Algorithmic Mediation, Learner Agency, and Cognitive Autonomy in AI-Enhanced Online Learning Environments

This article examined the psychological and behavioral implications of artificial intelligence integration in online learning environments, with a focus on algorithmic mediation and its influence on learner agency, cognitive autonomy, and self-regulated learning. AI-enhanced platforms increasingly employ personalization algorithms, predictive analytics, and adaptive feedback to structure learning pathways and guide learner behavior. Using a conceptual-analytical research design, the study synthesized peer-reviewed literature from psychology, behavioral sciences, and educational research to analyze how these algorithmic mechanisms interact with cognitive and motivational processes. The analysis drew on established theoretical frameworks in self-determination theory, cognitive psychology, and behavioral regulation to map algorithmic guidance onto learner perceptions of control, responsibility, and autonomy. Findings indicated that adaptive sequencing and continuous feedback may enhance perceived competence and reduce cognitive load, while simultaneously externalizing decision-making processes that are central to autonomous learning. Algorithmic nudging and data-driven recommendations were shown to recalibrate self-regulatory behaviors by shifting regulation from internally driven metacognitive processes toward externally mediated cues embedded in system design. Variations in the intensity of algorithmic guidance were associated with corresponding differences in perceived autonomy, intrinsic motivation, and reliance on external regulation. By clarifying the cognitive and behavioral mechanisms through which AI-mediated systems shape learning behavior, the study contributed to psychological understandings of autonomy in digital learning environments and provided a theoretically grounded perspective relevant to the design of autonomy-supportive AI-enhanced educational systems.