Process mining shows the ChatGPT group's revision loop centred on consulting ChatGPT, while the human-expert group strengthened connections between revising and reading, orientation and evaluation¶

Claim · All claims
Evidence · 1 study · causal r2 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? First-order Markov process maps show AI-group learners repeatedly returned to ChatGPT after revising and evaluating, whereas HE-group transitions linked revising with reading, orientation and evaluation. → Fan 2025

Evidence¶

Fan 2025¶

Fan, Y., Tang, L., Le, H., Shen, K., Tan, S., Zhao, Y., Shen, Y., Li, X., & Gasevic, D. (2025). Beware of Metacognitive Laziness: Effects of Generative Artificial Intelligence on Learning Motivation, Processes, and Performance. British Journal of Educational Technology. https://doi.org/10.1111/bjet.13544

q2 · i? · causal · r2

Process mining with a first-order Markov model (pMineR) on trace data from the revising stage, comparing AI vs CN and AI vs HE groups in Figure 4; red edges mark transition-probability differences larger than 10%. The AI group showed a pronounced loop involving Other, HC.EO and MC.E.

"This loop suggests that learners in the AI group predominantly relied on consulting ChatGPT during the revision stage to refine and assess their essays, thus making it their primary strategy."

Discussion¶

Related Claims¶

• During revision, supported groups (AI, human expert, checklist) engage more in elaboration, organisation and orientation, while AI and human-expert groups read less than the control group — related
• The ChatGPT group achieves significantly greater essay score improvement than the other three groups — related
• Perceived GenAI role is systematically associated with metacognitive dimensions activated: replacement-tool users show no evaluation or elaboration, collaborator users show predominantly higher-order activity — related
• Checklist tool use significantly increases Evaluation processes, an effect not observed in the ChatGPT and human-expert groups — related
• Human–GenAI collaborative interactions show promise for supporting metacognition and sociometacognition, based on an exploratory expert case study — related
• Only the Collaborator group shows a complete significant sequential metacognitive chain from orientation through elaboration; the Replacement Tool group shows only self-loops — related
• AI-section students showed tool-linked self-efficacy and attenuated metacognitive calibration, with the human-pair contrast group showing larger concept-map gains in cross-links and hierarchy — related
• Learners supported by ChatGPT, a human expert, checklist tools, or no support show no significant difference in post-task intrinsic motivation — related
• GenAI use collapses SRL cycle transitions, with metacognitive activity restructuring around the tool — possibly the same claim (merge candidate)
• Process mining of two fitness-typical groups suggests the group with more heterogeneous yellow markers and shared red markers worked more efficiently while both achieved the same summary outcome — related