Abstract
There is a well-known perceptual foreshortening in the depth dimension, even for relatively near distances (e.g., 3–10 m). However, when observers view two markers on the ground and attempt to walk the distance between them with their eyes closed, there is little evidence of foreshortening (Loomis et al., 1992). The present study explored this apparent discrepancy by using imagined walking of exocentric intervals. In an outdoor setting, observers binocularly viewed frontal and depth intervals of 137 cm, presented 3–10 m away; they then attempted to physically pace out the intervals with eyes closed, or imagine performing the same task. Walked distance (real walking only) and walking time (both real and imagined walking) were measured. For comparison, participants also performed a visual matching task in which they attempted to create a depth interval that matched a frontal interval. Results from this matching task showed that depth intervals were made about 40% longer than the corresponding frontal intervals, indicating that these observers showed large depth foreshortening. On the other hand, although walking time was significantly underestimated in imagined walking compared to real walking, both response modes were much more resistant to perceptual depth compression. These results suggest that a similar spatial representation underlies real and imagined walking, which is largely isotropic and dissociable from the anisotropic spatial representation underlying visual perception of exocentric intervals. However, the underestimation of walking time in imagined walking indicates some functional differences between real and imagined walking, conceivably attributable to the lack of self-motion signals during imagined walking.
Research supported by NIH 5R01NS052137-03.