Diffusion modeling reveals reinforcement learning impairments in gambling disorder that are linked to attenuated ventromedial prefrontal cortex value representations.

Abstract

Gambling disorder is associated with deficits in feedback-based learning tasks, but the computational mechanisms underlying these learning impairments are still poorly understood. Here, we examined this question using a combination of computational modeling and functional resonance imaging (fMRI) in individuals that regular participate in gambling (n=23, seven fulfilled 1-3 DSM 5 criteria, sixteen fulfilled 4 or more) and matched controls (n=23). Participants performed a stationary reinforcement learning task with two pairs of stimuli (80% vs. 20% reinforcement rates per pair). As predicted, the gambling group made significantly fewer selections of the optimal stimulus, while overall response times (RTs) were not significantly different between groups. We then used comprehensive modeling using reinforcement learning drift diffusion models (RLDDMs) in combination with hierarchical Bayesian parameter estimation to shed light on the computational underpinnings of this performance impairment. In both groups, an RLDDM in which both non-decision time and response threshold (boundary separation) changed over the course of the experiment accounted for the data best. The model showed good parameter and model recovery, and posterior predictive checks revealed that in both groups, the model reproduced the evolution of both accuracy and RTs over time. The learning impairment in the gambling group was attributable to a more rapid reduction in boundary separation over time, and a reduced effect of value-differences on the drift rate, compared to controls. The gambling group also exhibted substantially shorter non-decision times. Imaging analyses replicated earlier effects of prediction error coding in the ventral striatum and value coding in the ventro-medial prefrontal cortex, but there was no credible evidence for group differences in these regions. Taken together, our findings highlight the computational mechanisms underlying reinforcement learning impairments in gambling disorder.