The effects of immersion and body-based rotation on learning multi-level indoor virtual environments

Abstract

The goal of this study was to investigate how the immersion level of virtual environments (HMD vs. desktop) and rotation method (physical vs. imagined) affects wayfinding performance in multi-story virtual buildings and the development of multi-level cognitive maps. Twelve participants learned multi-level virtual buildings using three VE conditions (physical rotation HMD, physical rotation desktop and imagined rotation desktop). They were then tested on four cross-level tasks, including: pointing, route navigation, vertical navigation, and paper-based drilling. Results showed that performance on between-floor trials was reliably worse than for within-floor trials and that this difference was neither improved by the level of immersion of the display nor the rotation behavior used during navigation. Our data suggest that increasing the fidelity of these interface variables does not yield more accurate development of multi-level cognitive maps. Indeed, multi-level indoor wayfinding performance was as effective with the simplest and least expensive desktop display based purely on joystick navigation as the more complex VE platforms. These findings show that spatial cognition research in multi-level virtual buildings need not be limited to immersive VEs with physical body rotation which require considerable equipment cost and increased technical complexity.