SPOC互动话语中认知行为模式的识别及其适应性干预方法研究

SPOC作为高等教育教学模式的重要变革方向，其互动场景是促进个体认知加工的主要载体，但互动过程中的学习认知机理目前尚不明晰，尤其是话语认知行为规律及其干预机制仍有待深入研究，这对于解决学习者认知迷航和高阶思维难以激发等问题具有重要价值。本项目以话语认知表征、认知行为模式与学习成效三者间作用关系为切入点，研究SPOC互动话语中认知行为模式的识别和适应性干预方法，以深度解构话语认知规律及其对学习成效的影响机理。围绕认知行为的可计算、可辨识及群体认知的可控问题，研究互动话语中认知行为特征的提取方法，进行认知语义分析与概率计算；探究认知行为模式的识别方法，实现个体与群体认知状态的准确量化与表征；研究认知行为的适应性干预方法，助力于群体高阶认知的激发；开展实验验证与分析，对相关算法、模型及干预方案进行有效性验证。研究成果有望为互动教学情境的科学建构、学习认知规律的深度理解及智能化导学提供关键技术。

SPOC is an important reform direction of teaching mode in higher education, its interactive context is the main carrier to promote individual cognitive processing. However, the cognitive learning mechanism in the process of interaction is still unclear at present, especially the regularities and intervention mechanism of discourse cognitive behavior are needed to be further studied, which is of great value in solving the problems of learner cognitive confusion and difficulty in stimulating higher-order thinking. Taking relationship between cognitive discourse representation, cognitive behavioral patterns and learning outcome as the starting point, this project focuses the research on the recognition and adaptive intervention methods of cognitive behavioral patterns in SPOC interactive discourses, in order to deeply deconstruct the cognitive regularities in discourse and their impact mechanism on learning outcome. Focusing on cognitive behaviors’ computability and identifiability as well as group cognition’s controllability problems, research contents include 1) extracting cognitive behavioral features in interactive discourse to conduct cognitive semantic analysis and probability calculation; 2) exploring the recognition method of cognitive behavioral patterns to realize precise quantification and characterization of individual and group cognitive states; 3) studying the adaptive intervention method of cognitive behaviors to help stimulate group higher-order cognition; 4) conducting experimental verification and analysis to validate effectiveness of related algorithms, models and intervention schemes. The research results are expected to provide key technical support for the scientific construction of interactive teaching situations, in-depth understanding of learning cognitive regularities, and intelligent tutoring.