Prediction is believed to be an important component of cognition, particularly in natural language processing. It has long been accepted that recurrent neural networks are best able to learn prediction tasks when trained on simple examples before incrementally proceeding to more complex sentences. Furthermore, the counter-intuitive suggestion has been made that networks and, by implication, humans may be aided in learning by limited cognitive resources (Elman, 1993, Cognition). The current work reports evidence that starting with simplified inputs is not necessary in training recurrent networks to learn pseudo-natural languages; in fact, delayed introduction of complex examples is often an impediment. We suggest that the structure of natural language can be learned without special teaching methods or limited cognitive resources.

Recent Positron Emission Tomography (PET) and other studies have produced detailed information about the areas of the brain involved in word association tasks, their functional roles in learning word associations, and the changes in activity in these areas during learning. We present a dynamic neurond model that replicates observed human cognitive behavior in learning word as?sociations while satisfying salient neuroanatomical and neuropsychological constraints. The model captures the observed dynamics of cortico-thalamo-basal ganglionic loops.