Individual differences in human–AI collaboration for inclusive education: A multi-level psychological and educational framework

This research programme investigates how individual differences shape the effectiveness of human–AI collaboration in learning, with the goal of supporting diverse learners and advancing more inclusive, equitable educational practices. It proposes an integrated framework in which learner characteristics, AI interaction patterns, and multi-level contextual factors jointly impact educational processes and outcomes. Grounded in person–environment fit theory, the programme suggests that alignment between personal characteristics and AI affordances enhances learning and human functioning, whereas misalignment can constrain or unevenly distribute benefits. This has direct implications for inclusive education, where misalignment can amplify disparities in access, engagement, and outcomes. By addressing diverse learning needs through person–AI fit, the programme offers a pathway toward more equitable and responsive educational design in the current agenda to learn with AI. The studies span multiple methods (surveys, lab and online experiments, classroom implementations), academic domains (social sciences/humanities and STEM), and cultural contexts (e.g., more individualistic vs more collectivistic systems). Together, they form a coherent programme designed to examine when and for whom personally fitted human–AI collaboration is most effective. This registration outlines the overarching framework, the families of constructs to be examined, and the broad hypotheses guiding the empirical work. This program aims to enhance inclusive educational practice. Its goal is not to classify students into fixed types, but to identify personally meaningful human–AI interaction patterns that help teacher‑educators design differentiated, GenAI‑supported activities and guidance. These psychological characteristics serve only as flexible lenses for scaffolding diverse learners’ AI use.