Scene Awareness While Using Multiple Navigation Aids in AR Search

TL;DR

The paper investigates how displaying multiple navigation cues in mobile augmented reality affects users’ scene awareness and memory for real-world objects during a guided search. Using a within-subject design, participants searched for virtual gems in an indoor AR environment while using world-locked arrows, an on-screen radar, both, or neither, with a stability manipulation of the arrows; after each trial they completed a recall test. Results show that object recall declines when navigation aids are active, especially for radar, and that congruence and presence of objects modulate recall; these findings highlight perceptual and cognitive costs of AR guidance and motivate adaptive interfaces to preserve environment awareness. The study thus contributes evidence that adaptable cueing can balance navigational support with safe, attentive interaction in mobile AR settings, with implications for design and safety in real-world search and navigation tasks.

Abstract

Augmented reality (AR) allows virtual information to be presented in the real world, providing support for numerous tasks including search and navigation. Allowing users access to multiple navigation aids may help leverage the benefits of different navigational guidance methods, but may also have negative perceptual and cognitive impacts. In this study, users performed searches for virtual gems within a large-scale augmented environment while choosing to deploy two different navigation aids either independently or simultaneously: world-locked arrows and an on-screen radar. After completing the search, participants were asked to recall objects that may or may not have been present in the scene. The use of navigation aids impacted object recall, with impaired recall of objects in the environment when an aid was switched on. The results point at possible impact factors of object awareness in mobile AR and underscore the potential for adaptable interfaces to support users navigating the physical world.

Figures (2)

• Figure 1: Top: Experiment environment, Bottom: Some small, medium, and large (left to right) objects included in the recall tests. a) semantically congruent objects. b) semantically incongruent objects. Overall, our set of candidate objects for recall questions consisted of 18 virtual object models and another 18 names of objects that were always absent from the scene.
• Figure 2: (a) Object recall accuracy as a function of semantic congruence and object presence. Participants were worse at remembering objects that were present compared to identifying absent objects. Error bars are 95% CI. (b) recall accuracy (for objects present in the scene only) as a function of the three baseline conditions. Recall was worse with the radar compared to no aid.