关键词: 快感缺失, 亚临床高危群体, 心理干预, 强化学习模型, 控制感

Abstract: Anhedonia, characterized by marked diminishment or absence of pleasure, arises from impairments in one or more components of reward processing. Although serving as both a prodromal marker and risk factor for multiple psychiatric disorders, current pharmacological and psychological interventions demonstrate limited efficacy in ameliorating anhedonia, necessitating novel therapeutic paradigms. Recent studies revealed that reward learning constituted the critical nexus between reward anticipation and consummatory response, with learning occurring exclusively when individuals perceive behavioral agency over reward outcomes. Therefore, this study employs a multimodal approach to investigate this phenomenon. Specifically, it aims to 1) validate the phenomenon of perceived control dynamically modulating anhedonic reward learning through experimental paradigms and reinforcement learning modeling, 2) elucidate its neurocomputational mechanisms using functional magnetic resonance imaging (fMRI), and 3) conduct randomized controlled trials to empirically test the intervention efficacy of perceived control. The present study postulates the Dual-Layer Reward Learning Hypothesis with Internal-External Loops: subjective perception (rather than objective reality) of reward–behavior contingency strength not only facilitates immediate reward learning (Internal Loops) but also modifies long-term beliefs about reward–behavior associations, corrects reward learning biases, and ultimately alleviates anhedonia (External Loops). The findings are expected to elucidate pathophysiological mechanisms of anhedonia and establish novel therapeutic frameworks.

Key words: anhedonia, subclinical high-risk population, psychological intervention, reinforcement learning model, perceived control