Motivation Estimation of Socio-Emotional Competence Acquisition by Behavioral Modeling of Infants

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K11975
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61040:Soft computing-related
• Research Institution: Kyushu Institute of Technology
• Principal Investigator: HORIO Keiichi 九州工業大学, 大学院生命体工学研究科, 教授 (70363413)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 行動解析 / 行動モデリング / 強化学習 / モティベーション推定 / 自動アノテーション

Outline of Final Research Achievements

Regarding the development of socio-emotional competence in early childhood, by pursuing the mechanism using an information engineering approach, (1) development of technology to automatically quantify the actions and remarks of each infant from the captured video, (2) building a behavioral model of infants and extracting behavioral changes accompanying growth, (3) estimation of infant motivation that contributes to behavioral change by applying the concept of reverse reinforcement learning, were performed. By matching the behavior of infants measured using reinforcement learning as a model, it was clarified that the gaze and utterances of surrounding infants had an effect on changes in the behavior of each infant, such as changes in the frequency of utterances. It was shown that the surrounding prefectural boundaries and their degree of influence are different.

Free Research Field

知的情報処理

Academic Significance and Societal Importance of the Research Achievements

幼児教育において，非認知能力の重要性が注目されているが，本研究では，ディスカッションという限定的ではあるが社会性に関係した状況において，各幼児の行動変容にどのような環境および周囲の幼児の行動が影響を及ぼすかを検証するものである．今後，幼児の日常の行動を計測する技術，保育士や保護者による幼児の非認知能力の定積的評価と合わせて，本研究の方法論を発展させることができれば，非認知能力およびその発達の定量的な評価につながり，社会的意義は大きい．