Neural correlates of interactions between adaptive learning and hierarchical reasoning in repeated strategic games

• 成果类型: Article
• 署名作者: Feng, Jun; Jin, Jia; Zhao, Sasa; Derrington, Edmund; Qin, Xiangdong; Fu, Shiguang; Shen, Qiang; Dreher, Jean-Claude
• 署名单位: Nanjing Agricultural University; Shanghai International Studies University; South China Normal University; Universite Lyon 1; Centre National de la Recherche Scientifique (CNRS); Universite Lyon 1; Shanghai Jiao Tong University
• 刊物名称: GAMES AND ECONOMIC BEHAVIOR
• ISSN/ISSBN: 0899-8256
• DOI: 10.1016/j.geb.2026.02.011
• 发表日期: 2026
• 关键词: Cognitive hierarchy; temporoparietal junction; PREDICTION ERRORS; social cognition; brain; MODEL; computations; activation; mechanisms; inference

摘要

Repeated strategic interactions with consistent partners involve both adaptive learning and hierarchical reasoning. Hybrid models that incorporate reasoning into adaptive learning paradigms have been proposed to capture these processes and their interactions at the behavioral level. Here, we employ model-based functional magnetic resonance imaging (fMRI) to investigate the neural computations underlying decisions in repeated strategic interactions between two players using 11-20 money request games. A hybrid model explains medial prefrontal cortex activity more accurately than pure adaptive learning models. Additionally, we find that regions such as the rostral anterior cingulate, dorsolateral prefrontal cortex, right temporoparietal junction, precuneus, and anterior insula correlate with prediction error signals generated by adaptive, simulated adaptive, and sophisticated learning from the hybrid model. Together, our findings provide novel neural-level support for the interplay between adaptive learning and hierarchical reasoning in a repeated game with fixed partners.