Proceedings of the Annual Meeting of the Cognitive Science Society

In this paper we examine how a mechanism that learns wordclasses from distributional information can contribute to thesimulation of child language. Using a novel measure of nounrichness, it is shown that the ratio of nouns to verbs in youngchildren’s speech is considerably higher than in adult speech.Simulations with MOSAIC show that this effect can bepartially (but not completely) explained by an utterance-finalbias in learning. The remainder of the effect is explained bythe early emergence of a productive noun category, which canbe learned through distributional analysis.