Journal of Vision - Does perceptual-motor recalibration of locomotion depend on perceived self motion or the magnitude of optical flow?, by Thompson, Mohler, & Creem-Regehr

Volume 5, Number 8, Abstract 386, Page 386a

doi:10.1167/5.8.386

http://journalofvision.org/5/8/386/

ISSN 1534-7362

Does perceptual-motor recalibration of locomotion depend on perceived self motion or the magnitude of optical flow?

William B. Thompson	School of Computing, University of Utah, USA
Betty J. Mohler	School of Computing, University of Utah, USA
Sarah H. Creem-Regehr	Department of Psychology, University of Utah, USA

Abstract

The perceptual-motor calibration of human locomotion can

be manipulated by exposure to an environment in which the

visual flow associated with self-motion is altered relative

to biomechanical walking speed (Rieser, Pick, Ashmead and

Garing 1995, JEP:HPP). An open question remains as to whether

this recalibration is based on perception of the speed of

movement through the world or on the magnitude of optic flow

itself. We addressed this issue using a treadmill-based

virtual environment in which we could independently vary

actual walking speed and the simulated visual experience

of moving down a hallway. The hallway consisted of textured

walls and textureless floor and ceiling, so that visual flow

information was only available from the walls. Subjects were

exposed to one of two conditions. Actual walking speed was

1.2m/s in both cases. In one condition, visual information

corresponded to movement down a long hallway at a speed

one third less than the biomechanical rate of walking. In

the second condition, the visual information corresponded to

movement three times faster than in the first condition, down

a hallway that was three times larger. Because the scale of

the space was increased by the same amount as the increase

in velocity through the space, the magnitude of optic flow

remained essentially constant, though flow due to the walls

moved upward in the visual field. Perceptual-motor calibration

was evaluated by having subjects walk blindfolded to previously

viewed targets at 6, 8 and 10m before and after 10 minutes of

walking on the treadmill. For the visually slower condition,

subjects increased the distance they walked by an average of

10% between the pre and post tests. For the visually faster

condition, subjects decreased the distance they walked by

an average of 3%. These differences demonstrate that the

recalibration depended at least in part on visual perception

of the speed of self-movement, not just on the magnitude of

optic flow.

This work was supported by NSF grants IIS-00-80999 andIIS-01-21084.