A biological mapping of a learned avoidance behavior model to the basal ganglia

Abstract

In this paper we map a computational model of learned avoidance behavior in a one-way avoidance experiment to the biology of the basal ganglia. We extend our previous work to develop a more biologically accurate mapping. Learned avoidance behavior is a critical component of animal survival; thus, a model of animal learning should account for this phenomenon. Through long term potentiation and long term depression at the corticostriatal synapses, we propose that a prediction of the expected future benefit is generated by the animal. We map a reinforcement center of the model to the indirect pathway of the basal ganglia and a motor center to the direct pathway. Finally, we propose that an external reinforcement signal, in the form of pain caused by an electric shock, is transferred from the thalamus to the subthalamic nucleus.