A neural model of adaptive behavior

A neurally plausible model of adaptive behavior is developed in the Boolean domain that is simple enough to be formally approached, but general enough to address a number of interesting issues.

Abstract

A neurally plausible model of adaptive behavior is developed in the Boolean domain. Behavior is modeled as the application of specific operators in response to patterns of internal and external features. Behavioral completeness requires that any stimulus (feature pattern) potentially be able to trigger any response (set of operators). Learning completeness requires that any stimulus-response mapping be learnable. Goal based behavior evaluation determines what the appropriate mapping is. The resulting problem is simple enough to be formally approached, but general enough to address a number of interesting issues. Appropriate structures and learning processes at the neuron and assembly level are developed. The problems of learning speed, noise rejection and shared memory are addressed in some detail.