Modeling cortico-subcortical interactions during planning, learning, and voluntary control of actions

Recent computer modeling studies have made substantial progress toward specifying how adaptive cortical circuits involved in planning and voluntary control of actions may interact with adaptive sub-cortical circuits, notably those in the basal ganglia and the cerebellum. A key aspect of any comprehensive model is the proposed division of labor among frontal cortex, basal ganglia, and the cerebellum, all of which may be significantly engaged during most voluntary planning and action. This paper presents an emerging theoretical synthesis based on an interlocking set of formal computational hypotheses, which specify how local circuit features in the three areas interact, via neuro-anatomically established pathways within and between the areas, to solve fundamental learning and performance problems encountered during voluntary planning and action. Results of simulations that demonstrate the mutual coherence and explanatory power of the hypotheses, vis-a-vis extensive behavioral and neurophysiological data, were reviewed.