The role of dorsomedial striatum adenosine 2A receptors in the loss of goal-directed behaviour

Rationale Adenosine A(2A) receptors (A(2A)R) in the dorsal striatum have been implicated in goal-directed behaviour. While activation of these receptors with several methods has resulted in an insensitivity to outcome devaluation, particular explanations for how they disrupt behaviour have not been explored. We both confirm a role for A(2A) receptors in goal-directed responding and evaluate additional behavioural aspects of goal-directed control to more fully understand the role of A(2A) receptors in instrumental behaviour. Objectives To examine the effects of the adenosine A(2A) agonist CGS-21680 in the DMS on response-outcome encoding, updating representations of outcome value and on the ability to inhibit behaviour when reward is not available. Methods Male rats were trained to lever press for food reward. The A(2A)R agonist CGS-21680 was infused into the dorsomedial striatum either before an outcome devaluation test, prior to training with two distinct response-outcome associations or prior to a test of discriminative stimulus control over instrumental performance. Results Intra-DMS administration of CGS-21680 impaired sensitivity to outcome devaluation. CGS-21680 treatment did not impair acquisition of specific response-outcome associations, selective control of responding based on the presence of stimuli that signaled when reward was or was not available, discrimination between stimuli or lever choices nor did it influence the effect of devaluation on the amounts of food eaten in a consumption test. Conclusions CGS-21680 impairs the ability to modulate responding based on recent changes to outcome value, an effect that is not accounted for by impairments in behavioural inhibition, discrimination, encoding the specific outcome of a response or the effectiveness of specific satiety.

Automated summary

This study confirms the role of A(2A) receptors in goal-directed responding, and investigates their effects on response-outcome encoding, updating representations of outcome value, and the ability to inhibit behavior when reward is not available.