Over the last decade, augmented reality (AR) has been widely adopted in architectural education, yet it is still often treated as a visualization add-on rather than as an operative design medium. This paper argues that AR becomes pedagogically meaningful when it is anchored to physical or graphic artefacts so that overlays function not as final images, but as reversible instruments for testing, adjustment, and spatial verification. Building on reflection-in-action as a model of situated design learning, the study examines two teaching experiences: one focused on the AR-based translation of complex two-dimensional graphic fields into three-dimensional hypotheses, and another centred on kinematic reasoning through equilibrium and iterative adjustment. The article proposes that error within AR-based workflows has a double pedagogical role: first, as corrective feedback, when mismatch reveals imprecision, insufficient legibility, or unstable alignment in the target; and second, as generative design feedback, when recalibration and reconfiguration trigger new spatial hypotheses or bidirectional transfers between physical and digital models. Evidence is based primarily on analytic observation of documented episodes and on visual documentation of process transformations, complemented by a background evaluative scaffold and supplementary student feedback where available. Results indicate that AR can (a) increase the material and graphic precision of the supporting artefact; (b) strengthen spatial and kinematic understanding by making intermediate states and inconsistencies visible; and (c) turn mismatch and recalibration into operative parts of the design process itself. The paper therefore reframes AR in architectural education not as a representational endpoint, but as a medium of verification, adjustment, and projective transformation.
Morales et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: