Item

Abstract

A variety of sources of sensory information (e.g., visual, inertial and proprioceptive) are available for the estimation of walking speed. However, little is known about how they are integrated. We present a series of experiments, using a 2-IFC walking speed judgment task, investigating the relative contributions of the inertial and proprioceptive information. We used a circular treadmill equipped with a motorized handlebar, to manipulate inertial and proprioceptive inputs independently. In one experiment we directly compared walking-in-place (WIP) and walking-through-space (WTS). We found that WIP is perceived as slower than WTS. The WIP condition creates a special conflict situation because the proprioceptive cue indicates motion whereas the inertial cue indicates an absence of motion through space. In another experiment we presented a range of conflicts by combining a single proprioceptive input with different inertial inputs. We found that the inertial input is weighted more heavily when it indicates a faster walking speed than proprioception. Conversely, it receives less weight if it indicates a lower speed. This suggests that the inertial cue becomes more reliable with increasing velocity. Our findings show a more important role for inertial information in the perception of walking speed than has previously been suggested in the literature.