Developmental Psychology 1998, Vol 34, No. 3, 525-539 Copyright 199S by the American Psychological Association, Inc. 0012-1649/98/$3.00 Children's Use of Triadic Eye Gaze Information for "Mind Reading" Kang Lee, Michelle Eskritt, Lawrence A. Symons, and Darwin Muir Queen's University Five experiments examined children's use of eye gaze information for "mind-reading" purposes, specifically, for inferring another person's desire. When presented with static displays in the first 3 experiments, only by 4 years of age did children use another person's eye direction to infer desires, although younger children could identify the person's focus of attention. Further, 3-year-olds were capable of inferring desire from other nonverbal cues, such as pointing (Experiment 3). When eye gaze was presented dynamically with several other scaffolding cues (Experiment 4), 2- and 3-year- olds successfully used eye gaze for desire inference. Scaffolding cues were removed in Experiment 5, and 2- and 3-year-olds still performed above chance in using eye gaze. Results suggest that 2- year-olds are capable of using eye gaze alone to infer about another's desire. The authors propose that the acquisition of the ability to use attentional cues to infer another's mental state may involve both an association process and a differentiation process. A pair of eyes is a relatively simple stimulus that involves only two dark dots encircled by surrounding white areas. Yet, it conveys strong directional information that is unparalleled by any other dark-white contrasts (Cline, 1967; Gibson & Pick, 1963). Ethologists and other behavioral scientists have noted the important and unique role that eye gaze plays in inter- and intraspecies interactions. For many animals, eye gaze establishes dominance, initiates and terminates aggression and mating be- haviors, and sometimes indicates the location of food and signals the direction of an approaching predator {Argyle & Cook, 1976; Eibl-Eibesfeldt, 1989; Gomez, 1994b; Povinelli & Eddy, 1996a, 1996b; Rutter, 1984). Humans have extended the use of eye gaze to many other situations (for a review, see Kleinke, 1986, and Rutter, 1984). One of the main uses of eye gaze that appears to be unique to humans is to reveal another person's mental activities, or "mind- reading" (Baron-Cohen, 1994, 1995a). Eye gaze may be used to determine another individual's state of mind (e.g., focus of attention, knowledge, desire, and belief). Baron-Cohen (1995a, 1995b) theorized that the ability to use eye gaze is crucial to the development of a theory of mind, and that the lack of ihe sensitivity to eye gaze is related to impairments in social and Kang Lee, Michelle Eskritt, Lawrence A. Symons, and Darwin Muir, Department of Psychology, Queen's University, Kingston, Ontario, Canada. We gratefully acknowledge the contribution of the children who par- ticipated in the study and the children's parents and teachers for their kind cooperation. Thanks also to Julie Board and Lianne Fisher for comments on a draft of the article and to Krista Muir for her help in data collection. This research was supported by grants from the Social Sciences and Humanities Research Council (SSHRCC) of Canada, and the Natural Sciences and Engineering Research Council of Canada, and by a SSHRCC Postdoctoral Fellowship to Lawrence Symons. Correspondence concerning this article should be addressed to Kang Lee, Department of Psychology, Queen's University, Kingston, Ontario, Canada K7L 3N6. Electronic mail may be sent to kang@pavlov.psyc. queensuxa. cognitive abilities such as autism (Baron-Cohen, Campbell, Karmiloff-Smith, Grant, & Walker, 1995; Phillips, Baron-Co- hen, & Rutter, 1992). In the past two decades, it has been found that the develop- ment of theory of mind undergoes a dramatic shift around 4 years of age (Perner, 1992; Wellman, 1990). Four-year-olds rap- idly develop the ability to represent others' beliefs, and to under- stand representational change of their own beliefs, whereas 3- year-olds have difficulty with such concepts. This developmental pattern has been found in numerous studies (see Astington & Gopnik, 1991, for a review), although a few investigators have claimed that 3-year-olds have a fledgling understanding of oth- ers' beliefs (e.g., Chandler, Fritz, & Hala, 1989; Siegal & Beat- tie, 1991). Many researchers believe that the rapid development in chil- dren's understanding of belief and false beliefs around 4 years of age is an outcome of many earlier developments. Wellman (1990) suggested that one of the precursors to the development of the understanding of beliefs is the understanding of another's desire. Baron-Cohen (1994) added to Wellman's model a Shared Attention Mechanism (SAM) that enables children to infer others' desire through the use of eye gaze information. According to Baron-Cohen (1994, 1995a), SAM evolves from two mechanisms, an inlentionality detector and an eye direction detector, which both emerge during the first year of life. The Shared Attention Mechanism allows children to determine the object of a person's attention and, when combined with other information, why the person is attending to the particular object. The Shared Attention Mechanism forms the basis for the later development uf a theory-of-mind mechanism, the mechanism responsible for understanding beliefs (Leslie, 1994). To date, little empirical evidence exists to substantiate Baron- Cohen's claims regarding the SAM and young children's use of eye gaze information for "mind-reading" purposes. Never- theless, extensive studies have been conducted to examine the early development of children's sensitivity to eye gaze and their use of eye gaze information during face-lo-face interaction, joint attentional activity, and referential communication (for a review, 525