A Horse of a Different Color: Specifying With Precision Infants’ Mappings of Novel Nouns and Adjectives

Abstract

A precisely controlled automated procedure confirms a developmental decalage: infants acquiring English link count nouns to object categories well before they link adjectives to properties. Fourteen-and 18-month-olds (n = 48 at each age) extended novel words presented as count nouns based on category membership, rather than shared properties. When the same words were presented as adjectives, infants revealed no preference for either category-or property-based extensions. The convergence between performance in this automated procedure and in more interactive tasks is striking. Perhaps more importantly, the automated task provides a methodological foundation for (1) exploring the development of form-meaning links in infants acquiring languages other than English, and (2) investigating the time-course underlying infants' mapping of novel words to meaning.Infants' first words are greeted with special joy, perhaps because we share with Confucious the intuition that "The beginning of wisdom is to call things by their right name." This, after all, is no simple matter. Many different words can be applied to the same scene, and different kinds of words (e.g., nouns, adjectives, verbs) highlight different aspects of that scene (e.g., object categories, object properties, and events). Successful word-learning therefore requires identification of a novel words' grammatical form, discovery of its appropriate referent, and its appropriate extension to new referents.Some propose that an early link between nouns and object categories provides the foundation for the acquisition of other links between grammatical forms and meanings (Dixon, 1982;Gentner, 1982;Gleitman, 1990;Huttenlocher & Smiley, 1987;Maratsos, 1998;Talmy, 1985;Waxman, 1999a; Waxman & Lidz, 2006;Wierzbicka, 1986). Because adjectives regularly derive meaning from the nouns they modify (e.g., a hard mattress vs. a hard test), and because verbs derive meaning from the relations among nouns (e.g., A chases B vs. B chases A), predicates (both adjectives and verbs) are interpreted in conjunction with accompanying nouns (Dixon, 1982).There is now considerable empirical support for this theoretical perspective (e.g. Childers & Tomasello, 2006;Echols & Marti, 2004;Gasser & Smith, 1998;Gentner, 1982;Hall & Moore, 1997;Imai, Haryu, & Okada, 2005;Smith, Jones, & Landau, 1992;Waxman & Booth, 2001; Waxman, Lidz, Braun, & Lavin, under review). Take for example, Booth and Waxman (2003). In this study, 14-month-old infants were introduced to four toy objects, all from the same object category and embodying the same property (e.g., purple horses). The experimenter labeled these objects either with novel nouns (e.g., "These are blickets") or adjectives ("These are blickish"). Next, in a contrast phase, infants viewed an object from a different category and embodying a different property (e.g., an orange carrot experimenter noted that this contrast object was "..not a blicket" (noun condition) or " …not a blickish one" (adjective condition). Finally, the experimenter introdu...

Linguistic Indicators of a False Confession

“…Nouns, verbs and adjectives (in that order) represent the three largest word classes in English by type frequency, while verbs are the largest class by token frequency. Moreover, these three categories relate to each other in a notable way: adjectives and verbs often obtain their meaning from the nouns they modify or accompany (e.g., 'a difficult journey' versus 'a difficult person' and 'Jack follows Jill' versus 'Jill follows Jack) (Booth & Waxman, 2009). Nouns are the first grammatical category to be acquired by Englishspeaking children (Bornstein et al, 2004), and they are often the first to be compromised in age-related cognitive decline (Ahmed, Arnold, Thompson, Graham, & Hodges, 2008).…”

Young Children's Use of Contrast in Word Learning: The Case of Proper Names

“…In this study, we focused on children's use of information about contrastive exemplars: indications that a label for one object does not apply to another object (e.g., Booth & Waxman, 2009;Clark, 1987Clark, , 1988Waxman et al, 2009; see also Au & Laframboise, 1990). The role of this type of information in lexical and conceptual development is less well understood (Namy & Clepper, 2010), but its potential relevance for learning proper names is especially clear.…”

Fast mapping from argument structure alone

“…Based on the time-course of children's response to novel nouns (Booth & Waxman, 2009), we expected their response to become evident 2 to 3 sec after the onset of the novel verb at test.…”