Sharp, Paul B., Raymond J. Dolan, and Eran Eldar. 2020. “Cognitive Map Learning Is Disrupted in Compulsivity and Anxious Arousal.” PsyArXiv. June 14. doi:10.31234/osf.io/x29jq
Abstract: Both anxiety and compulsivity have been linked to a belief that the environment is uncontrollable. We lack a mechanistic explanation regarding how such a belief might form. Here, we propose a novel hypothesis that individuals with anxiety and compulsivity have difficulty in building probabilistic internal maps between actions and environmental states, specifically state transition learning. To address this hypothesis, we recruited a large (n=174) online sample of individuals exhibiting a range of psychopathology. We created a novel one-step revaluation task that isolated state transition learning from other processes with which it is typically confounded in computational psychiatry investigations. Results from the one-step revaluation paradigm demonstrated that both compulsivity and anxious arousal are associated with a basic disruption in state transition learning. To strengthen mechanistic inferences from this online study we developed a computational model of state transition learning and tested its validity by fitting it to a separate publicly available (n=1413) sequential decision-making data. In this independent dataset we show that compulsivity-related disruptions in state transition learning arise out of overly fast updating of state transitions estimates. We suggest that disrupted state transition learning is a promising computational phenotype that may shed light on the genesis of these pathologies as well as provide a potential target for prevention and intervention. Future work can determine if a deficit in transition learning represents a cause, consequence, or maintenance factor in compulsivity and anxious arousal.
Sunday, August 16, 2020
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment