Semantic reorganization: does language influence infants' perception of components of events?

Journal Title
Journal ISSN
Volume Title
University of Delaware
Infants appear to first discriminate and categorize a possibly universal set of event components like path or the trajectory of an action or figure , the entity carrying out the action. Only after being exposed to the ambient language will they privilege components such as these that are expressed in their native language. This process of semantic reorganization may be one of the mechanisms that underlie the acquisition of relational terms (e.g., verbs and prepositions). Studies support the notion of semantic reorganization (e.g., Göksun et al., 2010). In Japanese, two different verbs are necessary to describe a person crossing a bounded ground (e.g., street) versus an unbounded ground (e.g., field) while in English, the same verb - crossing - describes both types of grounds. Both Japanese and English 13-to 15-month-old infants detect Japanese ground distinctions while at 18-to 20-months, only Japanese-reared infants maintain sensitivity to this distinction (Göksun et al., 2011). While language is hypothesized to guide infants' progression from language-general to language-specific event perception, no prior studies have examined this hypothesis. To pursue this question, the present study investigated whether 13-to 15-month-old English-learning infants (Experiment 1a) and 21-to 24-month-old children (Experiment 2a) showed sensitivity to Japanese ground-path distinctions in the presence of general language. To examine whether language plays a role in English-reared children's weakened sensitivity to non-native event components, Experiment 1b assessed whether language experience of a specific type weakens 13-to 15-month-old infants' sensitivity to Japanese ground-path distinctions. To explore the plasticity of semantic reorganization, Experiment 2b investigated whether children's sensitivity can be heightened when novel prepositions presented in sentences attracted their attention to ground-path distinctions. In the presence of general language, Experiment 1a found that at 13 to 15 months of age, English-reared infants showed sensitivity to Japanese ground-path. Experiment 2a demonstrated that 21-to 24-month-old English-reared children no longer showed sensitivity to this event component, corroborating the findings of Göksun et al. (2011). Experiment 1b showed that when a novel word is paired with bounded and unbounded grounds, 13-to 15-month-old infants' sensitivity to Japanese ground-path becomes dampened, suggesting that language may be the driving force behind children's language-specific event perception. When novel words were paired uniquely with bounded and unbounded grounds, 21-to 24-month-old discriminated between Japanese ground-path categories (Experiment 2b), suggesting that semantic reorganization is malleable. This is one of the first studies to show that language can be used to heighten and dampen children's sensitivity to "non-native" event components such as ground type (Casasola et al., 2009; Hespos & Spelke, 2004). Investigating the mechanism underlying semantic reorganization may further our understanding of how children learn to talk about events in their native language and illuminates an ancient debate about the role of language in thought.