Xue Teng^1,2, Laurie M. Wilcox^1,3, Robert S. Allison^1,2; ¹Centre for Vision Research, York University, ²Department of Electrical Engineering and Computer Science, York University, ³Department of Psychology, York University

In virtual reality (VR) environments, our movements often differ from those in the equivalent physical world. One common discrepancy is visual gain which is a scaling difference between visual and kinesthetic motion. Previous studies have shown that observers reliably perceived the point of subjective stationarity (PSS) in a gain discrimination task during active self-motion. However, the PSS did not shift following prolonged adaptation to non-unity gain. Here we asked whether postural response adapts to gain manipulations. Three adaptation gains were tested in separate, counterbalanced blocks. Each block consisted of a 10-minute initial adaptation, followed by four test segments interleaved with three 2-minute top-up adaptation periods. During adaptation observers were immersed in a virtual room and continuously walked to grab objects and align them with corresponding markers at other locations. Their virtual motion was scaled by 0.67, 1 or 2 times their physical motion. During testing, quiet stance postural sway was recorded while the surrounding environment oscillated sinusoidally at 0.2 Hz over a peak-to-peak distance of 0.5 m, in either the front-back (N=18) or the left-right (N=14) direction. To isolate the visual perturbation effects, we also included a ‘stimulus absent’ condition in which observers viewed a dark HMD screen. Prior to starting, 60 seconds of baseline postural sway data was collected for both visual stimulus present and absent conditions. Results showed that postural sway in dark was larger after adaptation to a gain of 2 (in either motion direction). Further, power analysis at 0.2 Hz suggested that visually-elicited synchronous postural sway was larger under both non-unity gains along the front-back direction, suggesting that gain manipulations resulted in destabilization. Collectively these experiments suggest that gain manipulations produce adaptation in postural responses, while perceived stability does not shift. This dissociation suggests that postural recalibration to gain adaptation operates independently of perceptual mechanisms.

Acknowledgements: The authors wish to thank NSERC Canada for support under an Alliance Grant #ALLRP 570802-21, in partnership with Qualcomm Canada.