Research on Human-Machine Collaboration Behavior Patterns Supported by Generative Artificial Intelligence

Analyzing human-machine collaboration behavior is crucial for integrating generative artificial intelligence(GAI) into the work of pre-service teachers, thereby enhancing teaching efficiency and quality. This study focuses on 55 pre-service teachers, employing lag sequential analysis and content analysis to examine their recorded collaboration processes and reflective texts. The aim is to explore the behavior patterns and characteristics of human-machine collaboration based on GAI. The findings indicate that the sequence C2→C1/C5→C3/C4→ C2 represents an efficient human-machine collaboration behavior pattern, in which pre-service teachers ask GAI questions, refine its responses based on their understanding, and iteratively collaborate to optimize lesson plans. The advantages of GAI in this collaboration include efficiency, comprehensive knowledge, clear logic, and lack of temporal and spatial constraints. However, its disadvantages are poor comprehension, limited professionalism, lack of personalization and innovation, reduced sense of communication, and technological limitations. The results provide significant insights for optimizing educational practices through collaboration between pre-service teachers and GAI.