认知—情感—社会:多智能体支持复杂问题解决学习的实践理路与发展路向OA
Cognitive-Affective-Social:Practical Pathways and Future Directions of Multi-Agent-Supported Complex Problem-Solving Learning
复杂问题解决学习是践行从知识传授为重转向以能力提升为本理念的重要方式,而多智能体为理解和促进复杂问题解决学习提供了新的实践思路.研究聚焦多智能体如何支持复杂问题解决学习这一核心问题,首先立足相关理论框架解析了复杂问题解决学习的内涵,厘清其与项目式学习、探究性学习的差异,并从认知、情感、社会三个维度解构其内在过程;然后,系统梳理多智能体支持复杂问题解决学习的实践现状,剖析其在认知、情感、社会等方面的学习支持作用与研究局限.在此基础上,从辅助工具、资源环境、参与主体三重定位,系统阐释多智能体支持复杂问题解决学习的功能角色与实践理路,即:以助手角色提供认知支架引导深度学习,以专家角色实现人机协同的情感觉察与调节,以学伴角色促进人机认知共享与社会协作.最后,从深层认知影响、人机情感连接、社会学习结构三方面分析多智能体带来的问题挑战,以此提出多智能体支持复杂问题解决学习的未来发展路向,推动探索人机协同的作用机制.研究旨在搭建多智能体支持复杂问题解决学习的理论框架,明晰其实践理路与发展路向,为教育数字化进程中构建人机协同的个性化学习新样态提供参考.
Complex problem-solving learning is an important approach to shifting education from knowledge transmission to competency development,while multi-agent systems provide new possibilities for understanding and facilitating such learning.Focus-ing on how multi-agent systems support complex problem-solving learning,this study first analyzes the connotation of complex prob-lem-solving learning based on relevant theoretical frameworks,clarifies its differences from project-based learning and inquiry-based learning,and deconstructs its internal processes from cognitive,affective,and social dimensions.It then systematically reviews current practices of multi-agent-supported complex problem-solving learning and examines the learning support provided by multi-agent sys-tems in cognitive,affective,and social aspects,as well as the limitations of existing research.Based on that,this study explains the functional roles and practical pathways of multi-agent systems in supporting complex problem-solving learning from auxiliary tools,resource environments,and participating agents,.Specifically,multi-agent systems can serve as assistants that provide cognitive scaf-folding to guide deep learning,as experts that enable human-machine collaborative affective awareness and regulation,and as learn-ing partners that promote human-machine cognitive sharing and social collaboration.Finally,this study analyzes the challenges posed by multi-agent systems in terms of deep cognitive influence,human-machine affective connection,and social learning structures.It further proposes future directions for multi-agent-supported complex problem-solving learning,with the aim of advancing research on the mechanisms of human-machine collaboration.This study seeks to develop a theoretical framework for multi-agent-supported complex problem-solving learning,clarify its practical pathways and future directions,and provide insights for building new forms of human-machine collaborative personalized learning in the process of digitalization of education.
郝祥军;顾小清
南京师范大学教育科学学院(江苏南京210097)华东师范大学教育学部(上海200062)
社会科学
复杂问题解决学习人工智能多智能体认知与情感人机协同
Complex problem-solving learningArtificial intelligenceMulti-agent systemsCognition and affectHuman-ma-chine collaboration
《远程教育杂志》 2026 (3)
48-58,11
国家社会科学基金2025年度教育学青年项目"多智能体支持复杂问题解决学习的协同作用机制研究"(项目编号:CCA250275).
评论