A Self-Organizing Neural Network Model of the Acquisition of Word Meaning

A central debate in the domain of language acquisition is how children acquire the

meanings of words. Although numerous studies have addressed this question in the last

few decades, researchers have not yet reached any consensus. One important line of

disagreement is whether children can use contextual or structural knowledge from the

sentence to bootstrap their learning of the semantic contents of words. Proponents of the

syntactic bootstrapping hypothesis (e.g., Gleitman, 1990) argue that children can and do

make use of the structural information to learn word meaning, whereas advocates of the

semantic bootstrapping hypothesis (e.g., Pinker, 1994) are suspicious of such an approach

that relies on the child’s distributional analysis of the input.