Goal-directed behaviour and instrumental devaluation: a neural system-level computational model

Devaluation is the key experimental paradigm used to demonstrate the presence ofinstrumental behaviors guided by goals in mammals. We propose a neural system-levelcomputational model to address the question of which brain mechanisms allow thecurrent value of rewards to control instrumental actions. The model pivots on and showsthe computational soundness of the hypothesis for which the internal representationof instrumental manipulanda (e.g., levers) activate the representation of rewards (or"action-outcomes", e.g., foods) while attributing to them a value which depends onthe current internal state of the animal (e.g., satiation for some but not all foods).The model also proposes an initial hypothesis of the integrated system of key braincomponents supporting this process and allowing the recalled outcomes to bias actionselection: (a) the sub-system formed by the basolateral amygdala and insular cortexacquiring the manipulanda-outcomes associations and attributing the current value to theoutcomes; (b) three basal ganglia-cortical loops selecting respectively goals, associativesensory representations, and actions; (c) the cortico-cortical and striato-nigro-striatalneural pathways supporting the selection, and selection learning, of actions based onhabits and goals. The model reproduces and explains the results of several devaluationexperiments carried out with control rats and rats with pre- and post-training lesionsof the basolateral amygdala, the nucleus accumbens core, the prelimbic cortex, andthe dorso-medial striatum. The results support the soundness of the hypotheses of themodel and show its capacity to integrate, at the system-level, the operations of the keybrain structures underlying devaluation. Based on its hypotheses and predictions, themodel also represents an operational framework to support the design and analysis ofnew experiments on the motivational aspects of goal-directed behavior.