Generative AI-Supported Learning, Cognitive Offloading, and Metacognitive Awareness: Examining Changes in Students’ Learning Regulation

Abstract

This study examined the relationships among generative AI-supported learning, cognitive offloading, metacognitive awareness, and students’ learning regulation among university students in Pakistan. A quantitative cross-sectional research design was employed, and data were collected from 293 university students through a self-administered online questionnaire distributed using Google Forms. The collected data were organized in Microsoft Excel and analyzed using IBM SPSS Statistics. Descriptive statistics were used to summarize demographic characteristics, Pearson correlation was applied to examine relationships between variables, and simple linear regression was conducted to assess predictive relationships. The illustrative analysis indicated a positive association between generative AI-supported learning and metacognitive awareness (β = 0.478, p = .001), with the regression model explaining 22.8% of the variance in metacognitive awareness (R² = 0.228). Cognitive offloading was positively associated with students’ learning regulation (r = 0.436, p = .001), while metacognitive awareness positively predicted learning regulation (β = 0.568, p = .001), explaining 32.3% of its variance (R² = 0.323). These findings highlight the potential relevance of generative AI use and metacognitive awareness to students’ learning regulation. The study emphasizes that AI tools should support, rather than replace, independent thinking, critical evaluation, and active learning. Educators may use these insights to design learning activities that promote reflection, self-monitoring, and responsible AI use. The study also provides a basis for future research examining these relationships across different academic disciplines and educational settings. However, the cross-sectional design limits causal interpretations and does not establish changes in learning regulation over time.