The hierarchical organisation of cortical and basal-ganglia systems: A computationally-informed review and integrated hypothesis

Abstract

To suitably adapt to the challenges posed by reproduction and survival, animals need to learn to select when to perform different behaviours, to have internal criteria for guiding these learning processes, and to perform behaviours efficiently once selected. To implement these processes, their brains must be organised in a suitable hierarchical fashion. Here we briefly review two types of neural/behavioural/computational literatures, focussed, respectively, on cortex and on sub-cortical areas, and highlight their important limitations. Then we review two computational modelling works of the authors that exemplify the problems, brain areas, experiments, main concepts, and limitations of the two research threads. Finally we propose a theoretical integration of the two views, showing how this allows us to solve most of the problems found by the two accounts if taken in isolation. The overall picture that emerges is that the cortical and the basal ganglia systems form two highly-organised hierarchical systems working in close synergy and jointly solving all the challenges of choice, selection, and implementation needed to acquire and express adaptive behaviour.