The theory that cannot learn: A conceptual–analytical critique of modality-based learning styles

Learning-styles theory remains one of the most influential ideas in educational psychology despite decades of empirical criticism. The theory assumes that learners possess stable sensory modalities (visual, auditory, or kinesthetic) and that instruction tailored to these modalities improves learning outcomes. Although this premise has intuitive appeal, empirical research has repeatedly failed to confirm its central prediction. This conceptual-analytical paper examines the theoretical structure of learning-styles theory and evaluates whether the empirical literature supports its core claims. Drawing on conceptual analysis, psychometric evidence, and prior experimental studies, the paper reconstructs the theory's central assumptions and formalizes its primary empirical prediction: if learning styles are valid, instruction matched to a learner's modality should produce a crossover interaction in which matched instruction yields superior outcomes relative to mismatched alternatives. Using this criterion, the paper reviews evidence from controlled studies and systematic reviews, which consistently fail to detect the predicted interaction. The analysis further examines the psychometric instability of learning-style classifications and the absence of a clearly specified cognitive mechanism linking modality preference to learning outcomes. Taken together, these findings indicate that learning-styles theory lacks reliable empirical support and provides limited guidance for instructional design. The paper concludes by considering why the theory persists in educational practice despite sustained empirical challenges.