生成式人工智能辅助学习的作用机制研究——以“概率论与数理统计”课程为例

Research on the Mechanisms of Generative Artificial Intelligence-Assisted Learning—A Case Study of the Course “Probability Theory and Mathematical Statistics”

摘要: 为探究生成式人工智能辅助学习在“概率论与数理统计”课程中的作用机制，本研究综合运用信效度检验、回归分析与结构方程模型开展实证分析。研究构建并验证了涵盖提示工程能力、批判性思维等变量的AI辅助学习行为评价模型，并整合课程成绩与学习满意度形成综合学习成果指标。结构方程模型表明，“提示工程–批判性思维–综合学习成果”构成显著的中介作用路径。进一步分析发现，学生在不同学习任务与使用场景中对AI的使用方式和效果存在显著差异。学生的核心需求主要集中于提示工程能力的提升、批判性思维的培养以及AI与学科教学的深度融合。本研究揭示了AI辅助学习由技术使用向学习成效转化的内在机制，为优化智能教学设计与协同育人模式提供了实证支持。

Abstract: To investigate the mechanisms through which generative artificial intelligence facilitates learning in the course “Probability Theory and Mathematical Statistics”, this study conducted an empirical analysis using reliability and validity tests, regression analysis, and structural equation modeling. An evaluation model of AI-assisted learning behaviors was constructed and validated, incorporating key variables such as prompt engineering and critical thinking. Course achievement and learning satisfaction were further integrated to form a comprehensive learning outcome index. The results of the structural equation model indicate that prompt engineering competence, critical thinking, comprehensive learning outcomes constitutes a significant mediating pathway. Further analyses indicate notable differences in students’ patterns and effectiveness of AI use across different learning tasks and application scenarios. Students’ core needs are mainly focused on the enhancement of prompt engineering, cultivation of critical thinking, and the deep integration of AI with subject-specific teaching. This study indicates the intrinsic mechanism by which AI-assisted learning translates from technological use to learning outcomes, providing empirical support for optimizing intelligent instructional design and collaborative education models.