This proposal addresses the general problem of how children learn words. The objective is to examine in greater detail the interactive relationship between linguistic input (object names) and nonlinguistic cognition (prelinguistic categories). The issue is whether a child's hearing or using words serves to stimulate and/or shape nonlinguistic categories, or both. This proposal focuses on the influence of linguistic input on nonlinguistic categorization. Evidence regarding the relationship between linguistic input and nonlinguistic categorization has relied primarily on linguistic data from already verbal children. This reliance has restricted our understanding of this relationship in two ways. First, there is little empirical evidence that directly documents changes in nonlinguistic categorization as a function of linguistic input. Moreover, current data from already verbal children do not address possible developmental change in the relationship. Second, explanations of why language influences categorization assume the use of uniquely linguistic information by the child. However, early in development, more general cognitive factors may be important. The required studies using nonlinguistic inputs (e.g., music) have not been done. Moreover, it is generally held that language acquisition is a highly buffered, redundantly cued process. Thus, research programs aimed at identifying a coalition of variables and describing their interactive influences on categorization are needed.
The specific aims of this proposal are to examine the interactive influences of a complex of input-related (attention, labeling, fixation- input correlation, meaningfulness, and acoustic factors) and non-input related (age, category, typicality) variables on categorization. Input- related variables will be examined using both linguistic and nonlinguistic input. The habituation-discrimination paradigm will be used to directly measure nonlinguistic categorization. Experiments will focus on infants 7- 15 months of age. Increased understanding of the variable facilitating categorization and word learning will provide a necessary reference point from which to study language outcome in high-risk infants. Such studies, in turn, could considerably enhance our ability to identify very early those infants at risk for language impairment.
Roberts, K (1997) A preliminary account of the effect of otitis media on 15-month-olds' categorization and some implications for early language learning. J Speech Lang Hear Res 40:508-18 |