OBJECT FUNCTIONS AND NAMES IN INFANT CATEGORIZATION

Northwestern University, Evanston IL

Investigators

Abstract

Categorical representations are critically important for human action and cognition. Although infants have the highly adaptive ability to form categories (e.g. Quinn & Eimas, 1997; Mandler & McDonough, 1993), it is not yet clear how they succeed at doing so. Several researchers have proposed that there must be constraints on the kinds of information that infants attend to in developing categorical representations. Two particularly promising classes of constraints have been identified: object functions and object names. Recent theory and research suggest that object functions and names highlight commonalities among sets of objects so that infants respond to them categorically (Booth, 1998; Waxman & Markow, 1995). Unfortunately, research on object functions and naming has been conducted using very different methodology and stimuli. Thus, it is difficult to know whether the two constraints are equally effective under a variety of matched conditions. This proposal has two goals, the first of which is to evaluate whether the object function and name constraints operate effectively on the same stimuli, using the same method, in infants of the same age under a variety of conditions. The second goal is to explore interactions between the effects of object functions and names. These goals are pursued using an active familiarization and match to sample task. 14-month-old infants manipulate a series of objects from a single category that are presented with no identifying cues, novel names, novel functions or both. They are then asked to choose between a novel member of the familiarization category and an exemplar of a contrasting category to make a match to a familiar target object. If object functions and names do highlight categories, infants should choose the taxonomic matches more often when these cues are presented during familiarization than when they are not.