Scholarlink

We employ Reinforcement Learning (RL) models to unravel the mechanisms behind the learning behavior of two macaque monkeys engaged in a free-moving multi-target reaching task. The study was conducted using computer simulations reflecting the animal’s learning conditions, and compared with the actual arm movements recorded on two macaque monkeys on a Kinarm apparatus. Our paper thus provides important insights for the design of motor control learning systems, combining end-effector control design with the learning of motor chunks associations. Our research is of interest for the modeling and understanding of natural motor learning systems, but also heads toward the design of more “brain-inspired” adaptive robotic manipulators.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Non-instructed Motor Skill Learning in Monkeys: Insights from Deep Reinforcement Learning Models

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Laurène Carminatti,
  • Lucio Condro,
  • Alexa Riehle,
  • Sonja Grün,
  • Thomas Brochier,
  • Emmanuel Daucé

摘要

We employ Reinforcement Learning (RL) models to unravel the mechanisms behind the learning behavior of two macaque monkeys engaged in a free-moving multi-target reaching task. The study was conducted using computer simulations reflecting the animal’s learning conditions, and compared with the actual arm movements recorded on two macaque monkeys on a Kinarm apparatus. Our paper thus provides important insights for the design of motor control learning systems, combining end-effector control design with the learning of motor chunks associations. Our research is of interest for the modeling and understanding of natural motor learning systems, but also heads toward the design of more “brain-inspired” adaptive robotic manipulators.