Abstract
In this poster, four connectionist models for word learning are implemented. While the first three models simulate the associationistic, constraints/principles and social-pragmatic views, respectively, the final model simulates the interactionist perspective. From this perspective, each of the prior theories represents a different level of implementation: mechanism, neurological structure, or environmental context. An empirical study is reported that demonstrates the validity of this proposition. It is concluded that only by combining approaches is a complete explanation for word learning possible.