SHAPING SELF-REGULATED AND METACOGNITIVE LEARNING THROUGH ARTIFICIAL INTELLIGENCE AND LEARNING ANALYTICS

Abstract

Artificial intelligence (AI) and learning analytics are increasingly embedded in educational systems, yet their implications for self-regulated learning (SRL) and metacognitive development remain only partially understood. This study conducts a systematic literature review of 34 empirical and review articles investigating AI-driven learning analytics in formal education settings, focusing on their impact on SRL, motivation, and academic performance. Following PRISMA-aligned procedures, the review identifies studies involving AI-based analytics tools, standardized SRL or metacognitive assessments, and empirical designs with pre–post or comparison data. The results indicate that AI-driven systems such as predictive models, adaptive platforms, chatbots, and generative AI can strengthen core SRL processes, increase motivation and engagement, and improve academic outcomes across diverse subjects and educational levels. However, the review also reveals persistent implementation challenges, including infrastructural limitations, teacher preparedness, privacy and ethical issues, and potential over-reliance on AI that may weaken independent strategic thinking. The article concludes that AI-driven learning analytics can meaningfully support SRL when embedded within coherent pedagogical frameworks and supported by institutional policies, and it outlines directions for future research on mechanisms, equity, and long-term effects.