【作 者】：
田 浩，武法提
【关 键 词 】：
复杂问题解决； 认知投入； 同步生理响应事件； 动态演化； 学习预测
【栏 目】：
课程与教学
【中文摘要】：
当今社会的动态性对学习者的复杂问题解决能力提出了前所未有的要求，学习者解决复杂问题时的认知投入直接影响任务完成的效率和质量。研究以设计类问题解决为背景，采集学习者的皮肤电数据，提出一种基于同步生理响应事件的认知投入测量方法。研究重点探究了学习者在复杂问题解决中的认知投入动态演化特征，并分析其与个人及小组绩效的关联，最终使用随机森林算法构建绩效的预测模型。研究发现：小组在解决复杂问题时，同步生理响应事件频次逐渐增加；在个体层面，学习者的投入敏捷度和持久度展现出显著的动态演化；在小组层面，高绩效小组在投入强度和同步性上变化显著，而低绩效小组仅在投入持久度上体现出明显变化；在所有特征中，方案生成阶段的投入敏捷度是预测个人绩效的关键因素，而观点交流阶段的投入持久度则对小组绩效具有最佳预测效果。研究拓展了复杂情境下认知规律的识别方法，同时对提升学生复杂问题解决能力提供了实证依据。
【英文摘要】：
In today's society, the dynamic nature puts unprecedented demands on learners' complex problem solving ability, and learners' cognitive engagement in solving complex problems directly affects the efficiency and quality of task completion. The study collects learners' galvanic skin response data in the context of design-based problem-solving, and proposes a method for measuring cognitive engagement based on synchronous physiological response events. The study focuses on the dynamic evolution features of cognitive engagement in complex problem-solving, analyzes its association with individual and group performance, and ultimately uses the random forest algorithm to construct a predictive model for individual and group performance. It is found that the frequency of synchronous physiological response events gradually increases when the group solving complex problems. At the individual level, learners' engagement agility and persistence demonstrate significant dynamic evolution. At the group level, high-performance groups exhibit significant changes in engagement intensity and synchronicity, while low-performance groups show significant changes only in engagement persistence. Among all features, the engagement agility in the plan generation phase is a key factor to predict individual performance, and the engagement persistence in the perspective exchange phase has the best predictive effect on group performance. The study expands the method of identifying cognitive patterns in complex situations, and provides empirical evidence for improving students' abilities to solve complex problems.