High and low heutagogy orientations among mathematics pre-service teachers and their relation to academic AI-usage

Introduction This study investigates heutagogical orientations among pre-service mathematics teachers (PSTs) and their associations with AI academic usage frequency and demographic-academic variables (gender, age, academic year, degree, teacher-education program). Heutagogy emphasizes self-directed learning, reflection, cybergogy, and problem-solving self-efficacy in technology-rich environments.​ Methods Self-report data were collected from 78 PSTs using validated scales for heutagogy and AI usage. Cluster analysis identified distinct profiles, followed by hierarchical binary logistic regression to examine predictors, controlling for background variables. Results Two profiles emerged: high-heutagogy (elevated self-directed learning, reflection, cybergogy, problem-solving self-efficacy) and low-heutagogy. Frequent AI academic use more than doubled the odds of high-heutagogy membership, even after controls. First-year PSTs showed stronger heutagogical orientations than second-year PSTs. Discussion Findings highlight AI-supported environments' role in fostering autonomous, reflective engagement with mathematics among PSTs, influenced by program structures and study phase rather than linear development. Self-report and cross-sectional limitations preclude causality; longitudinal/intervention studies are recommended. Teacher education should integrate AI as a co-agent to cultivate heutagogical dispositions for technology-rich contexts.