ChatGPT as a student metacognitive reflection tool for recognizing errors and mistakes in mathematics at university level

The aim of this study was to investigate the contribution of ChatGPT in promoting metacognitive activities among university students. Nine first-year Albanian students attending the Calculus 2 course initially worked in class on a task involving maxima and minima of functions of two variables. Subsequently, at home, they reviewed their solutions by comparing them with those generated by ChatGPT and completed a metacognitive questionnaire. The analysis, based on metacognitive activity categories, shows that the combination of the questionnaire and the comparison with ChatGPT’s solutions fostered students’ metacognitive engagement, particularly in terms of monitoring, evaluation, and elaboration. This process also supported the identification of students’ mistakes and encouraged reflection on them. Overall, the activity proved effective in promoting self-criticism as well as reflection on mathematical concepts and on the use of AI as a didactical tool. The findings suggest that integrating AI-generated solutions with structured metacognitive prompts can effectively support reflective thinking in university mathematics, provided that the tasks are designed to elicit deeper strategic awareness. However, no student demonstrated engagement in planning-related metacognitive activity, indicating that the questionnaire design was not sufficient to prompt reflection on problem-solving strategies.