A Spiking Neuron Model of the Basal Ganglia Circuitry that Can Generate Behavioral Variability

Abstract

The cortical-basal ganglia circuitry has been implicated in action selection, action initiation, and generation of behavioral variability. However, underlying mechanisms for these functions still remain unresolved. In this paper, we propose a new spiking neuron model for the basal ganglia circuitry that includes different functions for the direct and indirect pathways: the indirect pathway selects an action to be executed and then the direct pathway determines the timing of the selected action. Computer simulations demonstrate that the basal ganglia circuitry supports action selection and self-timed initiation of actions. The circuitry can also generate trial-by-trial variations in the selection and timing of actions for exploration. Finally, these variations can optimally and independently be tuned by the strength of corticostriatal synapses for exploitation.