Artificial Intelligence and Learner Autonomy: A Meta-Analysis of Self-Regulated and Self-Directed Learning

Krishnashree Achuthan^1,2*

• ¹Amrita Vishwa Vidyapeetham (Amritapuri Campus), Kollam, India
• ²Amrita Vishwa Vidyapeetham Amrita School of Engineering Amritapuri, Kollam, India

As artificial intelligence (AI) becomes increasingly embedded in educational environments, understanding how it shapes learners’ capacity for self-regulated and self-directed learning has become a central question in contemporary learning science. The present study investigates the impact of artificial intelligence (AI)-based interventions on learners’ self-regulated learning (SRL) and self-directed learning (SDL), providing a comprehensive synthesis of 32 empirical studies comprising 92 effect sizes and 3,029 participants. Following PRISMA guidelines, the analysis examined overall SRL effects, SRL dimensions and phases, learning outcomes, and SDL. AI interventions demonstrated a large and statistically significant positive effect on overall SRL (g = 1.613, p = 0.032) and SDL (g = 1.111, p = 0.043), indicating that AI meaningfully enhances learners’ capacity to plan, monitor, and regulate their learning activities while fostering autonomy and persistence. Across SRL dimensions, AI produced moderate improvements in cognitive/metacognitive (g = 0.377, p = 0.0004) and motivational/affective regulation (g = 0.505, p = 0.013), though behavioral regulation outcomes remained inconsistent. Phase-level analyses revealed that AI interventions are most effective during the forethought phase - supporting goal setting, planning, and motivational readiness, while producing smaller but significant gains in self-reflection and variable effects during performance. AI-based systems also yielded moderate improvements in learning outcomes and achievement (g = 0.350, p = 0.034), confirming that enhanced regulation translates into measurable academic benefits. Moderator analyses indicated that AI effectiveness varied by learner, contextual, and design factors. Greater SRL gains were observed among older learners, longer intervention durations, and in language learning using interactive systems. Gender effects were minimal, suggesting equitable AI support across groups. Sensitivity and publication bias tests affirmed the robustness of findings. These results suggest that AI serves as an adaptive scaffold that strengthens cognitive, motivational, and reflective regulation, thereby promoting more autonomous and sustained learning.