Comparing Computational Cognitive Models of Generalization in a Language Acquisition Task