.png){kind=icon}
.png){kind=icon}
Return
Student-AI collaboration patterns in project-based learning: Temporal and structural characteristics
.png){kind=icon}
PRE
AI
X
DOI：10.1080/15391523.2026.2661697.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
As generative AI (GenAI) becomes increasingly integrated in project-based learning (PjBL), understanding the dynamic student-AI collaboration is essential. This study employed an integrated data analysis framework (e.g., sequence analysis, process mining) to identify strategic collaboration patterns with GenAI among 40 undergraduates in a semester-long PjBL course. Three patterns emerged, each with unique characteristics. Pattern 1 (strategic balanced collaborator), associated with the highest performance, involved initial autonomous exploration followed by an “assistance–evaluation” loop, whereas pattern 3 (dependent support-seeker) showed reliance on direct assistance with the lowest performance. Findings suggest that GenAI’s pedagogical value depends not on usage frequency but on the timing, quality, and intentionality of engagement, highlighting the need to cultivate relevant literacies that leverage GenAI while safeguarding core traits fundamental to meaningful PjBL.
Keywords:
Generative artificial intelligence
human-AI collaboration
project-based learning
learning analytics
Journal
IF:
5
Papers:
164
Citations:
2.7K
Organization
Cited Papers