A Crosslinguistic Study of Spatial Categorization from the Preverbal State to Adulthood

San Diego State University Foundation, San Diego CA

Investigators

Abstract

Choi, Soonja BCS-0091493 The relation between semantics and cognition: a crosslinguistic study of spatial categorization from the preverbal stage to adulthood This project investigates the relation between language-specific word meanings and conceptual classifications of spatial relations in English and Korean, from infancy to adulthood. The two languages differ significantly in the way they divide up spatial relations for purposes of language. For example, whereas English has words such as IN and ON to distinguish between containment (e.g., putting an apple IN a bowl) and support (e.g., putting a cup ON a table), Korean does not have such terms. Instead, Korean has words that distinguish between loose fit (e.g., apple in a bowl) and tight fit (e.g., video cassette in its case; ring on a finger) regardless of containment and support. The research explores two interrelated questions that arise from this finding: (1) How do children learn such language-specific word meanings so early?, and (2) how do these early-acquired word meanings influence spatial concepts at later stages? These questions will be studied by investigating the ways in which children and adults distinguish various types of spatial relations. More specifically, the studies ask whether infants who do not yet talk initially begin learning language with a more extensive set of spatial categories than is needed for the language they are learning, and whether older children and adults maintain sensitivity to the same extensive set of spatial categories or, as preliminary work suggests, they channel their attention primarily to distinctions they learned in acquiring their first language. Preliminary experiments suggest that preverbal infants (9, 11, and 14 months) are able to distinguish a wide range of spatial features, including those not relevant for the language being learned, but that over time, the language channels their attention to a more limited set of linguistically relevant features. The proposed work examines this possibility further with a set of spatial features involving containment and support. The studies employ a preferential looking design that measures the amount of time the participant looks at a particular scene, given a choice between two scenes. This method can be used with both young children (starting from 9 months) and adults, thus allowing systematic comparisons of the two populations. In addition, adult participants are asked to perform a classification task which requires that they verbally explain their decisions. This research will provide insight into the mechanisms that enable children to acquire language-specific word-meaning systems at a very young age, and the kinds of interactions that take place between language and concept formation in infants, children and adults.