Conceptual Domains of AI in Education for Neuroinclusive Measurement Using Topic Modelling

The increasing integration of artificial intelligence (AI) in educational contexts calls for measurement instruments that adequately capture learners’ experiences from a neuroinclusive perspective. This study examines the conceptual structure of recent AI-in-education research to inform early phases of scale development. A systematic Scopus search identified 411 peer-reviewed journal articles published between 2020 and 2025, which were analysed using Latent Dirichlet Allocation topic modelling. Four latent domains were identified, reflecting general educational discourse on AI, analytical research perspectives, measurement and assessment, and medical or clinical training contexts. Lexical frequency analyses further indicated a limited representation of emotional and epistemic constructs.