ABSTRACT
This chapter aims to integrate connectionist developmental research related to language. It presents CogSci 2002 a variety of new models in connectionist language acquisition, including an SRN model of generalization, a self-organizing model of categorical representation, and an encoder network of concept acquisition. The chapter examines the emergence of categorical representation from a developmental connectionist perspective. It argues that localized linguistic representations arise as a function of the brain’s organization and reorganization in response to characteristics of the environment in learning and development. A self-organizing neural network is used to explore the high-dimensional space of various linguistic categories, analyzing realistic natural language data. Although neural network models provide an adequate account of fuzzy concepts, they are, according to some, incapable of accounting for the acquisition and representation of crisp concepts as in, e.g., kinship terms.
