International Centre for Language and Communicative Development: Using a Developmentally Realistic Model of Word Class Acquisition to Simulate Developmental Changes in the Noun-richness of Children's Early Language Across English, Dutch and German, 2014-2020

We examine the success of developmental distributional analysis in English, German and Dutch. We embed the mechanism for distributional analysis within an existing model of language acquisition (MOSAIC) that encodes increasingly long utterances, and compare results against a measure of ‘noun richness’ in child speech. We show that, cross-linguistically, the mechanism’s success in building an early noun class is inversely related to the complexity of the determiner and noun gender system, and that merging of determiners gives very similar results across languages. These results suggest that children may represent grammatical categories at multiple levels of abstraction that reflect both the larger category as well as its finer structure. We also examine how a mechanism that learns word classes from distributional information can contribute to the simulation of child language. Using a novel measure of noun richness, it is shown that the ratio of nouns to verbs in young children’s speech is considerably higher than in adult speech. Simulations with MOSAIC show that this effect can be partially (but not completely) explained by an utterance-final bias in learning. The remainder of the effect is explained by the early emergence of a productive noun category, which can be learned through distributional analysis.