A Comparative Analysis of Cognitive Feedback Between GPT-4.0 and Teacher in Flipped Classrooms

This study investigated the efficacy of GPT-4.0 feedback for substitute teachers in flipped classroom environments. In the study, GPT-4.0 and course teachers delivered cognitive feedback through an online discussion platform respectively. At the end of each topic, 89 third-year university students (53 females) majoring in Educational Technology were required to complete a self-directed learning questionnaire, a learning satisfaction survey, and a learning performance assessment without being aware of the feedback providers' identities. The research results revealed that while GPT -4.0 exhibits some limitations in addressing objective questions, its capacity to enhance self-regulated learning and learner satisfaction is on par with traditional teacher-provided feedback. These findings underscore the imperative for ongoing research and refinement of artificial intelligence applications in educational technology. Future studies should focus on the broader role of artificial intelligence across diverse educational contexts and disciplines and assess its long-term effects.