A novel behavioral task for investigating a mixture of model-free and model-based reinforcement learning in rodents

2017 Fiscal Year Final Research Report

• Project/Area Number: 16K13505
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Experimental psychology
• Research Institution: Nagoya University
• Principal Investigator: Mizoguchi Hiroyuki 名古屋大学, 環境医学研究所, 講師 (70402568)
• Project Period (FY): 2016-04-01 – 2018-03-31
• Keywords: モデルフリー・モデルベース / 意思決定 / 行動選択

Outline of Final Research Achievements

Decision-making and reinforcement learning arise from the involvement of two distinct processes that control choice, known as model-based and model-free strategies. Model-based choices are prospective, whereas model-free choices are retrospective. In general, normal performance depends on the flexible integration of model-based and model-free control. In this research, we employed a two-step sequential visual discrimination learning of complex stimuli in touch-screen task to establish a novel behavioral system for investigating a mixture of model-free and model-based reinforcement learning in rodent. Rats chose option on the reward amount- and provability-dependent manner. However, we failed to find best fitting behavioral modeling and parameters. We need to continue our experiments in future studies.

Free Research Field

神経薬理学