Effects of question-chain prompting on self-regulated learning, metacognition, and achievement in a generative AI environment

Generative artificial intelligence (GenAI) offers learners immediate and personalized support, yet unguided interaction may encourage passive use and weaken self-regulated learning (SRL). This study examined whether question-chain prompting can support SRL in human-AI interaction. A quasi-experiment was conducted with 83 undergraduates enrolled in a television program production course. A control group used free-form questioning, whereas two experimental groups used either a progressive or a divergent question-chain strategy. Data included GenAI interaction logs, post-task reflections, metacognitive awareness measures, and course grades. This study found that both question-chain groups demonstrated significantly more learning-strategy and learning-monitoring behaviors than the control group. Reflection data further indicated differences in students’ approaches to human-AI collaboration, although these differences should be interpreted cautiously because the reflection prompts differed by condition. Moreover, both experimental groups scored significantly higher than the control group on knowledge of cognition, regulation of cognition, and academic achievement. No significant differences were found between the progressive and divergent question-chain approaches. The findings suggest that question-chain prompting, implemented as an instructional package of specific, sequenced, and required questions, can promote more active engagement during human-AI interaction and support metacognitive development and learning performance. Question chains offer a practical approach for guiding students’ use of GenAI in educational settings.