摘要
生成式人工智能技术拥有强大的代码生成功能,因此被广泛应用于各类编程任务。但是,生成式人工智能对学习者编程学习的影响尚不明确。该研究以某高校“计算思维”课程的学习者为研究对象,通过准实验研究的方式,探索在编程问题解决过程中应用生成式人工智能工具对学习者的自我效能感、认知模式和编程行为的影响。实验结果显示:与生成式人工智能工具的交互并未显著增强学习者的自我效能感;学习者在问题解决过程中的认知加工模式更为单一;在行为模式上更加依赖“复制粘贴”类行为,而缺少“修改调试”类行为。基于上述结论,研究进一步提出相关建议:在应用生成式人工智能工具支持编程学习时,教师需要采用更为丰富的策略来提升学习者的编程自我效能感;尝试融合“人-人协作”与“人-机协作”来完成编程任务;引导学习者采用更为多样化的编程行为模式来进行代码编写。上述结论与建议对于运用人工智能技术支持编程学习具有重要的参考价值。
Generative AI technology possesses powerful code generation capabilities and is therefore widely applied in various programming tasks.However,the impact of generative AI on learners’programming learning remains unclear.This study takes learners of a‘computational thinking’course at a certain university as the research object and employs a quasi-experimental research method to explore the eff ects of applying generative AI technology support tools on learners’self-efficacy,cognitive patterns,and programming behavior during the programming problem-solving process.The experimental results show that interaction with generative AI tools does not significantly enhance learners’self-efficacy;learners exhibit more monotonous cognitive processing patterns during problem-solving;and their behavioral patterns become more reliant on‘copy-paste’actions while lacking‘modifi cation and debugging’behaviors.Based on these fi ndings,the study further proposes the following recommendations:when applying generative AI tools to support programming learning,teachers should adopt more diverse strategies to enhance learners’programming self-effi cacy;attempt to integrate‘human-human collaboration’with‘human-machine collaboration’to complete programming tasks;and guide learners to adopt more diverse programming behavior patterns for code writing.The aforementioned conclusions and insights hold significant reference value for the application of AI technology to support programming learning.
作者
徐静
李泊文
吴林静
高喻
Xu Jing;Li Bowen;Wu Linjing;Gao Yu(Faculty of Artifi cial Intelligence in Education,Wuhan,Hubei,China 430079;Hubei Key Laboratory of Digital Education Central China Normal University,Wuhan,Hubei,China 430079;School of Intelligent Manufacturing,Kaifeng University,Kaifeng,Henan,China 475001)
出处
《数字教育》
2025年第6期9-16,共8页
Digital Education
基金
国家自然科学基金项目“数据驱动的在线学习协作会话过程监测与干预机制研究”(72174070)
数字教育湖北省重点实验室开放基金资助项目(F2024G02)。
关键词
生成式人工智能
编程学习
人机协作
认知模式
编程行为
Generative AI
Programming learning
Human-machine collaboration
Cognitive patterns
Programming behavior
