Presence, rather than prior exposure, is the more strongly indicated factor in the accurate perception of egocentric distances in real world co-located immersive virtual environments

In recent work [4,2], we have discovered that people are able to make surprisingly accurate judgments about egocentric distances in an immersive virtual environment (IVE) in the special case that the IVE represents a high fidelity model of the same physical space that the user is actually occupying, and the user has been able to unambiguously verify this by viewing the real space prior to donning the display upon which the corresponding virtual environment is presented. Through followup experiments in multiple locales, we have verified that the key factor enabling this distance perception accuracy is fact of co-location, rather than any particular characteristics of the physical environment [1].One possible interpretation of these intriguing results is that observers are better enabled to make accurate judgments of egocentric distance in an IVE when they are when they are cognitively 'immersed' or 'present' in the IVE -- i.e. when they readily accept the virtual environment as being 'equivalent' to the real world and are therefore prepared to act in the virtual world in the same way that they act in the real world [3]. However, another interpretation is also possible: it could be that people are able to make accurate judgments of egocentric distances in a virtual environment when they know that it exactly corresponds to a recently viewed real environment because they are able to form a metrically accurate mental model of the spatial structure of the real environment from their brief exposure to it, so that when they are subsequently presented with the virtual environment they simply calibrate their mental model of distances in the IVE to be consistent with their remembered model of the corresponding real environment.In order to disambiguate between the 'presence' hypothesis and the 'spatial memory' hypothesis, we conducted the following study. Using a between-subjects design, we asked observers to make judgments of egocentric distance via blind walking in a real room and in one of three different virtual models, each of which was described to the participants as representing a "high fidelity virtual model of that same room". However, only one of the virtual models was actually an identical match in size to the real room. Nine of our 23 participants viewed a virtual model in which each of the walls had been surreptitiously moved 3 ft inward towards the center of the room, and another nine viewed a virtual model in which each of the walls had been surreptitiously moved 3.75 ft outwards from the center of the room. In each case, the textures were touched up in Photoshop to effect the changes without scaling anything. The remaining 5 participants viewed the same sized virtual model, replicating our earlier study. Acoustic cues were muffled for all participants by a radio playing static, and no training or feedback was given at any time.If the 'presence' hypothesis holds, we would either expect that the performance of participants in each group would be about the same, or that distances would be slightly underestimated in each of the artificially manipulated rooms. However, if the 'spatial memory' hypothesis holds, then we would expect that distance judgments would tend to deviate in opposite directions in the smaller and larger rooms, relative to the in the same room: participants who experience the smaller virtual room should overestimate distances in the virtual environment, and participants who experience the larger room should underestimate distances in the virtual environment, relative to in the real room.Figure 1 shows the three virtual environments used in this study, and figure 2 shows the average relative error in the distance judgments made by each participant in each virtual environment (vertical axis), compared to their error in the real environment (horizontal axis). We can see that most participants who experienced the accurately sized virtual room made distance judgments that were nearly equivalent in the real and virtual environments, consistent with our earlier findings [2]. However, many of the participants who experienced the smaller room model, and nearly all of the participants who experienced the larger room model, judged distances to be shorter, on average, in the virtual world than in the real world. These trends are statistically significant, and seem to support the 'presence' hypothesis more strongly than the 'remembered size' hypothesis.