Tell-Tale Eyes: Children’s Attribution of Gaze Aversion as a Lying Cue Shiri Einav University of Oxford Bruce M. Hood University of Bristol This study examined whether the well-documented adult tendency to perceive gaze aversion as a lying cue is also evident in children. In Experiment 1, 6-year-olds, 9-year-olds, and adults were shown video vignettes of speakers who either maintained or avoided eye contact while answering an interviewer’s questions. Participants evaluated whether the speaker was telling the truth or lying on each trial. The results revealed that at both ages, children were more likely to attribute lying to speakers in the gaze aversion condition; however, the effect was significantly greater among 9-year-olds. Significant gender differences were also uncovered, with girls demonstrating strongest sensitivity to the gaze cue. Experiment 2 replicated the gender effect in 6-year-olds but found that when the speakers’ verbal responses were removed, boys’ use of the gaze cue increased and the gender difference disappeared. These findings indicate that at 6 years old, children interpret interpersonal gaze behavior as a socially informative cue. Furthermore, the growing appreciation of the stereotypic gaze behavior associated with lying and the reputed female advantage in gaze sensitivity may reflect differential processing of multimodal communication. Keywords: gaze perception, mental state attribution, nonverbal deception cues, social cognition, sex differences As adults, we regularly attend to people’s gaze behavior to obtain information about their mental states. Indeed, visual scan paths have revealed that when individuals observe a face or a social scene, the eyes represent the primary focus of their attention (Klin, Jones, Schultz, Volkmar & Cohen, 2002; Yarbus, 1967). The mentalistic inferences that adults readily draw from the eyes can be specific—for example, identifying what someone is refer- ring to or feeling at a given moment (Baron-Cohen, 1995; Baron- Cohen, Wheelwright & Jolliffe, 1997)— but have also been found to influence more global attributions about the nature of a social relationship, or core aspects of people’s personality, such as their credibility, dominance, confidence, and competence levels (for a review, see Argyle & Cook, 1976; Kleinke, 1986). There is a wealth of empirical evidence to suggest that sensi- tivity to gaze is an early emerging capacity. From birth, infants demonstrate awareness of the eyes, showing visual preference for open rather than closed eyes (Batki, Baron-Cohen, Wheelright, Connellan, & Ahluwalia, 2000) and for direct rather than averted gaze (Farroni, Csibra, Simion, & Johnson, 2002). Furthermore, 2-month-old infants, like adults, spend longer looking at the eyes than at any other internal facial feature (Maurer & Salapatek, 1976). By 2 years of age, infants can use gaze direction to relate an adult’s emotional display to a specific referent and to relate a novel verbal label to the target of the speaker’s gaze (Baldwin & Moses, 1994; Phillips, Wellman, & Spelke, 2002; Repacholi, 1998; Sodian & Theormer, 2004; Woodward, 2003). By the time they reach their 3rd year, children begin using gaze direction cues for making explicit judgments about people’s mental states, including focus of attention (Doherty & Anderson, 2000), desires (Baron-Cohen, Campbell, Karmiloff-Smith, Grant, & Walker, 1995; Lee, Eskritt, Symons, & Muir, 1998), and knowledge (Pratt & Bryant, 1990). In addition, 4- and 5-year-olds are able to make use of temporal aspects of gaze such as duration and frequency for inferring goal and preference (Einav & Hood, 2006; Montgomery, Bach, & Moran, 1998). So far, research has concentrated on children’s early use of object-directed gaze behavior for judging another’s mentalistic relation toward physical objects, but comparatively little is known about children’s use of interpersonal gaze behavior when making social appraisals. This is an essential issue to address given that the gaze channel plays such a central role in face-to-face social inter- action. The levels of eye contact that individuals display toward one another as they interact can reveal a lot about the relation that exists between them (e.g., distant, friendly, or hostile; for a review, see Kleinke, 1986). Being able to interpret such cues involves more than simply identifying an agent’s primary visual target—the typical required response in previous gaze-reading studies. Rather, the child needs to be sensitive to the quality of gaze shown by one person to another and to recognize its mentalistic significance. Very young infants can discriminate between direct eye contact and averted gaze (Farroni et al., 2002; Vecera & Johnson, 1995), Shiri Einav, Department of Experimental Psychology, University of Oxford, Oxford, England; Bruce M. Hood, Department of Experimental Psychology, University of Bristol, Bristol, England. This research is based on part of Shiri Einav’s doctoral dissertation and was supported by an Economic and Social Research Council Postgraduate Studentship and Medical Research Council of Great Britain Grant G0700419 to Bruce M. Hood. Grateful acknowledgment is extended to parents and children who gave their consent to participate in these experiments and to the kind cooperation of local schools. We thank Melanie Grady, Sharon Enav, Daniel Bruneau, and David Gould for their invaluable assistance as our video protagonists. We also thank Peter Mitchell and Elizabeth Sheppard for helpful comments on the manuscript. Correspondence concerning this article should be addressed to Shiri Einav, Department of Experimental Psychology, University of Oxford, South Parks Road, Oxford OX1 3UD, England. E-mail: shiri.einav@psy .ox.ac.uk Developmental Psychology Copyright 2008 by the American Psychological Association 2008, Vol. 44, No. 6, 1655–1667 0012-1649/08/$12.00 DOI: 10.1037/a0013299 1655