runeson & Frykholm, 1983). The studyof biological motion provides an impressiveexample of the inherently social nature ofperception, largely because of the frequentuse of point-light displays, which are ableto isolate the information extracted frommotion itself regardless of the visual or socialcontext. Overall, this work demonstrateshow the visual system is surprisingly attunedto the social and informational content ofspecific cues from both faces and bodies.

Vocal Cues

The voice is an abundant (albeit experi-mentally underappreciated) source of socialinformation as well as more basic informa-tion about the physical characteristics of thespeaker. Voices are enormously informativein isolation (as when one speaks to a strangeron the phone for the first time), but theyalso can serve as multimodal context forthe perception of someone’s face (as whenone finally meets someone previously onlyspoken to on the phone). Research on socialcategorizations and stereotyping largelyhas ignored vocal contributions, perhapsbecause of an implicit assumption that thevoice is not as salient a cue as the face.However, this assumption may be inaccuratefor some aspects of social perception, assome researchers suggest that vocal cuesmay be even more informative than facialcues in the case of emotion perception, due


� �

�


to their variability and ability to conveysubtle distinctions in emotional state (e.g., aloud approach-oriented anger versus a quietbrooding anger; Scherer, 2003).

An extensive body of research has stud-ied the basic information that is rapidlyinferred from vocal cues in isolation, includ-ing physical attributes, such as height andweight (Van Dommelen, 1993), body sizeand shape (Evans, Neave, & Wakelin, 2006),age (Hughes & Rhodes, 2010), and affectivestate (Bestelmeyer, Rouger, DeBruine, &Belin, 2010), all of which are gleaned fromthe voice in a generally accurate manner.Moreover, this information is availableto perceivers even when vocal cues arepresented for extremely brief durations(Latinus & Belin, 2012). A recent study hasshown that more socially consequential per-sonality information also is rapidly extractedfrom vocal cues (McAleer, Todorov, & Belin,2014). This study repeatedly presented par-ticipants with the word “hello” spoken bydifferent targets and found that participantsrapidly inferred traits such as trustworthiness,aggressiveness, competence, confidence, andattractiveness from these utterances. Highlyconsistent impressions of these traits werereached with exposure to vocal clips thatwere on average less than 400 ms in length,in keeping with the oft-reported consen-sus in personality judgments of strangersobserved in the face perception literature.The researchers found that specific aspectsof the acoustic input reliably covaried withperceived personality traits in a mannerthat depended on the gender of the speaker.For example, perceived dominance in malevoices seems to depend on decreases inpitch, while for females perceived domi-nance increases with increases in the pitchof the voice. However, there were somecommonalities, with the acoustic variableharmonic-to-noise ratio (indicating rough-ness) contributing to perceptions of valence

in both male and female speakers (McAleeret al., 2014).

Vocal cues also carry social categoryinformation in a way that appears to dependon specific characteristics of the acousticinput. Gender is rapidly and accuratelyperceived from vocal cues, which are partic-ularly distinct between males and femalesbecause of dimorphism in the body (Fitch &Giedd, 1999). Perceivers are very sensitiveto diagnostic gender cues in the voice andcan discriminate subtle differences in thefemininity versus masculinity of a voicewithin gender categories (e.g., feminine ver-sus masculine male voices), in a process thatautomatically activates relevant stereotypes(e.g., males with feminine vocal cues presentin the voice are expected to be feminine andpossess attributes stereotypically linked tofemininity, such as sensitivity and kindness;Ko, Judd, & Blair, 2006). Research also hasfound that race is categorized accuratelyfrom isolated vocal cues (Walton & Orlikoff,1994). African American voices tend to havelarger frequency perturbation (varying pitchin the voice) and amplitude perturbation(varying loudness in the voice) as well as sig-nificantly lower harmonic-to-noise ratio thanWhite voices. Participants were more suc-cessful at discriminating African Americanversus Caucasian speakers when these char-acteristics of the auditory signal were mostdistinct, suggesting a specific sensitivityto certain vocal cues in race perception(Walton & Orlikoff, 1994).

A growing body of research also hasshown how vocal cues interact with facialcues during social categorization of faces.Notable effects have been observed inemotion perception, in which sad facialexpressions are mistakenly perceived ashappy when they are accompanied bya happy voice, even when participants areinstructed to disregard the voice (de Gelder &Vroomen, 2000). The voice is also a salient


� �

�


multimodal cue in the categorization of aface’s gender, and studies have shown thatgender-congruent voices facilitate accu-rate detection of gender on a face (Smith,Grabowecky, & Suzuki, 2007) and incon-gruent voices can disrupt processing offace gender (Masuda, Tsujii, & Watanabe,2005; for a review, see Campanella &Belin, 2007). Work in gender categorizationalso has shown that vocal cues can biasprocessing of gender-atypical faces (e.g.,feminine male faces and masculine femalefaces). In one study (Freeman & Ambady,2011c), researchers employed a computermouse-tracking paradigm, measuring thetrajectory of computer mouse movementsas participants reached to click on a “male”or “female” category response in a gen-der categorization task. The stimuli wereslightly gender-atypical male and femalefaces accompanied by voices that were eithergender typical (e.g., a masculine male voice)or gender atypical (e.g., a feminine malevoice). The researchers found that whenfaces were accompanied by a sex-atypicalvoice, participants’ mouse movements con-tinuously deviated toward the oppositecategory response (e.g., participants werecontinuously attracted to the “female” cate-gory response when categorizing a male faceaccompanied by a feminine male voice). Thiswork suggests an important role for the voiceas a continual source of information duringperson perception, interacting with and evenbiasing an evolving visual interpretation asparticipants develop a stable categorization ofa face’s gender (Freeman & Ambady, 2011c).

Vocal cues comprise an important sourceof social and nonsocial information about anindividual. Humans are reliably attuned tosubtle distinctions in the acoustic propertiesof vocal cues that signal such information,ranging from perceptions of the body sizeof the speaker to inferences about that indi-vidual’s trustworthiness and competence.

As such, vocal cues signal important infor-mation about perceptual targets in isolationand provide a source of multimodal contextthat can enhance or constrain the perceptionof faces. In such paradigms, participants inte-grate information present in the voice intotheir categorical judgments of face stimuli,even when asked to disregard the vocal input.These findings together indicate an importantrole for vocal cues in developing stable per-ceptions of other people, both on their ownand embedded in the variety of bottom-upsensory cues discussed in this section. Wenow turn to a discussion of top-down factors,which can impact, guide, and bias the personperception process.

TOP-DOWN FACTORS IN THEENVIRONMENT AND HARBOREDIN THE PERCEIVER

Because of the abundance of informationpresent in sensory input, the perceptualsystem is necessarily strategic. In additionto processing cues from the environment,person perception also must utilize thevisual and social context in the environmentand preexisting perceptual heuristics in theobserver to make sense of ongoing sensoryinput. These additional top-down factors canbe social or nonsocial in nature and can takethe form of extraneous perceptual input (i.e.,external cues in the environment that guidethe perception of a target stimulus, suchas the surrounding context influencing theperception of a face) or inputs from the per-ceiver (i.e., motivations and expectations thatstructure and potentially bias the processingof novel stimuli). Many of these influenceson perception have not always been treatedas such: Stereotypes, for example, were longconsidered to be triggered after the per-ception of an associated stimulus (Allport,1954), while contemporary approaches


� �

�

Top-Down Factors in the Environment and Harbored in the Perceiver 441

appreciate the influence that stereotypes canhave on a visual percept before it has fullystabilized and reached conscious awareness(Freeman & Ambady, 2011a). Indeed, ahallmark of top-down effects is that theyco-occur with and even constrain basicvisual processing, despite often consisting ofhigh-level information, such as social factorsthat intuitively seem distinct from perception.

Visual Context

In the realm of person perception, “context”encompasses a broad range of visual aspectsof the environment that provide a sourceof expectations and predictions about thesocial targets likely to be perceived in thatenvironment. A basic and intuitive exampleof context in person perception is the visualscene in which a person is encountered,which is certainly relevant for determiningtheir identity. For example, the process ofdeciding whether someone who looks likeyour boss is actually your boss likely willdiffer depending on whether the personis encountered in an office setting or in anightclub. However, in person perception,“context” is multifaceted. The immediatevisual context inherent to an individual (e.g.,their clothes or the positioning of their body)can provide visual context for the perceptionof their face. A multimodal cue, such asa person’s voice (discussed previously inthe “Vocal Cues” subsection), can impactand even bias perception of their face, andone social category that a person belongsto can serve as context for the perceptionof that person’s other group and categorymemberships due to overlapping physicalcues (discussed earlier in the “Facial Cues”subsection) or stereotypes (discussed laterin the “Stereotypes” subsection). Here werestrict our discussion to the influence ofthe immediate visual context and scene onperception of an individual’s face.

Even cues inherent to the individual (e.g.,hair and clothing) can be understood as asource of visual context, supplying a sourceof expectation and prediction for the percep-tion of someone’s face and identity. Hairstylein particular has been studied in the context ofsocial categorization, and studies have shownthat racially ambiguous faces are more likelyto be categorized as Black when they havea stereotypically Black hairstyle. Followingthis categorization, the faces subsequentlyare perceived to have more Afrocentric cueson the face (MacLin & Malpass, 2001).Clothing also can bias race perception byexerting a contextual cue to the social statusof an individual, eliciting predictions aboutthe person’s race. In one study, researcherspresented participants with faces morphedalong a Black–White continuum, each ofwhich was presented with low-status attire(e.g., a janitor uniform) or high-status attire(e.g., a business suit). In a mouse-trackingparadigm, participants categorized the facesas either White or Black while their com-puter mouse trajectories were recorded. Thestudy found that low-status attire biasedperceptions toward the Black category whilehigh-status attire biased perceptions towardthe White category. When race and statuswere stereotypically incongruent (e.g., aWhite face with low-status attire or a Blackface with high-status attire), participants’mouse movements showed a continuousattraction to the opposite category, indicat-ing that the social status associated withclothing exerted a top-down influence on thevisual categorization of race. This effect wasgreater for more racially ambiguous faces(Freeman, Penner, Saperstein, Scheutz, &Ambady, 2011).

The visual scene also provides a usefulsource of prediction about the types of indi-viduals likely to be encountered in a givenenvironment, which in turn can bias theperception of individuals in certain contexts.


� �

�


For example, salient cultural contexts providea predictive framework for the potential racesand ethnicities likely to be encountered inthat context (such as a Shinto shrine elicitingthe prediction that Japanese individuals willbe encountered nearby). Behavioral workhas shown that race categorization is facili-tated by race-congruent contexts, such thatAsian faces are categorized more rapidlyand accurately in Asian scenic contexts, andthat race-incongruent contexts interfere withrace categorization (Freeman, Ma, Han, &Ambady, 2013). The scenic context alsomay play a particularly important role inemotion perception. Given the ambiguousand highly variable nature of emotional facialexpressions, perceiving discrete emotionson another’s face seems particularly relianton contextual and multimodal cues (Barrett,Mesquita, & Gendron, 2011), as discussedearlier in the subsections on bodily (e.g.,de Gelder, 2005) and voice (e.g., Scherer,2003) cues. There is a great deal of behav-ioral evidence to suggest that identical facialexpressions of emotion are perceived differ-ently depending on the visual scene in whichthey are encountered (e.g., a neutral context,such as standing in front of a house, or a fear-ful context, such as a car crash; Righart &de Gelder, 2008). Similar effects occur whenparticipants are just given prior knowledgeabout the social context emotional facialexpressions were originally displayed in(Carroll & Russell, 1996). Social informationimmediately present in a scene also caninfluence emotion perception. For example,emotion perception can be influenced bythe facial expressions of other individualsin a visual scene (Masuda et al., 2008), aneffect that is impacted by perceivers’ culturaldifferences in sensitivity to context (Ito,Masuda, & Hioki, 2012). These results arewidely consistent with insights in the visionscience community about the inherentlypredictive nature of perception, rendering

these processes particularly prone to expecta-tions guided by the environment (Bar, 2004;Summerfield & Egner, 2009).

Prior Knowledge and Familiarity

Most of the research considered thus far hasdealt with the knowledge that can be inferredfrom a complete stranger based on their faceand body. However, a great deal of earlywork on face perception focused instead onthe ability to recognize the identity of pre-viously encountered faces. Although takenfor granted on a daily basis, the ability torapidly recognize another person’s identity isan incredible feat, given the large variety ofsocial contexts and visual perspectives a per-son can be encountered in, from dim lighting,to a new haircut, to age-related changes in theface. Indeed, research has demonstrated thatchanges in viewpoint, lighting, and distancerarely cause difficulty in the recognition offamiliar faces, but such variables greatlyimpede recognition of recently learned faces(Hancock, Bruce, & Burton, 2000). Thevisual features of a face can serve as a potenta priori source of expectations about a per-son, as we have observed, but familiarity withand prior knowledge about a person exertadditional forces on person perception in theform of affective responding and spontaneousretrieval of semantic knowledge related tothat person (Gobbini & Haxby, 2007). Classicearly work showed that familiarity with anindividual’s identity sensitizes perception tocategorically perceive that identity (Beale &Keil, 1995). Researchers morphed the facesof famous politicians (e.g., John F. Kennedy,Bill Clinton) to create a continuum of facestimuli from one individual to the other.When prompted to categorize the faces byidentity, participants perceived face identitycategorically, with an abrupt shift from oneidentity to the other despite the objectivelygradient nature of the stimuli. Importantly,


� �

�


participants’ self-reported familiarity withthe identities in question predicted thedegree to which they perceived the identitiescategorically.

Prior knowledge about an individual alsocan influence person perception, even ifthere is no firsthand familiarity with thetarget. In some cases, this influence seemsto arise from the spontaneous extraction ofperson-specific semantic knowledge. Forexample, when participants are asked tojudge the personality traits of faces, they areinfluenced by information previously pairedwith that face, even when that information(and its face pairing) is not explicitly recalled(Uleman, Blader, & Todorov, 2005). Suchprior knowledge has an even more powerfulbiasing effect on trait attribution when theknowledge is affectively salient (e.g., knowl-edge that the target has previously engagedin disgusting behaviors; Todorov, Gobbini,Evans, & Haxby, 2007). However, the impactof prior knowledge also may occur at ear-lier processing stages, influencing visualprocessing via top-down attentional routes.Recent research has shown that the affectiveand social content about prior knowledgelearned about a person can influence visualperception of their face, even at sublimi-nal levels of processing (Anderson, Siegel,Bliss-Moreau, & Barrett, 2011). In this study,participants rehearsed face stimuli that werepaired with negative, positive, or neutralinformation and subsequently completed abinocular rivalry task. In binocular rivalrytasks, perceptually dissimilar images arepresented to the left and right visual fields ofa participant, and one of the two images even-tually dominates conscious visual perception(Blake, 2001). Although this dominanceis not always long lasting, the duration ofperceptual dominance and the particularstimulus that is most likely to dominateperception often are interpreted as evidencefor a top-down attentional bias toward that

particular percept. The researchers in thisstudy found that faces previously pairedwith negative social affective informationwere more likely to dominate in binocularrivalry, and dominate for a longer period oftime. Fascinatingly, this work suggests thatprior knowledge about a person can becomeactivated during preconscious processing ofa target, in turn modulating attention andperception to enhance processing of theface. Thus, the perceptual system appearsparticularly sensitive to recognizing knownothers in the environment and biasing atten-tion toward individuals who have previouslybeen associated with socially and affectivelysalient knowledge. These characteristics offace perception in particular are consistentwith the idea that the perceptual system ismost attuned to information in the environ-ment that is motivationally relevant to theperceiver.

Stereotypes

Stereotypes are comprised of conceptualknowledge about social groups—the types oftraits, behaviors, and physical features mem-bers of a specific social group are expectedto display (Allport, 1954). Stereotypes are acomplex example of top-down influences onperson perception since in many cases theyare linked directly to specific bottom-up cues.Although all influences of stereotypes on per-ception depend in part to the expectations andsocial knowledge of the perceiver, they alsorely on specific features of the environment(i.e., facial cues) that have become associatedwith such expectations (e.g., Afrocentricfacial cues trigger stereotypes associatedwith the “Black” race category, includingnegative stereotypes, such as hostility, andmore positive stereotypes, such as athleti-cism). As discussed at length in this chapter,when specific visual cues become linked tolikely relevant outcomes in the environment,


� �

�


they also provide a source of prediction andexpectation that influences visual processing.A fundamental aspect of stereotypes is thatthey are generalized to all members of a socialgroup, and researchers traditionally assumedthat these assumptions were triggered afterplacing an individual in a social category,as a cognitive strategy to guide effectiveand appropriate social interaction (Allport,1954). However, contemporary work hascome to appreciate the ability of stereotypesto continuously guide perception as well,before the process of social categorizationis complete.

Social categorization has a powerful orga-nizing effect on perception. For example,racially ambiguous faces that have beencategorized as “Black” are subsequentlyperceived to have a darker skin tone (Levin &Banaji, 2006) and more Afrocentric facialcues (MacLin & Malpass, 2001). However,accumulating evidence suggests that thiskind of top-down feedback from stereotypesto the visual system also can occur beforesocial categorizations are complete. In onestudy, researchers presented participants withface stimuli that were morphed to be highlysex typical (e.g., masculine male face) or sexatypical (e.g., feminine male face). Partic-ipants were tasked with stereotyping thesetargets by choosing one of two adjectives(e.g., “caring,” “aggressive”) that they feltwas most stereotypically associated with thetarget face. By recording the trajectories ofparticipants’ computer mouse movements enroute to one of the two responses, researchersfound that when participants were stereo-typing atypical targets (e.g., male faceswith some female cues), mouse movementscontinuously deviated toward the adjectivestereotypically associated with the compet-ing category (e.g., the stereotypically femaleadjective “caring”). When stereotyping anatypical target, the correct social category(“male”) and the incorrect social category

(“female”) both become tentatively acti-vated as potential ways to categorize a face.This study provided initial evidence thatstereotypes associated with specific socialcategories also are activated before a stablecategorization has been reached (Freeman &Ambady, 2009).

Stereotypes also can guide categorizationsof perceptually ambiguous groups, as in thedomain of sexual orientation. Some work hasshown that participants utilize specific facialcues when tasked with categorizing sexualorientation (Rule & Ambady, 2008a; Ruleet al., 2008), but these cues often reflect atop-down stereotypic heuristic. Specifically,in one set of studies, researchers found thatfaces were more likely to be categorizedas gay or lesbian when a greater degreeof gender “inversion” was present on theface—gender incongruency among multiplegendered facial cues (Freeman, Johnson,Ambady, & Rule, 2010). Thus, due to theculturally pervasive stereotype that gay menare feminine and lesbian women are mas-culine, the presence of feminine cues on amale face or masculine cues on a femaleface consistently biased categorization. Thisis finding consistent with other researchthat found similar effects among multiplegendered bodily cues (Johnson et al., 2007)and is an example of the impact of top-downstereotypes on social categorization as wellas the link between stereotypic assumptionsand specific cues present in the environment.

Intersectionality effects, discussed brieflyin the subsection titled “Facial Cues,” haveattracted increasing attention in the litera-ture as an interesting example of stereotypefeedback on visual perception and socialcategorization. Although multiple socialcategories certainly can intersect becauseof an overlap in diagnostic phenotypic cues(e.g., Afrocentric features also contain cuesto the “male” sex category; Johnson et al.,2012), multiple social categories can become


� �

�


linked based on an overlap in stereotype asso-ciations. When cues on a face activate onesocial category and its associated knowledgestructures and expectations (e.g., male cueson a face activating the concept “male” andassociated expectations about an individual,such as “aggressive”), this facilitates thecategorization of other social categories thatincidentally share similar stereotypic expec-tations (e.g., the race category “Black,” whichshares the implicit expectation of aggressionwith the “male” category). Behaviorally,these stereotype overlaps can cause onecategory (e.g., “male”) to facilitate recog-nition of another category along a differentdimension (e.g., “Black”). They can pro-vide a top-down source of prediction onthe categorization of ambiguous stimuli,such that gender-ambiguous Black facesconsistently are categorized as males. Thisoverlap also means that certain individuals,such as Black females, experience stereotypeincongruence during social categorization.Stereotype overlap between the “Black” and“male” categories means that Black femalefaces partially activate the “male” category,which impedes sex categorization (as withAsian male faces, which partially activatethe “female” category; Johnson et al., 2012).Similar effects have been shown betweenintersecting race and emotion categories(e.g., “Black” and “angry”; Hugenberg &Bodenhausen, 2004) and sex and emotioncategories (e.g., “female” and “joy”; Hesset al., 2000). In this case, the conceptualstructure of stereotypes and their inherentlypredictive nature allow social categoriesand the facial cues associated with them toprovide a visual context for the perceptionof other social categories, even those fromorthogonal dimensions (for a review, seeJohnson & Freeman, 2010). Indeed, a recentstudy showed that individual differences instereotype overlap (e.g., similarity in concep-tual content of stereotypes for the “Black”

and “male” social categories) predicted theamount of perceptual intersectionality effectsas measured with computer mouse-tracking(Stolier & Freeman, 2016).

Motivation

Motivation has been explored extensivelyas an influence on perception harbored inthe perceiver. Transient motivational statescan influence visual perceptions, such thatindividuals will perceive ambiguous stimuliin line with whatever interpretation will havea positive outcome for them (Balcetis &Dunning, 2006; Voss et al., 2008). How-ever, motivation also can be thought of as achronic tuning of perception toward whateveraspects of the environment are most adap-tively relevant or useful, as we addressedin the introduction in our discussion of theenvironmental cues preferentially attended toby perceivers. Typically, these motivationalinfluences bear weight on perception evenwhen they reside outside of conscious aware-ness. For example, participants are morelikely to identify an impoverished image ofa gun as a gun when primed with a Blackface, because of the association of AfricanAmericans with crime in the United Statesand the motivation to recognize and respondto potential threats in the environment(Eberhardt, Goff, Purdie, & Davies, 2004).In the realm of social categorization,researchers have shown that unconsciousbiological factors, such as fertility in women(Johnston, Arden, Macrae, & Grace, 2003),and conscious motivational states, suchas sexual desire (Brinsmead-Stockham,Johnston, Miles, & Macrae, 2008), can mod-ulate the process of sex categorization toincrease efficiency (speed and accuracy) ofthe recognition of potential mates.

Individual differences in overall vigilanceto certain social cues also modulate per-son perception. These can be thought of as


� �

�


individual differences in “attunements” toparticular aspects of the environment, in theGibsonian sense. For example, adults highin attachment anxiety have a speed-accuracytrade-off in perceiving emotional expressionson faces: Although they are hypervigilant tothe presence of emotion cues on a face andfast to identify an emotion when a neutralface dynamically changes its expression,these assessments typically are less accuratethan those made by nonanxious individuals(Fraley, Niedenthal, Marks, Brumbaugh, &Vicary, 2006). As with visual context, per-ceivers rely more on motivation and otherperceiver inputs when the target of perceptionis ambiguous. (For a discussion, see Paukeret al., 2010.) For example, individuals highin stigma consciousness (the expectation ofbeing stereotyped and stigmatized by others;Pinel, 1999) are more likely to interpret anambiguous emotional expression as express-ing contempt (Inzlicht, Kaiser, & Major,2008). On the other end of the spectrum,White individuals high in racial prejudice aresignificantly more likely to categorize raciallyambiguous angry faces as Black (Hugen-berg & Bodenhausen, 2004; Hutchings &Haddock, 2008). Although the intersectionbetween the emotion concept “anger” andthe race category Black is common in theUnited States, as previously discussed, Whiteindividuals high in racial prejudice have anexaggerated version of this effect due to theirparticular motivation to recognize what theyview as a threatening social group.

Intergroup Processes

Humans possess an intrinsic drive to formsocial groups and behave in ways that sustainthose groups. From this tendency to supportand protect the in-group while remain-ing suspicious of the out-group, a numberof perceptual consequences emerge whencategorizing an individual as an in-group

or out-group member. Indeed, a growingbody of research suggests a fundamentaldivergence in the early visual processingof in-group versus out-group faces. A sub-stantial amount of early evidence for anin-group/out-group processing distinctioncame from observations of the “cross-raceeffect,” a robust phenomenon where recog-nition memory is better for own-race ratherthan other-race faces (Meissner & Brigham,2001). A great deal of research looked intothe underlying mechanisms of this effect,with evidence accumulating to suggest possi-ble contributions of perceptual expertise andgreater familiarity with the racial in-group(MacLin & Malpass, 2001), increased indi-viduation (versus mere categorization) ofown-race faces (Hugenberg, Young, Bern-stein, & Sacco, 2010), and, importantly,poorer encoding of other-race faces due toa divergence in the way other-race faces areprocessed visually. In particular, findingssuggest that perception of racial out-groupfaces relies on featural processing (encod-ing of isolated focal cues on a face) versusconfigural processing (better encoding ofthe gestalt spatial layout of a face; Michel,Rossion, Han, Chung, & Caldara, 2006)and that the effect may manifest at a verylow level in differences in how attentionis allocated to different parts of the face(Hills & Lewis, 2006, 2011; Hills & Pake,2013). Early evidence began to suggest thatthis shift in processing was the fundamentalmechanism behind the cross-race effect, overand above a lack of familiarity or exper-tise with bottom-up cues more prevalent inracial out groups. For example, researchersshowed that identical racially ambiguousfaces, which contain both own-race andother-race cues, were processed more holis-tically whenever categorized as own race(Michel, Corneille, & Roisson, 2007).

Meanwhile, research also accumulatedto suggest that the cross-race effect might


� �

�


be one instance of a broader cross-categoryeffect, where there is simply better encodingof in-group versus out-group faces in allcases. A powerful paradigm used to inves-tigate these and other effects in intergrouprelations is the minimal group paradigm, inwhich participants are assigned to an arbi-trary “minimal” group with which they haveno prior knowledge, familiarity, or stereo-types (Tajfel, 1970). Studies have shown thatwhen such arbitrary coalitions are created inan experimental setting, participants rapidlyadapt to this new context and even come tospontaneously categorize individuals alongnew arbitrary category dimensions morereadily than they do by race. Researchershave speculated that these effects show thatrace dominates perception only because it issuch a visually salient category dimension,not necessarily because it is the most impor-tant or primary of human coalitional divisions(Kurzban, Tooby, & Cosmides, 2001). Otherwork has shown that these effects extendto evaluative dimensions, such that partici-pants show implicit favoritism for minimalin-group members (Van Bavel & Cunning-ham, 2009). Importantly, cross-categoryeffects also have been shown for minimalgroups, such that participants have betterrecognition memory for in-group versusout-group faces even when these groups areminimal (Van Bavel, Packer, & Cunningham,2012). In both of these studies, the mini-mal groups included both Black and Whiteindividuals.

The impressive effects observed in mini-mal group paradigms hint at a social-cognitiveinfluence on face perception that is drivenmainly by group membership. Indeed, thesalience of group identity also can impactthe perception of natural groups, such asrace. Chiao, Heck, Nakayama, and Ambady(2006) explored this by priming biracial indi-viduals with either a Black or White identity.When primed with their Black identities,

participants subsequently performed a visualsearch task with Black and White faces muchlike Black participants (i.e., faster visualsearch times for Black faces). The oppositeeffect was observed when participants wereprimed with their White identities, suggestingthat top-down effects of group membershipon perception can be modulated by individualdifferences in the salience of group identity.These effects also are influenced by endur-ing characteristics of the perceiver, such asindividual differences in lay beliefs abouttraits and categories (i.e., essentialism ver-sus incrementalism). For example, biracialindividuals tend to have more flexible con-ceptions of group membership, which leadsto differences in their reliance on context cuesto categorize faces along the race dimension.Specifically, individuals belonging to onlyone race tend to have more essentialistconceptions of group identity, leading toa greater reliance on contextual cues (e.g.,verbal race labels) to disambiguate process-ing of racially ambiguous faces (Pauker &Ambady 2009).

However, there also is some evidence forthe role of familiarity and personal expe-rience with out-group members in cross-category perception. An extensive literaturehas documented these effects by examin-ing perceptual boundaries between socialcategories—the point in a continuum ofmorphed faces where participants perceivea transition from one category to another.Importantly, these boundaries are malleableand can be influenced by very recent context.For example, repeated exposure to male facesmoves the gender category boundary towardthe male category (making participants moreconservative about categorizing faces asmale, ultimately categorizing more facesas female; Webster, Kaping, Mizokami, &Duhamel, 2004). Similarly, the categoryboundary between White and Asian facesshifts toward the White category when target


� �

�


faces are presented in a sequence of Whitefaces. These shifts also can be induced bymore chronic changes in exposure and fre-quency of encounters with out-group faces.For example, Asian college students show ashift in their perceptual boundary betweenthe White and Asian categories when theylive in the United States for approximately1 year (Webster et al., 2004). In general,researchers find that people tend to shiftboundaries toward their own social iden-tity, which potentially reflects an adaptivemechanism—humans are conservative aboutcategorizing novel individuals as in-groupmembers, particularly when such individualsare ambiguous in-group membership, sincemistakes in this categorization could becostly (Tajfel & Turner, 1979).

Finally, it is important to note the diffi-culty in separating genuine top-down effectson perception from top-down influences onpostperceptual processes, such as responsebiases or recognition behavior. This issue hasled some researchers to criticize the idea thattop-down factors are able to penetrate per-ceptual representations (Firestone & Scholl,2016), drawing on older ideas of func-tional modularity (Fodor, 1983; Pylyshyn,1999). In this ongoing debate, we believemore implicit behavioral measures (such asmouse-tracking) and neuroimaging methodsthat can better assess perceptual represen-tations (e.g., multivariate fMRI) and do notrequire explicit responses may be helpful inaddressing to what extent higher-order socialcognition can shape lower-level perceptualrepresentations.

COMPUTATIONAL AND NEURALMECHANISMS

A great deal of work in social neurosciencehas aimed to determine which aspects of thebrain’s visual processing regions are selective

for faces and bodies. This work built onprominent early models of face processing(Bruce & Young, 1986), which focused onisolating the neural systems responsible forprocessing invariant qualities of a face, suchas a face’s identity, versus variant qualities,such as dynamic facial expressions. Researchon the neural systems of face perceptionin turn built on the incredible progress inneuroscience describing the structure ofbasic visual processing. After visual sensoryinformation hits the retina, it is relayed viaoptic nerve fibers that largely terminate inthe lateral geniculate nucleus of the thala-mus. The lateral geniculate nucleus in turnrelays information to primary visual cortex(striate cortex), where it initially retainsits retinotopic coding (i.e., purely veridicalneural processing of the light on the retina)(Bullier, 2002). Visual information is furtherprocessed by two extrastriate pathways,known as the dorsal and the ventral visualstreams (Goodale & Milner, 1992). Theventral visual stream, also known as the“what” stream, comprises ventral aspectsof occipitotemporal cortex, key processingregions for the visual recognition and cate-gorization of objects. A great deal of workin social and cognitive neuroscience hasfocused on delineating which aspects of theventral visual stream are selective for facesand bodies and how these neural systemsinteract with larger-scale brain networks.Indeed, the increasing capabilities of socialand cognitive neuroscience to characterizeindividual cognitive functions in terms of thecollaborative activity of multiple large-scalebrain networks stands to revolutionize theorydevelopment in social and cognitive psychol-ogy, including the field of person perception(Barrett & Satpute, 2013)

In person perception, the ventral visualstream is primarily responsible for process-ing static facial cues (Haxby et al., 2000).In particular, early feature-based processing


� �

�

Computational and Neural Mechanisms 449

is undertaken by the occipital face area(OFA), with higher-level representation ofthe configural properties of a face occur-ring in the fusiform gyrus (FG). The FGin particular has been the focus of intensescrutiny regarding its precise role in faceprocessing. This debate follows from thediscovery of a functionally defined region inthe FG commonly referred to as the fusiformface area (FFA; Kanwisher, McDermott, &Chun, 1997), which appears strongly selec-tive for faces and has been posited as aface perception module. Regardless of thespecific computations performed by the FFAand its status as a truly distinct functionalmodule, the FG/FFA has undisputed primacyin the neural processing of faces (Haxbyet al., 2000). Moreover, because of theventral visual stream’s role in categorizingvisual objects in general, and its particu-lar sensitivity to faces, it is unsurprisingthat these regions are routinely implicatedin social category representation. Multi-voxel pattern analyses (MVPA) of fMRIdata, which are able to isolate the uniquepatterns of neural activity associated withspecific stimulus conditions (Norman, Polyn,Detre, & Haxby, 2006), are consistentlyable to isolate specific representations forsocial categories in the race (Contreras,Banaji, & Mitchell, 2013) and sex (Kaul,Rees, & Ishai, 2011) dimensions. Similarly,processing of static bodily cues is subservedby the extrastriate body area and fusiformbody area (Peelen & Downing, 2007). How-ever, when bodies provide a disambiguatingcontext for the social categorization offaces, the meaningful social category infor-mation still appears to be encoded in theFG/FFA (Cox, Meyers, & Sinha, 2004).Additionally, fascinating work shows thatgender-specific olfactory cues (i.e., com-pounds that mimic sex hormones) also elicitFG activity (Savic, Berglund, Gulyas, &Roland, 2001), suggesting a more general

sensitivity to social category informationin the FG.

The ventral visual stream is also sensitiveto several of the top-down factors discussedpreviously. An important aspect of visual pro-cessing is that sensory input becomes moreconstrained by conceptual and predictive fac-tors along more anterior aspects of the ventralvisual stream (Grill-Spector & Weiner, 2014).Although early feature processing occurs inthe OFA, more anterior regions, such asthe FG, involved in forming higher-orderperceptual characteristics of a face largelyare constrained by dense structural and func-tional connections with higher-order regions,such as the orbitofrontal cortex (OFC),which comprises the most ventral aspects ofthe prefrontal cortex (PFC; Zanto, Rubens,Thangavel, & Gazzaley, 2011). Some evi-dence also suggests that the anterior temporallobe, at the end of the ventral visual process-ing stream, houses amodal representations offace identity (Anzellotti & Caramazza, 2014).However, even the FG appears to be sensi-tive to complex information about a face’sidentity (e.g., prior knowledge and famil-iarity; Rotshtein, Henson, Treves, Driver, &Dolan, 2005). Moreover, the FG shows anin-group/out-group distinction even whenthe groups in question are minimal groups,suggesting that intergroup effects in the FGreflect higher-level coding of perceptualtargets beyond visual features or familiarityeffects (Van Bavel et al., 2008). It is worthnoting that in this case, “in-group/out-groupdistinction” simply means that there wasgreater activity in the right FG. Since FGactivity reflects higher-level processing offaces, including configural processing andenhanced encoding, this simple distinc-tion in activity could suggest a potentialcontributing mechanism to cross-categoryeffects. However, examining neural activ-ity just in terms of overall mean activationacross a brain region sometimes obscures a


� �

�


more nuanced story. A recent fMRI studydivided participants into mixed-race minimalgroups and showed that even though theFG shows increased activity for minimalin-group faces, face race still is representa-tionally distinct in the FG as measured byMVPA (Ratner, Kaul, & Van Bavel, 2013).MVPA also reveals an interesting patternof effects for individuals who are implicitlyprejudiced, who show more distinct (i.e.,dissimilar) neural representations of own-and other-race faces in the FG (Brosch,Bar-David, & Phelps, 2013). Since evalua-tive bias causes individuals to see out-groupmembers as more similar (and thus moredifferent from in-group members; Ostrom &Sedikides, 1992), these findings suggest thatthese individual differences in evaluative biasare detectable at the level of neural repre-sentations of social categories and reflect aperceptual bias.

Recent work also has used MVPAapproaches to examine how stereotype inter-sectionality effects can bias neural repre-sentations of faces in the FG and OFC.Importantly, the OFC is theorized to constrainface representations in the FG via automaticactivation of stereotypes and associated pre-dictions and expectations (Knutson, Mah,Manly, & Grafman, 2007; Milne & Grafman,2001). In a recent set of studies, researcherspresented participants with faces crossed ongender, race, and emotion categories. Partic-ipants also completed a mouse-tracking taskthat indexed individual differences in per-ceptual biases (e.g., subjective perceptions ofsimilarity between the “Black” and “male”categories) as well as a stereotype contentstask to assess individual differences in con-ceptual knowledge (i.e., stereotypes) abouteach social category. Categories exhibitinggreater conceptual similarity (e.g., Blackand Angry) predicted greater perceptualbiases, and this biased similarity was in turnreflected in the representational similarity

of categories’ patterns of neural activity inthe FG and OFC. For example, a participantwhose stereotypes overlapped between the“Black” and “male” categories tended to besee faces belonging to those categories moresimilarity (perceptual biases assessed withmouse-tracking), and FG and OFC neuralpatterns representing the “Black” and “male”categories were consistently more similar,even when controlling for any bottom-upphysical similarity (Stolier & Freeman,2016). This study provides evidence thatstereotype overlap indeed can bias relativelylow-level perceptual representations of a facein a top-down direction.

Although extrastriate regions along theventral visual stream are widely acknowl-edged substrates of static cue processing,with important implications for social cate-gorization, dynamic facial expressions andhuman movement in general appear to beprocessed primarily by the superior tem-poral sulcus (STS), an aspect of the lateraltemporal lobe (Haxby et al., 2000). MVPAapproaches implicate the STS in representingemotion category information (Said, Moore,Engell, Todorov, & Haxby, 2010b), whichlargely depends on dynamic facial cues. TheSTS also has been implicated in fMRI stud-ies of biological motion (Grossman et al.,2000). Ongoing research is attempting toresolve how static and dynamic cues areintegrated into stable percepts in the brain.The STS processes and represents both staticand dynamic cues but seems to receive thisinformation through divergent pathways. Forexample, transcranial magnetic stimulation(which temporarily produces effects similarto a focal lesion) to the OFA reduces STSresponses to static cues but not to dynamiccues (Pitcher, Duchaine, & Walsh, 2014),suggesting that the STS receives informa-tion about static cues from the OFA butreceives information about dynamic cuesfrom another, as-yet undescribed pathway.


� �

�

Computational and Neural Mechanisms 451

This is particularly interesting, given the factthat there are no direct white matter path-ways from the OFA to the STS (Gschwind,Pourtois, Schwartz, Van De Ville, & Vuilleu-mier, 2012), further complicating the natureof the relationship. Additional work isnecessary to disentangle the processesthat integrate static and dynamic cues inperson perception.

Researchers interested in trait attributionand impression formation have focused alarge amount of their work on the medialPFC (MPFC), which appears to represent theabstract knowledge about personality traitsand mental states inferred about anotherperson (Mitchell, Banaji, & Macrae, 2005).This work has isolated an interesting dis-tinction between ventral and dorsal aspectsof the MPFC, observing that DMPFC hasa more central role in integrating personknowledge with trait attributions gleanedfrom faces (Ferrari et al., 2016). How-ever, the visual perception of personalitytraits, as discussed previously, depends onmany of the same perceptual mechanismsresponsible for representing social categoryinformation, and as such, similar regionslike the FG also are recruited during traitattributions, such as judgments of trust-worthiness (Todorov & Engell, 2008) andbaby-facedness (which correlate with judg-ments of competence/dominance; Zebrowitz,Luevano, Bronstad, & Aharon, 2009). Theamygdala, a subcortical structure locateddeep in the medial temporal lobes, alsoplays an interesting role tin perceptions oftrustworthiness. The amygdala is respon-sive to trustworthy and untrustworthy faces(Engell, Haxby, & Todorov, 2007), andtracks trustworthiness cues even when facesare presented subliminally (Freeman, Stolier,Ingbretsen, & Hehman, 2014). The amygdalaseems preferentially involved in spontaneoustrait inference (i.e., rapid personality judg-ments made without prior knowledge about

faces; Todorov & Engell, 2008), consistentwith interpretations of amygdala activity asreflecting motivational salience in the envi-ronment more generally, signaling relevantdetails in the visual field and promptingincreased processing (W. A. Cunningham &Brosch, 2012). Consistent with recent evi-dence of the role of facial averageness intrustworthiness judgments (Sofer et al.,2015), the amygdala also has been shown torespond to the atypicality of faces broadly,which may contribute to its observed rolein tracking trustworthiness (Said, Dotsch, &Todorov, 2010a).

Although much work in neuroimaging hasfocused on specific regions that seem selec-tively tuned to faces, recent work in socialand cognitive neuroscience has examined thecollaborative role of multiple domain-generalneural regions in face perception, with a par-ticular focus on the regions responsible forimplementing top-down effects. To char-acterize a large-scale network involved intop-down face perception, researchers inone study “trained” participants to recognizefaces embedded in visual noise. Participantssaw faces increasingly obscured by noiseand, on critical trials, saw only noise, withno face present. However, participants stillreported the ability to detect faces in the purenoise trials (Li et al., 2009). Researchersexamined these trials to explore face process-ing from a completely top-down direction(i.e., when literally no bottom-up facialcues were visually present), and they foundgreater activity in bilateral FFA and OFA.They additionally examined which regionswere most connected functionally with theFFA during illusory face trials and foundconsistently increased connectivity withthe left STS, bilateral OFC, left anteriorcingulate cortex (ACC, involved in con-flict monitoring; Botvinick, Braver, Barch,Carter, & Cohen, 2001), left dorsolateral pre-frontal cortex (DLPFC, involved in response


� �

�


inhibition; MacDonald, Cohen, Stenger, &Carter, 2000), left inferior parietal lobule(IPL), and left premotor cortex (Li et al.,2009). Based on further analysis of bothintrinsic and task-based functional connec-tivity between these regions (Li et al., 2010),the researchers outline a specific model fortop-down face processing, wherein the OFCshifts feature processing in the OFA basedon predictive assumptions, which can thencause false or biased detection of featuresin the environment, in turn biasing perceptsformed in the FG.

More work is needed to further character-ize the top-down face-processing network,but it is interesting to note in the meantimethat these neural regions were implicatedpreviously in downstream evaluative aspectsof social categorization, such as implicitevaluation (e.g., amygdala, OFC) and regu-latory processes exerted on these evaluativebiases (e.g., ACC, DLPFC; for a review,see Amodio, 2014). However, additionalrecent research converges with the idea ofa top-down face-processing network, impli-cating these regions in mechanisms of socialcategorization when faces simply are viewedpassively. A recent neuroimaging studyexamined neural responses when participantspassively viewed face stimuli crossed on raceand emotion. Participants also completed amouse-tracking task to determine individualdifferences in subjective stereotype overlapbetween race and emotion categories. Asdiscussed previously, race and emotion cat-egories are linked perceptually because ofstereotypic associations (e.g., Black facesare expected to be angry, White faces areexpected to be joyful). As faces becamemore stereotypically incongruent (e.g., joyfulBlack faces and angry White faces), activityin the ACC increased linearly, and func-tional connectivity increased between theACC and FG, reflecting the need to resolvea processing conflict: Due to stereotypes,

participants expect to see angry Black faces,and regulatory processing is required tomanage the conflict presented by a joy-ful Black face. Importantly, in individualshigh in stereotype overlap, the DLPFC alsoshowed increased activity in response tostereotype-incongruent targets, suggestinga potential inhibition of the initial (incor-rect) stereotypical prediction of the face’semotion (Hehman, Ingbretsen, & Freeman,2014c). The fact that regions involved inconflict monitoring and response inhibitionresponded when participants were viewingfaces passively is an important insight into thelarge-scale mechanisms of face perception.

Ongoing findings from social neuro-science and social vision have inspiredtheoretical models that account for thedynamic interactivity between bottom-upand top-down factors observed in personperception. In particular, a computationalmodel has provided a framework for under-standing how perceptual cues interact withtop-down factors during social categorizationand how the precise nature of these interac-tions can result in many of the perceptualbiases reviewed in this chapter (Freeman &Ambady, 2011a). The dynamic interactive(DI) model describes computationally themultiple levels of person perception and howthese different levels interact dynamically.At the cue level, the model accounts forvisual (face, body, and visual contextual) andauditory (voice) cues, which have ascendantconnections to higher-level nodes at thecategory and stereotype level in additionto feedback connections from those nodes.The category level models social categories,such as sex, race, age, and emotion, whichhave bidirectional connections to nodes atthe stereotype level, allowing stereotypes toprovide a dynamic constraint on early-levelvisual processing of social categories, aswe have observed. At each level, nodesare activated in parallel and share lateral


� �

�

Downstream Consequences of Person Perception 453

connections. For example, individual facialcues can facilitate processing of one another,or they can inhibit one another via feedbackconnections from stereotype nodes linked toa particular facial feature. The model alsoaccounts for higher-level cognitive states,such as goals, motivations, and attention,which provide an additional constraint onprocessing. For example, a typical behav-ioral experiment that tasks a participant withcategorizing race activates race-related atten-tional and goal-oriented processing, whichcan guide and constrain processing at alllevels of the model.

Thus far, the DI model appears to modelsuccessfully many of the top-down effects onperson perception discussed in this chapter,including complex interactions betweenperceptual cues and stereotypes, such asintersectionality (Johnson et al., 2012) andovergeneralization effects (Zebrowitz et al.,2003). Recent work also has used the DImodel to demonstrate the role of interra-cial exposure in influencing perceptions ofracially ambiguous individuals. Using a largeonline sample from across the United States,researchers used computer mouse-trackingto show that White participants who live inparts of the country with higher exposureto Black individuals show stable coactiva-tion of the White and Black race categorieswhen categorizing a mixed-race individual.In contrast, White participants who live inareas with low exposure to Black individualsshowed highly unstable coactivation of theWhite and Black categories, with abruptshifts between the two categories reflected intheir computer mouse movements en routeto the “White” or “Black” category labels onthe screen. Additional modeling work sug-gested that these unstable dynamics arose inlow-exposure individuals because they holdstereotypes that White and Black individualsare highly dissimilar. Thus, when encoun-tering a mixed-race face, bottom-up visual

processing attempts to push the system toan in-between point between the White andBlack categories, while top-down concep-tual knowledge in a low-exposure perceivertries to rapidly pull the system into a givenrace category, creating unstable dynamics(Freeman, Pauker, & Sanchez, 2016). Impor-tantly, unstable shifts in race categoryactivation also predicted participants’ eval-uative bias toward mixed-race individuals,demonstrating that such downstream effectscan be represented by individual differencesin the structure and dynamics of the broadersystem (Freeman et al., 2016).

The strength of the DI model may lie inmodeling the system as massively interactive,allowing bottom-up perceptual cues, con-ceptual processing, and high-level cognitivestates to interact flexibly in real time duringthe processing of visual input. Future workis required to refine the model to account foradditional cognitive and affective states of theperceiver (e.g., goals, emotional state, atten-tion) and how these inputs affect lower-levelperception. To further situate our discussionof person perception in the complex andvariable nature of everyday social contexts,we now briefly discuss the impact that certainaspects of the person perception processcan have on interpersonal interaction anddownstream behavior.

DOWNSTREAM CONSEQUENCESOF PERSON PERCEPTION

Historically, social psychologists studiedthe downstream consequences of personperception and social categorization whilecognitive psychologists and neuroscientistsstudied the perceptual processes leading tothese categorizations and impressions. Muchearly work in social psychology discussedthe consequences of social categorization andstereotyping, recognizing the ability of these


� �

�


processes to guide behavior and evaluativedimensions in an unintended and implicitmanner (Bodenhausen & Macrae, 1998).However, this approach assumes a perceptualendpoint: Perceptual processes give rise toan initial impression or categorization, whichin turn triggers related stereotypes and theirassociated biasing power on behavior andinteraction. Moreover, researchers also typi-cally assumed that stereotypes were appliedto the same degree to all individuals placedin a given social category (Fiske & Taylor,1991). However, many of the ongoing con-tributions summarized in this chapter—frominterdisciplinary “social vision” approachesto current theoretical models of personperception—suggest that these processes arebest thought of as interactive and continu-ous, and many of the downstream effectsof person perception (such as attitudes andbiases that can come to pervade societyat an institutional level) are rooted in thesubtle dynamics of earlier processing stages(Freeman & Ambady, 2011a).

As hinted at in the earlier discussion offacial cues, visual context, and stereotypes,there is a complex relationship between fea-tures of the face and their related stereotypes.Consistent with these insights, emergingwork shows that attitudes and behaviorsemerging from stereotypes and related traitassumptions are not applied categoricallyto all members of a social category but areoften nuanced inferences that depend on thedegree to which an individual displays spe-cific category-relevant facial features (Blair,Chapleau, & Judd, 2004a). For example,members of marginalized racial groupsexperience more stigma and bias when theypossess facial cues that are highly typical anddiagnostic for their race (Maddox, 2004).In many cases, these biases are enormouslyconsequential for members of marginalizedsocial groups. For example, there exists anunfortunate and well-documented “shooter

bias,” in which law enforcement officers aremore likely to shoot individuals belonging tosocial categories commonly stereotyped ashostile and aggressive, such as African Amer-icans (Correll, Park, Judd, & Wittenbrink,2002; but see James, Vila, & Daratha, 2013;James, Klinger, & Vila, 2014). Additionally,researchers found that Black felony offenderswith more Afrocentric facial features aregiven longer criminal sentences, even whencontrolling for the number and severity ofcrimes committed and the individuals’ pastcriminal histories (Blair, Judd, & Chapleau,2004b). These featural biases underscorethe importance of studying the impact ofbottom-up cues in person perception.

Facial cues also can lead to consequen-tial social outcomes when they introduceinconsistencies and incongruities into socialcategorization. For example, women withmasculine facial features are less likely tobe elected to political office in conservativestates of the United States (Hehman et al.,2014a). There are also several negativesocial outcomes for individuals belonging tointersectional social categories (e.g., Blackindividuals and Asian individuals, because ofthe association of the “Black” category withmasculinity and the “Asian” category withfemininity). In a collection of studies, Galin-sky, Hall, & Cuddy (2013) found that Blackwomen and Asian men are less successfulin heterosexual romantic relationships, andBlack individuals in general are more likelyto be represented in social institutions thatprivilege masculinity (such as business lead-ership and sports). These patterns were foundto be consistent between lab-based paradigmsand large-scale data from the United StatesCensus and the NCAA Student-AthleteEthnicity Report (Galinsky et al., 2013).

Rapid trait-based inferences, whichdepend on invariant facial cues and often aresubject to systematic bias, also can result ina number of consequences for the targets of


� �

�

References 455

perception. The consequences of perceptionsof dominance and competence have beenwidely studied in the context of leadershipselection and compensation. For example, anumber of studies show that individuals withfaces judged to be competent are more likelyto be elected to political office (Ballew &Todorov, 2007). In the business domain, suchindividuals are much more likely to be hiredby successful companies and receive highersalaries (Rule & Ambady, 2008b). Addition-ally, individuals with untrustworthy-lookingfaces are more likely to receive guilty verdictsin court, even when there is scant evidenceof their guilt (Porter, ten Brinke, & Gustaw,2010). On the other end of the spectrum,baby-faced individuals (who are consistentlyperceived as more trustworthy and honest)are more likely to win their legal casesin court (Zebrowitz & McDonald, 1991).Although these insights are discouraging, itis remarkable that early perceptual biasesin person perception can come to drive per-vasive trends in society at an institutionallevel. Further work is needed to understandhow these insights can be used to implementpolicies to reduce these systematic trends.

CONCLUSION

The human perceptual system is remarkablyattuned to social information in the environ-ment. As such, a particular attentional focuson faces, bodies, and voices emerges earlyin development and remains a potent forcein social perception and interaction through-out the life span. Emerging from the vastcognitive resources deployed to make senseof the social environment is an incrediblehuman propensity to infer quickly and oftenaccurately information from the qualities ofanother person’s face and body, although,as we have discussed, the efficiency of thissystem sometimes renders it vulnerable to

systematic biases. Since interpersonal per-ception, communication, competition, andcooperation are such integral parts of sociallife, these biases can be incredibly costly,marring social behavior. Recent theoreti-cal insights, motivated by interdisciplinarycollaboration among the social, cognitive,neural, and vision sciences, have provided adeeper understanding and appreciation forhow the complexity of the social environmentis perceived and understood. The emergingscientific understanding of person perceptionyields a fascinating picture: a complex anddynamic system encompassing interactionamong the perceptual cues available in theenvironment and the cognitive systems inher-ent to the perceiver. In this chapter, we havebut scratched the surface of this fascinatingresearch domain, and we look forward tofuture work building an increasingly richunderstanding of how we perceive and shapethe social world.

REFERENCES

Alaei, R., & Rule, N.O. (2016). Accuracy ofperceiving social attributes. In J. A. Hall, M.Schmid Mast, & T. V. West (Eds.), The socialpsychology of perceiving others accurately(pp. 125–142). New York, NY: CambridgeUniversity Press.

Allport, G. W. (1954). The nature of prejudice.Oxford, UK: Addison-Wesley.

Ambadar, Z., Schooler, J. W., & Cohn, J.F. (2005).Deciphering the enigmatic face: the impor-tance of facial dynamic in interpreting subtlefacial expressions. Psychological Science, 16,403–410.

Ambady, N., Hallahan, M., & Conner, B. (1999).Accuracy of judgments of sexual orientationfrom thin slices of behavior. Journal of Person-ality and Social Psychology, 77, 538–547.

Ambady, N., & Rosenthal, R. (1993). Half aminute: predicting teacher evaluations fromthin slices of nonverbal behavior and physical


� �

�


attractiveness. Journal of Personality and SocialPsychology, 64(3), 431–441.

Amodio, D. M. (2014). The neuroscience of prej-udice and stereotyping. Nature Reviews Neuro-science, 15(10), 670–682.

Anderson, E., Siegel, E. H., Bliss-Moreau, E., &Barrett, L. F. (2011). The visual impact ofgossip. Science, 332, 1446–1448.

Anzellotti, S., & Caramazza, A. (2014). The neuralmechanisms for the recognition of face identityin humans. Frontiers in Psychology, 5, 672.

Atkinson, A. P., Dittrich, W. H., Gemmell, A. J., &Young, A. W. (2004). Emotion perception fromdynamic and static body expressions in point-light and full-light displays. Perception, 33,717–746.

Atkinson, J. W., & Walker, E. L. (1956). The affili-ation motive and perceptual sensitivity to faces.Journal of Abnormal Psychology, 53, 38–41.

Balcetis, E., & Dunning, D. (2006). See what youwant to see: motivational influences on visualperception. Journal of Personality and SocialPsychology, 91, 612–625.

Ballew, C. C., & Todorov, A. (2007). Predict-ing political elections from rapid and unreflec-tive face judgments. Proceedings of the NationalAcademy of Sciences, 104, 17948–17953.

Bar, M. (2004). Visual objects in context. NatureReviews Neuroscience, 5, 617–629.

Bargh, J. A., Chen, M., & Burrows, L. (1996).Automaticity of social behavior: direct effectsof trait construct and stereotype activation onaction. Journal of Personality and Social Psy-chology, 71, 230–244.

Barrett, L. F., Mesquita, B., & Gendron, M. (2011).Context in emotion perception. Current Direc-tions in Psychological Science, 20, 286–290.

Barrett, L. F., & Satpute, A. B. (2013). Large-scalebrain networks in affective and social neu-roscience: towards an integrative functionalarchitecture of the brain. Current Opinion inNeurobiology, 23, 361–372.

Beale, J. M., & Keil, F. C. (1995). Categoricaleffects in the perception of faces. Cognition, 57,217–239.

Berry, D. S., & McArthur, L. Z. (1986). Perceivingcharacter in faces: the impact of age-related

craniofacial changes on social perception.Psychological Bulletin, 100, 3–18.

Bestelmeyer, P. E. G., Rouger, J., DeBruine,L. M., & Belin, P. (2010). Auditory adapta-tion in vocal affect perception. Cognition, 117,217–223.

Biederman, I. (1987). Recognition-by-compo-nents: a theory of human image understanding.Psychological Review, 94, 115–147.

Blair, I. V., Chapleau, K. M., & Judd, C. M.(2004a). The use of Afrocentric features as cuesfor judgment in the presence of diagnostic infor-mation. European Journal of Social Psychology,34, 1–10.

Blair, I. V., Judd, C. M., & Chapleau, K. M.(2004b). The influence of Afrocentric facial fea-tures in criminal sentencing. Psychological Sci-ence, 15, 674 – 679.

Blake, R. (2001). A primer on binocular rivalry,including current controversies. Brain andMind, 2, 5–38.

Bodenhausen, G. V., & Macrae, C. N. (1998).Stereotype activation and inhibition. In J. R.Wyer (Ed.), Advances in social cognition(Vol. 11, pp. 1–52). Mahwah, NJ: Erlbaum.

Botvinick, M., Braver, T., Barch, D., Carter,C., & Cohen, J. (2001). Conflict monitoring andcognitive control. Psychological Review, 108,624–652.

Brewer, M. B. (1988). A dual process modelof impression formation. Advances in SocialCognition, 1, 1–36.

Brinsmead-Stockham, K., Johnston, L., Miles,L., & Macrae, C. N. (2008). Female sexualorientation and menstrual influences on personperception. Journal of Experimental SocialPsychology, 44, 729–734.

Brosch, T., Bar-David, E., & Phelps, E. A. (2013).Implicit race bias decreases the similarity ofneural representations of black and white faces.Psychological Science, 24(2), 160–166.

Brown, E., & Perrett, D. I. (1993). What gives aface its gender? Perception, 22, 829–840.

Bruce, V., & Young, A. W. (1986). A theoreticalperspective for understanding face recognition.British Journal of Psychology, 77, 305–327.


� �

�

References 457

Bruner, J. S., & Goodman, C. C. (1947). Valueand need as organizing factors in perception.Journal of Abnormal and Social Psychology, 42,33–44.

Bullier, J. (2002). Neural basis of vision. In H.Pashler & S. Yantis (Eds.), Stevens’ handbook ofexperimental psychology, Vol. 1: Sensation andperception (pp. 1–40). Hoboken, NJ: Wiley.

Campanella, S., & Belin, P. (2007). Integratingface and voice in person perception. Trends inCognitive Sciences, 11, 535−543.

Carré, J. M., & McCormick, C. M. (2008).In your face: facial metrics predict aggres-sive behaviour in the laboratory and in varsityand professional hockey players. Proceedingsof the Royal Society B: Biological Sciences,275(1651), 2651–2656.

Carré, J. M., McCormick, C. M., & Mondloch,C. J. (2009). Facial structure is a reliable cue ofaggressive behavior. Psychological Science, 20,1194–1198.

Carroll, J. M., & Russell, J. A. (1996). Do facialexpressions signal specific emotions? Judgingemotion from the face in context. Journal of Per-sonality and Social Psychology, 70, 205–218.

Chiao, J., Heck, H. E., Nakayama, K., & Ambady,N. (2006). Priming race in biracial observersaffects visual search for Black and White faces.Psychological Science, 17, 387–392.

Contreras, J. M., Banaji, M. R., & Mitchell, J. P.(2013). Multivoxel patterns in fusiform facearea differentiate faces by sex and race. PLoSONE, 8(7), e69684.

Correll, J., Park, B., Judd, C. M., & Wittenbrink, B.(2002). The police officer’s dilemma: using eth-nicity to disambiguate potentially threateningindividuals. Journal of Personality and SocialPsychology, 83(6), 1314–1329.

Cox, D., Meyers, E., & Sinha, P. (2004). Contextu-ally evoked object-specific responses in humanvisual cortex. Science, 304(5667), 115–117.

Cunningham, M. R. (1986). Measuring thephysical in physical attractiveness: Quasi-experiments on the sociobiology of femalefacial beauty. Journal of Personality and SocialPsychology, 50, 925–935.

Cunningham, W. A., & Brosch, T. (2012). Moti-vational salience: amygdala tuning from traits,

needs, values, and goals. Current Directions inPsychological Science, 21(1), 54–59.

Cutting, J. E., & Kozlowski, L. T. (1977). Rec-ognizing friends by their walk: gait perceptionwithout familiarity cues. Bulletin of the Psycho-nomic Society, 9, 353–356.

Cutting, J. E., Proffitt, D. R., & Kozlowski,L. T. (1978). A biomechanical invariant for gaitperception. Journal of Experimental Psychol-ogy: Human Perception and Performance, 4,357–372.

de Gelder, B. (2005). Toward a neurobiology ofemotional body language. Nature Reviews Neu-roscience, 7, 242–249.

de Gelder, B., & Vroomen, J. (2000). The percep-tion of emotions by ear and by eye. Cognitionand Emotion, 14, 289−311.

Downing, P. E., Bray, D., Rogers, J., & Childs, C.(2004). Bodies capture attention when nothingis expected. Cognition, 93, B27–B38.

Eberhardt, J. L., Goff, P. A., Purdie, V. J., &Davies, P. G. (2004). Seeing black: race, crime,and visual processing. Journal of Personalityand Social Psychology, 87, 876–893.

Engell, A. D., Haxby, J. V., & Todorov, A.(2007). Implicit trustworthiness decisions: auto-matic encoding of face properties in the humanamygdala. Journal of Cognitive Neuroscience,19(9), 1508–1519.

Erdelyi, M. H. (1974). A new look at the new look:perceptual defense and vigilance. PsychologicalReview, 81(1), 1–25.

Evans, S., Neave, N., & Wakelin, D. (2006).Relationships between vocal characteristics andbody size and shape in human males: an evo-lutionary explanation for a deep male voice.Biological Psychology, 72, 160–163.

Ferrari, C., Lega, C., Vernice, M., Tamietto, M.,Mende-Siedlecki, P., Vecchi, T., Todorov, A., &Cattaneo, Z. (2014). The dorsomedial prefrontalcortex plays a causal role in integrating socialimpressions from faces and verbal descriptions.Cerebral Cortex, 26(1), 156–165.

Firestone, C., & Scholl, B. J. (2016). Cognitiondoes not affect perception: evaluating the evi-dence for “top-down” effects. Behavioral andBrain Sciences, 39: e229.


� �

�


Fiske, S. T., Cuddy, A. J. C., & Glick, P.(2007). Universal dimensions of social cogni-tion: Warmth and competence. Trends in Cog-nitive Sciences, 11, 77–83.

Fiske, S. T., & Taylor, S. E. (1991). Social Cogni-tion, 2nd ed. New York, NY: McGraw-Hill.

Fitch, W. T., & Giedd, J. (1999). Morphologyand development of the human vocal tract: astudy using magnetic resonance imaging. Jour-nal of the Acoustical Society of America, 106,1511–1522.

Fodor, J. A. (1983). The modularity of mind: anessay on faculty psychology. Cambridge, MA:MIT Press.

Fraley, R. C., Niedenthal, P. M., Marks, M. J.,Brumbaugh, C. C., & Vicary, A. (2006). Adultattachment and the perception of emotionalexpressions: probing the hyper-activating strate-gies underlying anxious attachment. Journal ofPersonality, 74, 1163–1190.

Freeman, J. B., & Ambady, N. (2009). Motions ofthe hand expose the partial and parallel activa-tion of stereotypes. Psychological Science, 20,1183–1188.

Freeman, J. B., & Ambady, N. (2011a). A dynamicinteractive theory of person construal. Psycho-logical Review, 118(2), 247–279.

Freeman, J. B., & Ambady, N. (2011b). Handmovements reveal the time-course of shape andpigmentation processing in face categorization.Psychonomic Bulletin & Review, 18, 705–712.

Freeman, J. B., & Ambady, N. (2011c). Whentwo become one: temporally dynamic integra-tion of the face and voice. Journal of Experimen-tal Social Psychology, 47, 259–263.

Freeman, J. B., Johnson, K. L., Ambady, A., &Rule, N. O. (2010). Sexual orientation per-ception involves gendered facial cues. Person-ality and Social Psychology Bulletin, 36(10),1318–1331.

Freeman, J. B., Ma, Y., Han, S., & Ambady, N.(2013). Influences of culture and visual contexton real-time social categorization. Journal ofExperimental Social Psychology, 49, 206–210.

Freeman, J. B., Pauker, K., & Sanchez, D. T.(2016). A perceptual pathway to bias: interracialexposure reduces abrupt shifts in real-time race

perception that predict mixed-race bias. Psycho-logical Science, 27(4), 502–517.

Freeman, J. B., Penner, A. M., Saperstein, A.,Scheutz, M., & Ambady, N. (2011). Looking thepart: social status cues shape race perception.PLoS ONE, 6, e25107.

Freeman, J. B., Stolier, R. M., Ingbretsen, Z. A., &Hehman, E. A. (2014). Amygdala respon-sivity to high-level social information fromunseen faces. Journal of Neuroscience, 34(32),10573–10581.

Freire, A., Lewis, T. L., Maurer, D., & Blake,R. (2006). The development of sensitivity tobiological motion in noise. Perception, 35(5),647–657.

Frischen, A., Bayliss, A. P., & Tipper, S. P. (2007).Gaze cueing of attention: visual attention, socialcognition, and individual differences. Psycho-logical Bulletin, 133, 694–724.

Galinsky, A. D., Hall, E. V., & Cuddy, A. J. (2013).Gendered races: implications for interracialmarriage, leadership selection, and athleticparticipation. Psychological Science, 24(4),498–506.

Gangestad, S. W., Simpson, J. A., DiGeronimo,K., & Biek, M. (1992). Differential accuracy inperson perception across traits: examination ofa functional hypothesis. Journal of Personalityand Social Psychology, 62, 688–698.

Gibson, J. J. (1979). The ecological approachto visual perception. Boston, MA: HoughtonMifflin.

Gilbert, D.T. (1998). Ordinary personology. InD. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.),The handbook of social psychology (4th ed.).New York, NY: McGraw Hill.

Gobbini, M. I., & Haxby, J. V. (2007). Neural sys-tems for recognition of familiar faces. Neuro-psychologia, 45, 32–41.

Goldberg, L. R. (1990). An alternate “descrip-tion of personality”: the big-five factor structure.Personality Processes and Individual Differ-ences, 59(6), 1216–1229.

Goodale, M. A., & Milner, A. D. (1992). Separatepathways for perception and action. Trends inNeurosciences, 15, 20–25.


� �

�

References 459

Grammer, K., & Thornhill, R. (1994). Human(Homo sapiens) facial attractiveness and sex-ual selection: the role of symmetry and average-ness. Journal of Comparative Psychology, 108,233–242.

Grill-Spector, K., & Weiner, K. S. (2014). Thefunctional architecture of the ventral tempo-ral cortex and its role in categorization. NatureReviews Neuroscience, 15(8), 536–548.

Grossman, E., Donnelly, M., Price, R., Pickens, D.,Morgan, V., Neighbor, G., & Blake, R. (2000).Brain areas involved in perception of biologi-cal motion. Journal of Cognitive Neuroscience,12(5), 711–720.

Gschwind, M., Pourtois, G., Schwartz, S., VanDe Ville, D., & Vuilleumier, P. (2012). White-matter connectivity between face-responsiveregions in the human brain. Cerebral Cortex,22(7), 1564–1576.

Hancock, P. J. B., Bruce, V., & Burton, A. M.(2000). Recognition of unfamiliar faces. Trendsin Cognitive Sciences, 4, 330–337.

Harrigan, J. A., Harrigan, K. M., Sale, B. A., &Rosenthal, R. (1996). Detecting anxiety anddefensiveness from visual and auditory cues.Journal of Personality, 64, 675–709.

Haselhuhn, M. P., & Wong, E. M. (2012). Badto the bone: facial structure predicts unethicalbehaviour. Proceedings of the Royal Society B:Biological Sciences, 279(1728), 571–576.

Hassin, R., Bargh, J. A., & Uleman, J. S.(2002). Spontaneous causal inferences. Journalof Experimental Social Psychology, 38,515–522.

Haxby, J. V., Hoffman, E. A., & Gobbini, M. A.(2000). The distributed human neural system forface perception. Trends in Cognitive Sciences, 4,223–233.

Haxby, J. V., Hoffman, E. A., & Gobbini, M. A.(2002). Human neural systems for face recog-nition and social communication. BiologicalPsychiatry, 51, 59–67.

Hehman, E., Carpinella, C. M., Johnson, K. L.,Leitner, J. B., & Freeman, J. B. (2014a).Early processing of gendered facial cues pre-dicts the electoral success of female politicians.Social Psychology and Personality Science, 5,815–824.

Hehman, E., Flake, J. K., & Freeman, J. B. (2015).Static and dynamic facial cues differentiallyimpact the consistency of social evaluations.Personality and Social Psychology Bulletin, 41,1123–1143.

Hehman, E., Ingbretsen, Z. A., & Freeman, J. B.(2014c). The neural basis of stereotypic impacton multiple social categorization. PsychologicalScience, 15, 342–345.

Hehman, E., Leitner, J. B., & Gaertner, S. (2013).Enhancing static facial features increases intimi-dation. Journal of Experimental Social Psychol-ogy, 49, 747–754.

Hehman, E., Leitner, J. B., & Freeman, J. B.(2014b). The face-time continuum: lifespanchanges in facial width-to-height ratio impactaging-associated perceptions. Personality andSocial Psychology Bulletin, 40, 1624–1636.

Helmholtz, H. (1867). Concerning the percep-tion in general. In Readings in the historyof psychology. East Norfolk, CT: Appleton-Century-Crofts.

Hess, U., Adams, R. B., & Kleck, R. E. (2004).Facial appearance, gender, and emotion expres-sion. Emotion, 4, 378–388.

Hess, U., Adams, R. B., & Kleck, R. E. (2005).Who may frown and who should smile?Dominance, affiliation, and the display of hap-piness and anger. Cognition & Emotion, 19,515–536.

Hess, U., Senécal, S., Kirouac, G., Herrera, P.,Philippot, P., & Kleck, R. E. (2000). Emotionalexpressivity in men and women: Stereotypesand self-perceptions. Cognition & Emotion,14(5), 609–642.

Hill, H., Bruce, V., & Akamatsu, S. (1995). Per-ceiving the sex and race of faces: the role ofshape and colour. Proceedings of the RoyalSociety B, 261, 367–373.

Hills, P. J., & Lewis, M. B. (2006). Reducing theown-race bias in face recognition by shiftingattention. Quarterly Journal of ExperimentalPsychology, 59, 996–1002.

Hills, P. J., & Lewis, M. B. (2011). Reducing theown-race bias in face recognition by attentionalshift using fixation crosses preceding the lowerhalf of a face. Visual Cognition, 19, 313–339.


� �

�


Hills, P. J., & Pake, J. M. (2013). Eye-trackingthe own-race bias in face recognition: revealingthe perceptual and socio-cognitive mechanisms.Cognition, 129, 586–597.

Hugenberg, K., & Bodenhausen, G.V. (2004).Ambiguity in social categorization: the role ofprejudice and facial affect in race categorization.Psychological Science, 15, 342–345.

Hugenberg, K., Young, S. G., Bernstein, M. J., &Sacco, D. F. (2010). The categorization-individuation model: an integrative account ofthe other-race recognition deficit. PsychologicalReview, 117(4), 1168–1187.

Hughes, S. M., & Rhodes, G. (2010) Making ageassessments based on voice: the impact of thereproductive viability of the speaker. Journal ofSocial, Evolutionary, & Cultural Psychology, 4,290–304.

Hutchings, P. B., & Haddock, G. (2008). Lookblack in anger: the role of implicit prejudicein the categorization and perceived emotionalintensity of racially ambiguous faces. Jour-nal of Experimental Social Psychology, 44,1418–1420.

Inzlicht, M., Kaiser, C. R., & Major, B. (2008). Theface of chauvinism: how prejudice expectationsshape perceptions of facial affect. Journal ofExperimental Social Psychology, 44, 758–766.

Ito, K., Masuda, T., & Hioki, K. (2012). Affectiveinformation in context and judgment of facialexpression: Cultural similarities and variationsin context effects between North Americans andEast Asians. Journal of Cross-Cultural Psychol-ogy, 43, 429–445.

James, L., Klinger, D. & Vila, B. (2014). Racialand ethnic bias in decisions to shoot seenthrough a stronger lens: experimental resultsfrom high-fidelity laboratory simulations. Jour-nal of Experimental Criminology, 10, 323–340.

James, L., Vila, B., & Daratha, K. (2013). Influ-ence of suspect race and ethnicity on decisionsto shoot in high fidelity deadly force judgmentand decision-making simulations. Journal ofExperimental Criminology, 9, 189–212.

Johansson, G. (1973). Visual perception of biolog-ical motion and a model for its analysis. Percep-tion and Psychophysics, 14, 201–211.

Johansson, G. (1975). Visual motion perception.Scientific American, 232, 76–89.

Johnson, K. L., & Freeman, J. B. (2010). A “newlook” at person construal: seeing beyond dom-inance and discreteness. In E. Balcetis & D.Lassiter (Eds.), The social psychology of visualperception (pp. 253–272). New York, NY:Psychology Press.

Johnson, K. L., Freeman, J. B., & Pauker, K.(2012). Race is gendered: how covarying phe-notypes and stereotypes bias sex categorization.Journal of Personality and Social Psychology,102, 116–131.

Johnson, K. L., Gill, S., Reichman, V., & Tassi-nary, L. G. (2007). Swagger, sway, and sex-uality: judging sexual orientation from bodymotion and morphology. Journal of Personalityand Social Psychology, 93, 321–334.

Johnston, L., Arden, K., Macrae, C. N., & Grace,R. C. (2003). The need for speed: the menstrualcycle and person construal. Social Cognition,21, 89–99.

Kanwisher, N., McDermott, J., & Chun, M. M.(1997). The fusiform face area: a module inhuman extrastriate cortex specialized for faceperception. Journal of Neuroscience, 17(11),4302–4311.

Kaul, C., Rees, G., & Ishai, A. (2011). The genderof face stimuli is represented in multiple regionsin the human brain. Frontiers in Human Neuro-science, 4, 238.

Kenny, D. A. (1991). A general model of con-sensus and accuracy in interpersonal perception.Psychological Review, 98, 155–163.

Kluver, M., Hecht, H., & Troje, N. F. (2016). Inter-nal consistency predicts attractiveness in bio-logical motion walkers. Evolution and HumanBehavior, 37(1), 40–46.

Knutson, K. M., Mah, L., Manly, C. F., & Graf-man, J. (2007). Neural correlates of automaticbeliefs about gender and race. Human BrainMapping, 28(10), 915–930.

Ko, S. J., Judd, C. M., & Blair, I. V. (2006). Whatthe voice reveals: within- and between-categorystereotyping on the basis of voice. Personalityand Social Psychology Bulletin, 32, 806−819.

Kozlowski, L.T., & Cutting, J. E. (1977). Rec-ognizing the sex of a walker from a dynamic


� �

�

References 461

point-light display. Perception and Psycho-physics, 21, 575–580.

Kunda, Z., & Thagard, P. (1996). Forming impres-sions from stereotypes, traits, and behaviors: aparallel-constraint-satisfaction theory. Psycho-logical Review, 103, 284 –308.

Kurzban, R., Tooby, J., & Cosmides, L. (2001).Can race be erased? Coalitional computationand social categorization. Proceedings of theNational Academy of Sciences of the UnitedStates of America, 98(26), 15387–15392.

Langlois, J. H., Kalakanis, L., Rubenstein, A. J.,Larson, A., Hallamm, M., & Smoot, M. (2000).Maxims or myths of beauty? A meta-analyticand theoretical review. Psychological Bulletin,126, 390–423.

Langlois, J. H., & Roggman, L.A. (1990). Attrac-tive faces are only average. Psychological Sci-ence, 1, 115–121.

Latinus, M., & Belin, P. (2012). Perceptual audi-tory aftereffects on voice identification usingbrief vowel stimuli. PLoS ONE, 7, e41384.

Lefevre, C. E., Lewis, G. J., Perrett, D. I., & Penke,L. (2013). Telling facial metrics: facial width isassociated with testosterone levels in men. Evo-lution & Human Behavior, 34, 273–279.

Levin, D. T., & Banaji, M. R. (2006). Distortionsin the perceived lightness of faces: the role ofrace categories. Journal of Experimental Psy-chology: General, 135(4), 501–512.

Li, J., Liu, J., Liang, J., Zhang, H., Zhao, J., Huber,D. E.,…Shi, G. (2009). A distributed neuralsystem for top-down face processing. Neuro-science Letters, 451(1), 6–10.

Li, J., Liu, J., Liang, J., Zhang, H., Zhao, J.,Rieth, C. A.,…Lee, K. (2010). Effective con-nectivities of cortical regions for top-down faceprocessing: a dynamic causal modeling study.Brain Research, 1340, 40–51.

Little, A. C., Burt, D. M., Penton-Voak, I. S., &Parrett, D. I. (2001). Self-perceived attractive-ness influences human female preferences forsexual dimorphism and symmetry of male faces.Proceedings of the Royal Society of London B,268, 39–44.

Macchi, C. V., Turati, C., & Simion, F. (2004).Can a nonspecific bias toward top-heavy

patterns explain newborns’ face preference?Psychological Science, 15(6), 379–383.

MacDonald, A. W., Cohen, J. D., Stenger, V. A., &Carter, C. S. (2000). Dissociating the role ofthe dorsolateral prefrontal and anterior cingulatecortex in cognitive control. Science, 228(5472),1835–1838.

MacLin, O. H., & Malpass, R. S. (2001). Racialcategorization of faces: the ambiguous-race faceeffect. Psychology, Public Policy and Law, 7,98–118.

Maddox, K. B. (2004). Perspectives on racial phe-notypicality bias. Personality and Social Psy-chology Review, 8(4), 383–401.

Masuda, T., Ellsworth, P. C., Mesquita, B., Leu, J.,Tanida, S., & Van de Veerdonk, E. (2008). Plac-ing the face in context: cultural differences in theperception of facial emotion. Journal of Person-ality and Social Psychology, 94, 365–381.

Masuda, S., Tsujii, T., & Watanabe, S. (2005).An interference effect of voice presentation onface gender discrimination task: evidence fromevent-related potentials. International CongressSeries, 1278, 156−159.

McAleer, P., Todorov, A., & Belin, P. (2014).How do you say “hello”? Personality impres-sions from brief novel voices. PLoS ONE, 9(3):e90779.

McArthur, L. Z., & Baron, R. M. (1983). Towardan ecological theory of social perception. Psy-chological Review, 90, 215–238.

Meissner, C. A., & Brigham, J. C. (2001). Thirtyyears of investigating the own-race bias in mem-ory for faces: a meta-analytic review. Psychol-ogy, Public Policy, and Law, 7, 3–35.

Michel, C., Corneille, O., & Roisson, B. (2007).Race categorization modulated holistic face pro-cessing. Cognitive Science, 31, 911–924.

Michel, C., Rossion, B., Han, J., Chung, C., &Caldara, R. (2006). Holistic processing is finelytuned for faces of one’s own race. PsychologicalScience, 17, 608–615.

Milne, E., & Grafman, J. (2001). Ventromedialprefrontal cortex lesions in humans eliminateimplicit gender stereotyping. Journal of Neuro-science, 21(12): RC150.

Mitchell, J. P., Banaji, M. R., & Macrae, C. N.(2005). General and specific contributions of


� �

�


the medial prefrontal cortex to knowledge aboutmental states. NeuroImage, 28, 757–762.

Montepare, J. M., & Dobish, H. (2003). The con-tribution of emotion perceptions and their over-generalizations to trait impressions. Journal ofNonverbal Behavior, 27, 237–254.

Montepare, J. M., & Zebrowitz, L. A. (1993). Across-cultural comparison of impressions cre-ated by age-related variations in gait. Journal ofNonverbal Behavior, 17, 55–68.

Montepare, J. M., & Zebrowitz, L. A. (1998).“Person perception comes of age”: the salienceand significance of age in social judgments.In M. Zanna (Ed.), Advances in experimentalsocial psychology (Vol. 30, pp. 93–163). SanDiego, CA: Academic Press.

Nelson, C. A. (2001). The development and neu-ral bases of face recognition. Infant and ChildDevelopment, 10, 3–18.

Norman, K. A., Polyn, S. M., Detre, G. J., &Haxby, J. V. (2006). Beyond mind-reading:multi-voxel pattern analysis of fMRI data.Trends in Cognitive Sciences, 10, 424–430.

Oosterhof, N. N., & Todorov, A. (2008). The func-tional basis of face evaluation. Proceedings ofthe National Academy of Sciences of the UnitedStates of America, 105, 11087–11092.

Ostrom, T. M., & Sedikides, C. (1992). Out-grouphomogeneity effects in natural and mini-mal groups. Psychological Bulletin, 112(3),536–552.

Pauker, K., & Ambady, N. (2009). Multiracialfaces: how categorization affects memory at theboundaries of race. Journal of Social Issues, 65,69–86.

Pauker, K., Rule, N. O., & Ambady, N. (2010).Ambiguity and social perception. In E. Bal-cetis & D. Lassiter (Eds.), The social psychologyof visual perception (pp. 7–26). New York, NY:Psychology Press.

Peelen, M. V., & Downing, P. E. (2007). The neuralbasis of visual body perception. Nature ReviewsNeuroscience, 8, 636–648.

Pinel, E. C. (1999). Stigma consciousness: the psy-chological legacy of social stereotypes. Jour-nal of Personality and Social Psychology, 76(1),114–128.

Pitcher, D., Duchaine, B., & Walsh, V. (2014).Combined TMS and fMRI reveal dissocia-ble cortical pathways for dynamic and staticface perception. Current Biology, 24(17), 2066–2070.

Porter, S., ten Brinke, L., & Gustaw, C. (2010).Dangerous decisions: the impact of first impres-sions of trustworthiness on the evaluation oflegal evidence and defendant culpability. Psy-chology, Crime, and Law, 16, 477–491.

Price, C. J., & Humphreys, G. W. (1989). Theeffects of surface detail on object categorizationand naming. Quarterly Journal of ExperimentalPsychology, 41A, 797–828.

Pylyshyn, Z. (1999). Is vision continuous withcognition?: The case for cognitive impenetrabil-ity of visual perception. Behavioral and BrainSciences, 22(3), 341–365.

Ratner, K. G., Kaul, C., & Van Bavel, J. J. (2013).Is race erased? Decoding race from patterns ofneural activity when skin color is not diagnos-tic of group boundaries. Social Cognitive andAffective Neuroscience, 8, 750–755.

Re, D. E., & Rule, N. O. (2016). Predicting firmsuccess from the facial appearance of chiefexecutive officers of non-profit organizations.Perception, 45, 1137–1150.

Reed, C. L., Stone, V., Bozova, S., & Tanaka, J.(2003). The body inversion effect. Psychologi-cal Science, 14, 302–308.

Richardson, M. J., & Johnston, L. (2005). Personrecognition from dynamic events: the kinematicspecification of individual identity in walk-ing style. Journal of Nonverbal Behavior, 29,25–44.

Richeson, J. A., & Shelton, J. N. (2005). Thinslices of racial bias. Journal of NonverbalBehavior, 29, 211–236.

Righart, R., & de Gelder, B. (2008). Rapid influ-ence of emotional scenes on encoding of facialexpressions: an ERP study. Social Cognitiveand Affective Neuroscience, 3, 270–278.

Rotshtein, P., Henson, R. N. A., Treves, A., Driver,J., & Dolan, R. J. (2005). Morphing Marilyninto Maggie dissociates physical and identityface representations in the brain. Nature Neuro-science, 8, 107–113.


� �

�

References 463

Rule, N. O., & Ambady, N. (2008a). Brief expo-sures: male sexual orientation is accurately per-ceived at 50ms. Journal of Experimental SocialPsychology, 44, 1100–1105.

Rule, N. O., & Ambady, N. (2008b). The face ofsuccess: inferences from chief executive offi-cers’ appearance predict company profits. Psy-chological Science, 19, 109–111.

Rule, N. O., Ambady, N., Adams, R. B., & Macrae,C. N. (2008). Accuracy and awareness in theperception and categorization of male sexualorientation. Journal of Personality and SocialPsychology, 95, 1019–1028.

Runeson, S., & Frykholm, G. (1983). Kine-matic specification of dynamics as an informa-tional basis for person-and-action perception:expectation, gender recognition, and deceptiveintention. Journal of Experimental Psychology:General, 112, 585–615.

Said, C. P., Dotsch, R., & Todorov, A. (2010a).The amygdala and FFA track both social andnon-social face dimensions. Neuropsychologia.48(12), 3596–3605.

Said, C. P., Moore, C. D., Engell, A. D., Todorov,A., & Haxby, J. V. (2010b). Distributed rep-resentations of dynamic facial expressions inthe superior temporal sulcus. Journal of Vision,10(5), 11.

Savic, I., Berglund, H., Gulyas, B., & Roland, P.(2001). Smelling of odorous sex hormone–likecompounds causes sex-differentiated hypotha-lamic activations in humans. Neuron, 31,661–668.

Scheib, J. E., Gangestad, S. W., & Thornhill,R. (1999). Facial attractiveness, symmetry, andcues to food genes. Proceedings of the RoyalSociety of London B, 266, 1913–1917.

Scherer, K R. (2003). Vocal communicationof emotion: a review of research paradigms.Speech Communication, 40, 227–256.

Simion, F., & Di Giorgio, E. (2015). Face per-ception and processing in early infancy: inbornpredispositions and developmental changes.Frontiers in Psychology, 6(969).

Smith, E. L., Grabowecky, M., & Suzuki, S.(2007). Auditory–visual crossmodal integrationin perception of face gender. Current Biology,17, 1680−1685.

Sofer, C., Dotsch, R., Wigboldus, D. H. J., &Todorov, A. (2015). What is typical is good:the influence of face typicality on perceivedtrustworthiness. Psychological Science, 26(1),39–47.

Stirrat, M., & Perrett, D. I. (2010). Valid facialcues to cooperation and trust: male facial widthand trustworthiness. Psychological Science, 21,349–354.

Stolier, R. M., & Freeman, J. B. (2016). Neuralpattern similarity reveals the inherent intersec-tion of social categories. Nature Neuroscience,19, 795–797.

Summerfield, C., & Egner, T. (2009). Expectation(and attention) in visual cognition. Trends inCognitive Sciences, 13(9), 403–409.

Sutherland, C. A. M., Rowley, L. E., Amoaku,U. T., Daguzan, E., Kidd-Rossiter, K. A.,Maceviciute, U., & Young, A. W. (2015). Per-sonality judgments from everyday images offaces. Frontiers in Psychology, 6(1616), 1–11.

Tajfel, H. (1970). Experiments in intergroupdiscrimination. Scientific American, 223(5),96–102.

Tajfel, H., & Turner, J. C. (1979). An integrativetheory of intergroup conflict. In W. Austin &S. Worchel (Eds.), The social psychology ofintergroup relations (pp. 33–47). Monterey, CA:Brooks/Cole.

Thornhill, R., & Gangestad, S. W. (1993). Humanfacial beauty: averageness, symmetry, and para-site resistance. Human Nature, 4, 237–269.

Todorov, A. (2011). Evaluating faces on socialdimensions. In A. Todorov, S. T. Fiske, D. A.Prentice (Eds.), Social neuroscience: towardunderstanding the underpinnings of the socialmind (pp. 54–76). New York, NY: OxfordUniversity Press.

Todorov, A., & Engell, A. D. (2008). The role ofthe amygdala in implicit evaluation of emotion-ally neutral faces. Social Cognitive and AffectiveNeuroscience, 3(4), 303–312.

Todorov, A., Gobbini, M. I., Evans, K. K., &Haxby, J. V. (2007). Spontaneous retrieval ofaffective person knowledge in face perception.Neuropsychologia, 45, 163–173.

Troje, N. F. (2003). Gender and attractiveness frombiological motion. Journal of Vision, 3(9), 86a.


� �

�


Trope, Y. (1986). Identification and inferentialprocesses in dispositional attribution. Psycho-logical Review, 93, 239–257.

Uleman, J. S., Blader, S., & Todorov, A. (2005).Implicit impressions. In R. Hassin, J. S.Uleman, & J. A. Barge (Eds.), The new uncon-scious (pp. 362–392). New York, NY: OxfordUniversity Press.

Van Bavel, J. J., & Cunningham, W. A. (2009).Self-categorization with a novel mixed-racegroup moderates automatic social and racialbiases. Personality and Social PsychologyBulletin, 35, 321–335.

Van Bavel, J. J., Packer, D. J., & Cunningham,W. A. (2008). The neural substrates of in-groupbias: a functional magnetic resonance imaginginvestigation. Psychological Science, 19(11),1131–1139.

Van Bavel, J. J., Swencionis, J. O’Connor,R., & Cunningham, W. A. (2012). Motivatedsocial memory: belonging needs moderate theown-group bias in face recognition. Journal ofExperimental Social Psychology, 48, 707–713.

Van den Stock, J., Righart, R., & de Gelder, B.(2007). Body expressions influence recognitionof emotions in the face and voice. Emotion, 7,487–494.

Van Dommelen, W. A. (1993). Speaker height andweight identification: a re-evaluation of someold data. Journal of Phonetics, 21, 337–341.

Voss, A., Rothermund, K., & Brandtstadter, J.(2008). Interpreting ambiguous stimuli: separat-ing perceptual and judgmental biases. Journalof Experimental Social Psychology, 44, 1048–1056.

Walton, J. H., & Orlikoff, R. F. (1994). Speakerrace identification from acoustic cues in thevocal signal. Journal of Speech, Language, andHearing Research, 37(4), 738–745.

Watson, D. (1989). Strangers’ ratings of thefive robust personality factors: evidence of a

surprising convergence with self-report. Jour-nal of Personality and Social Psychology, 57,120–128.

Webster, M. A., Kaping, D., Mizokami, Y., &Duhamel, P. (2004). Adaptation to natural facecategories. Nature, 428, 557–561.

Weisbuch, M., & Ambady, N. (2011). Thin-slicevision. In R. B. Adams, N. Ambady, K.Nakayama, & S. Shimojo, The science of socialvision (pp. 228–247). New York, NY: OxfordUniversity Press.

Yin, R. K. (1969). Looking at upside-downfaces. Journal of Experimental Psychology, 81,141–145.

Zanto, T. P., Rubens, M. T., Thangavel, A., &Gazzaley, A. (2011). Causal role of the pre-frontal cortex in top-down modulation of visualprocessing and working memory. Nature Neu-roscience, 14(5), 656–661.

Zebrowitz, L. A., Fellous, J. M., Mignault, A., &Andreoletti, C. (2003). Trait impressions asovergeneralized responses to adaptively signifi-cant facial qualities: evidence from connection-ist modeling. Personality and Social PsychologyReview, 7, 194–215.

Zebrowitz, L. A., Luevano, V. X., Bronstad,P. M., & Aharon, I. (2009). Neural activationto babyfaced men matches activation to babies.Social Neuroscience, 4(1), 1–10.

Zebrowitz, L. A., & McDonald, S. M. (1991).The impact of litigants’ baby-facedness andattractiveness on adjudications in small claimscourts. Law and Human Behavior, 15(6),603–623.

Zebrowitz, L. A., & Montepare, J. M. (2006).The ecological approach to person percep-tion: evolutionary roots and contemporary off-shoots. In M. Schaller, J. A. Simpson, & D. T.Kenrick (Eds.), Evolution and social psychol-ogy (pp. 81–113). New York, NY: PsychologyPress.

Psychology and Neuroscience of Person...

Documents

Transcript of Psychology and Neuroscience of Person...