Abstract
Numerous experiments investigating egocentric distance perception in immersive virtual environments have found that distances are underestimated in virtual environments relative to in the real world. However we have recently discovered a notable exception to this rule: when the virtual environment represents an exact 3D replica of the occupied real environment, and when the participant is allowed to unambiguously verify this by viewing the real environment before donning the display on which the corresponding virtual environment is presented, distances are judged nearly equivalently in the virtual and real worlds.
Two explanations are possible: 1) the accuracy of participants' distance judgments is related to their sense of ‘presence’ in the virtual environment, which increases when they have unambiguous evidence that the presented virtual environment corresponds to the occupied real environment; or 2) participants are able to use metric information gleaned from their prior exposure to the real space to calibrate the size of the corresponding space presented via the head mounted display. To disambiguate, we have designed a follow up experiment in which we surreptitiously vary the positions of the walls in the virtual environment so that the floor space (distances between opposing walls) is either slightly less or slightly greater in the virtual environment than in the real environment. If the first hypothesis holds, then we should not expect to see much degradation in the accuracy of participants' distance judgments as a result of these manipulations; however if the second hypothesis holds, then the accuracy of distance judgments should decline.
This research was supported by NSF (IIS-0313226)