BUCLD 37 Proceedings To be published in 2013 by Cascadilla Press Rights forms signed by all authors Word-Mapping in Autism: Evidence for Backwards Bootstrapping of Social Gaze Strategies Kristina E. Patrick, Felicia Hurewitz, and Amy E. Booth * Although much is known about how typically developing (TD) children first learn words, understanding of the word-learning strategies used by children diagnosed with autism spectrum disorders (ASD) is still evolving. Language abilities are often delayed in children with ASD, but most do develop some functional communication skills (NIH, 2010). Children with ASD use many of the same language strategies as TD children, but they do so later in development. For instance, Swensen, Kelley, Fein, and Naigles (2007) found that children with ASD understood SVO word order before they were able to produce it and demonstrated a noun bias (preferentially attaching labels to nouns) in a manner similar to language acquisition processes of TD children who were several months older. Although there is some evidence for similar, yet delayed, language acquisition processes in children with ASD, it remains unclear whether they learn early words using some of the same strategies or entirely different strategies from those of TD children. More specifically, questions remain regarding the influence of core ASD-related socio-pragmatic deficits on children’s word-learning strategies and skills. Social understanding appears to play a critical role in supporting the early accumulation of vocabulary by TD children. Numerous studies have indicated that children as young as 17 months of age follow the gaze of a speaker to learn new words (Baldwin, 1991; Baldwin, 1993a; Baldwin, 1993b). Although simple associative processes surely contribute to infants’ success at mapping words to their referents based on direction of eye gaze, evidence suggests that infants are also keenly attuned to speakers’ communicative goals in these contexts (Woodward, 2003). In the absence of clear social indicators of a speaker’s referential intent, TD children do not pair novel labels and objects (Baldwin et al., 1996; Waxman & Gelman, 2009). Indeed, gaze following appears to be a core strategy used to infer speaker intention. * Kristina E. Patrick, Drexel University, kep66@drexel.edu. Felicia Hurewitz, Drexel University, fdh23@drexel.edu. Amy E. Booth, Northwestern University, a- booth@northwestern.edu. We would like to thank our lab managers, Ruth Zajicek and Kathy McGroarty-Torres; our research assistants, Amy Krischer, Jennifer Mongiat, Carolyn Moy, Vanessa Dopker, and Lauren Erikson; Michael Pesch for programming support; and the children and their families who participated in this study.