Abstract
Multiple studies have reported that visually directed actions beyond near-space are performed in virtual environments (VEs) as if distances were perceived as 20%–50% smaller than intended. Although there is a general subjective sense that VE spaces seem smaller than real spaces, there has been little systematic work looking at other spatial dimensions in VEs beyond distance. Measuring perceived affordances is a way to assess body-based space perception in the context of action without requiring real movement. We demonstrate the feasibility of using affordance judgments to probe spatial perception in VEs and compare perceived affordances with perceived size and distance to the same targets. Judgments were made in a virtual classroom viewed through a head-mounted display. Two poles were always present, varying on each trial in diameter, spacing, and distance from the viewpoint. In a within-subject design, standing participants first performed a size judgment by holding out their hands to indicate the apparent spacing between the columns. The screen was blanked, new columns appeared at a different distance with either very narrow or wide spacing, and subjects requested the columns to be moved closer together or farther apart until they felt that they were just wide enough to walk through without turning their shoulders. After another screen blank, a third set of poles appeared. Participants closed their eyes and walked to the apparent location between the two poles. Affordance judgments of passability averaged 1.3× actual shoulder width, a ratio more conservative than that seen in comparable real world results (Warren and Whang, 1987, 1.16× shoulder width). Size judgments averaged 13% overestimation and were constant across distance. Consistent with prior work, distance was underestimated by about 20%. The utility of using affordance judgments in VEs and implications of the apparent discrepancy between scaling of size and distance judgments will be discussed.
This work was supported by NSF grant 0121084.