BUCLD 35 Proceedings To be published in 2011 by Cascadilla Press Rights forms signed by all authors Verb Learning from Syntax Alone at 21 Months Sudha Arunachalam, Emily Escovar, Melissa Hansen, and Sandra R. Waxman * To discover the meaning of a novel verb, toddlers are able to glean important information from the syntactic context in which it appears, including the number of participants expected to be involved in the event, and broad aspects of the relationship in which they will stand to each other (e.g., Fisher, 1996; Gleitman, 1990; Naigles, 1990). But verbs are often uttered in the absence of the events they describe (Tomasello & Kruger, 1992). To learn a new verb from such an encounter, when no visual scene is present, toddlers must be able to posit an initial representation for the verb from its linguistic context alone. Recent evidence indicates that even if toddlers hear a novel verb in the absence of any accompanying event, syntactic information does support their acquisition of its meaning (Arunachalam & Waxman, 2010; Yuan & Fisher, 2009). For example, Arunachalam & Waxman introduced 27-month-olds to dialogues incorporating novel verbs either in transitive sentences (e.g., John mooped the doggie) or intransitive sentences (e.g., John and the doggie mooped), but absent any relevant referent scene. Next, toddlers viewed two candidate scenes: a) two participants performing synchronous actions (e.g., boy and girl each wave one arm), and b) two participants performing a causative action (e.g., boy spins girl). When asked to point to “mooping,” toddlers who had heard transitive sentences chose the causative scene; those who had heard intransitive sentences did not. Here, we advance this work in two ways. First, we ask whether toddlers as young as 21 months can also use syntactic context to assign verb meaning. We chose this age group because previous evidence suggests that toddlers under two years of age are sensitive to word order cues (Gertner et al., 2006), but not argument structure (Hirsh-Pasek & Golinkoff, 1996) when learning a novel verb’s meaning. Second, we examine the time-course underlying toddlers’ resolution of novel verb meaning. We measure toddlers’ eye gaze as they hear the novel verbs during the test phase, and ask how long they require to shift their attention to the appropriate scene. NB: Although the time-course for lexical processing of familiar words has been well-studied in toddlers (e.g., Fernald et al., 2006), there is scant evidence on processing of newly-learned words. * Northwestern University. Corresponding author E-mail address: s-arunachalam@northwestern.edu (S. Arunachalam)