Abstract
Acquiring a precise understanding of three-dimensional anatomical relationships remains a fundamental challenge in medical education. Traditional pedagogical methods, such as two-dimensional atlases and physical models, often impose high cognitive demands as students mentally translate flat images into multi-dimensional spatial layouts. While Augmented Reality (AR) offers a promising alternative, the choice of display modality significantly influences user experience and learning efficiency. This study presents a custom-designed Spatial Augmented Reality (SAR) projection system that overlays dynamic, three-dimensional digital instructions directly onto a physical anatomical mannequin. We conducted a randomized controlled trial (N = 60) comparing our SAR system against Hand-Held Augmented Reality (HHAR) and a traditional Two-Dimensional Screen (2DS) during a multi-organ assembly task. Spatial cognitive load was evaluated using the NASA Task Load Index (NASA-TLX) and a specialized Spatial Cognitive Load Scale (SCLS), alongside objective performance metrics including task completion time and positioning accuracy. The results reveal that the SAR system significantly reduced mental demand, temporal demand, and extraneous spatial cognitive load compared to both the HHAR and 2DS modalities. Furthermore, participants in the SAR condition exhibited faster assembly times and fewer positioning errors. These findings suggest that co-locating instructional overlays directly onto physical targets mitigates the split-attention effect, offering a highly effective and ergonomic paradigm for anatomical and clinical training.
