Title | Credit Assignment during Movement Reinforcement Learning |
Authors | Dam, Gregory Kording, Konrad Wei, Kunlin |
Affiliation | Univ Rio Grande, Dept Behav Sci, Rio Grande, OH USA. Northwestern Univ, Dept Phys Med & Rehabil, Rehabil Inst Chicago, Chicago, IL 60611 USA. Peking Univ, Beijing Engn Res Ctr Intelligent Rehabil Engn, Minist Educ, Dept Psychol,Key Lab Machine Percept, Beijing 100871, Peoples R China. |
Keywords | COORDINATION ADAPTATION VARIABILITY TASK |
Issue Date | 2013 |
Publisher | plos one |
Citation | PLOS ONE.2013,8,(2). |
Abstract | We often need to learn how to move based on a single performance measure that reflects the overall success of our movements. However, movements have many properties, such as their trajectories, speeds and timing of end-points, thus the brain needs to decide which properties of movements should be improved; it needs to solve the credit assignment problem. Currently, little is known about how humans solve credit assignment problems in the context of reinforcement learning. Here we tested how human participants solve such problems during a trajectory-learning task. Without an explicitly-defined target movement, participants made hand reaches and received monetary rewards as feedback on a trial-by-trial basis. The curvature and direction of the attempted reach trajectories determined the monetary rewards received in a manner that can be manipulated experimentally. Based on the history of action-reward pairs, participants quickly solved the credit assignment problem and learned the implicit payoff function. A Bayesian credit-assignment model with built-in forgetting accurately predicts their trial-by-trial learning. |
URI | http://hdl.handle.net/20.500.11897/391985 |
ISSN | 1932-6203 |
DOI | 10.1371/journal.pone.0055352 |
Indexed | SCI(E) SSCI |
Appears in Collections: | 心理与认知科学学院 机器感知与智能教育部重点实验室 北京市智能康复工程技术研究中心 |