Egocentric distance perception in real and HMD-based virtual environments

Abstract

We conducted four experiments on egocentric depth perception using blind walking with a restricted scanning method in both the real and a virtual environment. Our viewing condition in all experiments was monocular. We varied the field of view (real), scan direction (real), blind walking method (real and virtual), and selfrepresentation (virtual) over distances of 4 meters to 7 meters. The field of view varied between 21.1 • and 13.6 • . The scan direction varied between near-to-far scanning and far-to-near scanning. The blind walking method varied between direct blind walking and an indirect method of blind walking that matched the geometry of our laboratory. We varied self-representation between having a selfavatar (a fully tracked, animated, and first-person perspective of the user), having a static avatar (a mannequin avatar that did not move), to having no avatar (a disembodied camera view of the virtual environment). In the real environment, we find an effect of field of view; participants performed more accurately with larger field of view. In both real and virtual environments, we find an effect of blind walking method; participants performed more accurately in direct blind walking. We do not find an effect of distance underestimation in any environment, nor do we find an effect of self-representation.

The Effect of Environmental Features, Self-Avatar, and Immersion on Object Location Memory in Virtual Environments

“…It extends previous work on spatial perception in VEs (Ellis and Menges, 1998;Waller et al, 1998;Loomis and Knapp, 2003;Thompson et al, 2004;Interrante et al, 2008;Ries et al, 2008;Phillips et al, 2009Phillips et al, , 2010Mohler et al, 2010;Lin et al, 2011) by suggesting an experimental task in which participants are asked to learn and recall a series of object configurations in concurrently occupied virtual and real environments.…”

“…The experimental task involves several judgments, including distance estimation, and it is not clear exactly what strategies participants use to learn object locations (Hartley et al, 2004). Previous work has shown that distance estimation is impaired within immersive VR, although including a self-avatar and increasing confidence in fidelity can reduce this impairment (Loomis and Knapp, 2003;Thompson et al, 2004;Interrante et al, 2006Interrante et al, , 2008Ries et al, 2008Ries et al, , 2009Phillips et al, 2009Phillips et al, , 2010Mohler et al, 2010;Lin et al, 2011). Our results suggest that the level of immersion is extremely important for accurate object location learning and recall, and that higher environmental fidelity may reinforce learning transfer from VEs to the real world.…”

Egocentric distance perception in HMD-based virtual environments

“…The protocol was the same as Experiment Four in Lin et al [2011] except for the viewing condition. In this experiment, subjects performed two methods of blind walking: direct blind walking and indirect blind walking .…”

“…We conducted a followup experiment to the work of Lin et al [2011]. The experimental protocol was the same as that of Experiment Four in Lin et al [2011] except the viewing condition was binocular instead of monocular.…”

“…The experimental protocol was the same as that of Experiment Four in Lin et al [2011] except the viewing condition was binocular instead of monocular. In that work there was no distance underestimation, as has been widely reported elsewhere, and we were motivated in this experiment to see if stereoscopic effects in head-mounted displays (HMDs) accounted for this effect.…”