AI-DRIVEN ADAPTIVE RESOURCE SUGESSTION IN DIGITAL LEARNING ENVIRONMENTS: INSIGHTS FROM EDUCATIONAL PSYCHOLOGY

Abstract

This study examines the effectiveness of an AI-driven adaptive resource delivery system in enhancing learning outcomes within a digital reading environment. Drawing on principles from educational psychology—particularly cognitive load theory and individualized scaffolding—the system adjusted instructional materials in real time according to learners’ performance indicators. A randomized controlled experiment was conducted with 152 undergraduate students assigned to either an adaptive learning condition or a fixed-resource control condition. Both groups studied identical content, but only the adaptive group received personalized adjustments such as difficulty modulation, targeted hints, and additional practice opportunities. Results showed that learners in the adaptive condition demonstrated significantly greater gains in reading comprehension than those in the control condition, as evidenced by a significant Time × Group interaction and higher post-test scores. Moreover, adaptive learners reported significantly lower cognitive load, indicating more efficient cognitive processing. These findings provide empirical support for integrating AI-based adaptive systems into digital learning environments and highlight the value of aligning algorithmic personalization with established theories in educational psychology.