Development of theory of mind from self-other cognition facilitated by maturational constraints: A constructive approach

2015 Fiscal Year Final Research Report

• Project/Area Number: 25700027
• Research Category: Grant-in-Aid for Young Scientists (A)
• Allocation Type: Partial Multi-year Fund
• Research Field: Intelligent robotics
• Research Institution: Osaka University
• Principal Investigator: Nagai Yukie 大阪大学, 工学研究科, 特任准教授 (30571632)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 認知発達ロボティクス / 予測学習 / 自他認知 / 心の理論 / 神経回路モデル / 確率モデル

Outline of Final Research Achievements

This study has investigated the underlying mechanism for social cognitive development. Inspired by neuroscience and cognitive science studies, we have suggested sensorimotor predictive learning as a principle for cognitive development and constructed computational models using neural networks and probabilistic models to verify the principle. Our experiments using robots demonstrated how predictive learning leads to development of various types of cognitive functions. For example, our robots improved the ability to recognize others’ actions based on the ability to produce own actions, generated helping behaviors through the minimization of prediction error, and acquired the ability to recognize and generate emotions through multimodal predictive learning. These results yield new insights into underlying mechanisms for human cognitive development.

Free Research Field

認知発達ロボティクス