Artificial Intelligence-Supported Adaptive Learning and Cognitive Engagement in Higher Education: A Comparative Mixed-Methods Analysis of Blended Instructional Transformation Across Two International Universities

Abstract

The rapid integration of artificial intelligence (AI) within higher education has intensified global debates concerning pedagogical transformation, cognitive engagement, institutional adaptation, and educational equity. While contemporary educational technology scholarship increasingly emphasizes AI-supported adaptive learning systems, existing learning sciences research remains fragmented regarding the institutional and pedagogical conditions under which AI integration produces meaningful learning transformation. This study investigates how AI-supported adaptive learning environments reshape learner engagement, instructional interaction, and educational outcomes through a comparative mixed-methods analysis of two higher education institutions: a research-intensive university in the United States and a digitally reform-oriented university in Finland. Drawing upon sociocultural learning theory, self-regulated learning frameworks, and cognitive engagement theory, the study analyzes comparative institutional datasets, learning analytics records, classroom interaction observations, curriculum documents, and student performance indicators collected between 2023 and 2025. The findings demonstrate that AI-supported learning environments improve instructional personalization and student participation only when embedded within coherent pedagogical ecosystems characterized by instructor facilitation, collaborative learning structures, and institutional digital governance. The comparative evidence further indicates that technological sophistication alone does not predict educational effectiveness; rather, instructional mediation, learner autonomy support, and institutional capacity significantly shape cognitive adaptation and academic resilience. This article contributes to learning sciences scholarship by developing a comparative framework linking adaptive instructional systems, pedagogical transformation, collaborative cognition, and institutional learning resilience. The study also provides policy-relevant implications for higher education governance, teacher professional development, and equitable digital learning implementation in increasingly AI-mediated educational environments.