Events

Augmented reality (AR) can make spatial information perceptually available during physical activity, but providing additional visual guidance does not necessarily lead to better spatial learning. We examine how AR visualization should configure focal and contextual information to develop spatial cognition and enable its transfer to subsequent performance. Drawing on focus-and-context visualization theory, we theorize the joint effects of two fundamental AR design features: present-state superimposition, which determines the amount of contextual spatial structure made visible, and future-state preview, which determines how strongly focal spatial transformations are externalized. We conduct two laboratory experiments. Experiment 1 uses AR-guided Rubik’s Cube training to show that the effects of animated versus discrete previews depend critically on whether present-state information is partially or fully superimposed. Experiment 2 examines whether the spatial cognition developed during AR training transfers to a dynamic drone-flight simulation requiring continuous spatial orientation, updating, and action. The findings demonstrate that AR configurations that produce superior immediate performance are not necessarily those that produce superior transfer to dynamic spatial performance. We contribute a focus-and-context account of AR design that explains how visualization configurations shape spatial cognition, attention allocation, and performance transfer, thereby shifting AR research from asking whether visual augmentation helps to determining how AR should be configured for different cognitive and performance objectives.