Answers are grounded in this paper and cite their sources.
Tests spatial memory in virtual and augmented reality, suggesting reality modality has limited impact.
Virtual reality (VR) is a powerful tool for researching human spatial memory, but potential differences between user behavior in virtual versus real environments are often confounded by multiple entangled factors such as locomotion type (e.g. physical vs. virtual), interface (e.g. keyboard vs. motion), and the user’s surrounding physical environment. A less explored factor is Reality Modality—effects that arise from knowing an environment is virtual rather than real. While reality modality is widely assumed to matter, its effects have rarely been tested independently due to challenges with isolating it. Here, we tested participants on the same spatial memory task in three matched conditions designed to isolate its effects—(1) Real-world passthrough Augmented Reality (AR) (2) immersive VR reconstructed by scanning the same real environment and using the same headset and interface in the same physical room, and (3) a traditional desktop setup. We found no significant differences between AR and VR in subjective measures such as immersion, enjoyment, difficulty, cybersickness, or in the objective measure of spatial memory accuracy. In contrast, both outperformed the desktop condition across all measures. These results suggest that if VR environments are naturalistic enough in design, interface and movement affordances, reality modality itself has limited influence on spatial memory and therefore spatial memory performance in them will match real world performance despite taking place in VR. Our findings increase our basic science understanding of the parameter of Reality Modality and the disentangled contribution of different parameters to spatial memory accuracy, with implications for the design of extended reality applications and for generalizing results along the virtuality-reality continuum.
No takes yet. Share an insight, caveat, or question.
Mayrose et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: