Exploring large language models as an integrated tool for learning, teaching, and research through the Fogg Behavior Model: a comprehensive mixed-methods analysis

Large language models (LLMs) are a recent advancement in artificial intelligence that has the potential to revolutionize learning, teaching, and research. Still, there is room for improvement regarding how effectively LLMs could be incorporated into these environments. This study investigated the role of LLMs, specifically ChatGPT, in learning, teaching, and research contexts. To understand how motivation, ability, and triggers influence the behavior of undergraduate students, teachers, and research scholars toward LLMs, the Fogg Behavior Model (FBM) is adopted. The study revealed that the behavior of students and researchers to apply LLMs in their respective domains was greatly influenced by their motivation and ability. However, teachers exhibited little interest in incorporating any LLMs into their pedagogical strategies. In addition to these results, participants identified limitations of ChatGPT in learning, teaching, and research fields. These insights contribute valuable perspectives on the practical implementation and effectiveness of LLMs in diverse academic and research contexts. The study sheds light on the potential benefits and challenges of integrating LLMs into educational and research environments. The findings emphasize the importance of accounting for motivational factors and individual abilities when applying such models. The study’s findings offer invaluable insights for educators and researchers to harness the potential of LLMs in educational and research environments while mitigating their limitations.

大语言模型（Large Language Models，LLMs）是人工智能领域近年的重要进展，具备革新学习、教学与科研模式的潜力。但目前在如何将大语言模型高效融入此类学术场景方面，仍存在优化空间。本研究以ChatGPT为典型代表，探究大语言模型在学习、教学及科研场景中的作用。为明晰动机、能力与触发因素如何影响本科生、教师及科研学者对大语言模型的使用行为，研究采用了福格行为模型（Fogg Behavior Model，FBM）作为分析框架。结果显示，学生与科研学者在各自领域应用大语言模型的行为，显著受自身动机与能力的驱动；但教师群体对将大语言模型融入教学策略的意愿普遍较低。除上述结论外，调研参与者还指出了ChatGPT在学习、教学及科研领域存在的局限性。上述调研结果为大语言模型在各类学术与科研场景中的落地实践与效能发挥提供了极具价值的参考视角，同时阐明了将大语言模型融入教育与科研环境所蕴含的潜在价值与面临的挑战。研究结果强调，在应用此类模型时，需充分考量动机因素与个体能力差异。本研究的调研结果可为教育工作者与科研人员提供宝贵参考，助力其在教育与科研场景中充分释放大语言模型的潜力，同时有效规避其局限性。