Item

Abstract

Background: Dopamine is implicated in signalling modelfree (MF) reward prediction errors and various aspects of model-based (MB) planning and choice. Recently, we showed that cooperative interactions between MB and MF systems include guidance of MF credit assignment by MB inference. Here, we test a novel hypothesis that enhancing dopamine levels boosts the guidance of MF credit assignment via MB inference. Methods: Within-subject data from n¼62 healthy individuals was collected in a double-blinded, placebo-controlled design using levodopa with a task that allows a separate measurement of MB and MF contributions to choice, and specifically probes guidance of MF learning based on MB knowledge of the task’s transition structure. Data was analysed using mixed effects and computational reinforcement learning models. Results: We found that levodopa enhanced retrospective guidance of MF credit assignment via MB inference (p¼.01), without impacting on MF and MB contributions to choice per se. This drug effect positively correlated with working memory (r¼.28, p¼.03), but only in a context where reward needed to be recalled for MF credit assignment. The dopaminergic enhancement in MB-MF interactions was negatively correlated with dopamine-dependent change in MB choice (r¼-.35, p¼.04), possibly reflecting a trade-off between these two components of behavioural control. Conclusions: Our findings demonstrate that dopamine boosts MB inference during guidance of MF learning, supported in part by working memory but in trade-off with a dopaminergic enhancement of MB choice. The findings highlight a novel role for a dopaminergic influence on MB-MF interactions.