Different Head-Sway Responses to Optic Flow in Sitting and Standing With a Head-Mounted Display

Abstract

We investigated postural responses (head displacements) and self-motion perception (vection) to radial and lateral optic flows while sitting and standing by using a head-mounted display. We found that head displacement directions varied across postures. In the standing posture, radial optic flow generally produced the opposed head displacement against the perceived vection direction, consistent with the literature; however, in the sitting posture, the optic flow generally produced the following head displacement in the vection direction. In the standing posture, responses were evident soon after the onset of the optic flow presentation but became less clear in the latter half of a trial. The results, while less clear for lateral flows, were similar for both flow types. Our findings suggest partially distinct processes underlying vection and postural control.

Keywords: HMD; optic flow; postural control; self-motion; sitting; vection; virtual reality.

FIGURE 1

Schematic structures of the three-dimensional cloud (upper panels) and optic flow stimulus generated by the camera movement in each structure (lower panels). The upper left panel shows the eye representing the virtual camera moving through the central hole, generating radial optic flow (lower left panel). In the upper right panel, the eye moved laterally along the half-cylinder, producing lateral optic flow (lower right panel). Notably, the bottom panels show dots in several successive frames to illustrate motion.

FIGURE 2

Mean time series of head displacement relative to the base position throughout the optic flow presentation for each posture. The top and middle panels plot the sitting and standing posture results, respectively. The left panels plot the findings for the expansion/contraction optic flow, and the positive and negative values along the vertical axis represent anterior and posterior displacements, respectively. The right panels plot the results for leftward/rightward optic flow, and the positive and negative values represent rightward and leftward displacements, respectively. Ribbons represent 95% confidence intervals obtained through the bootstrap method. Shaded regions in the top and middle panels highlight the period where a significant cluster was found with the cluster-based permutation test for expansion–contraction and rightward–leftward comparisons for each posture. The bottom panels highlight the period where a significant cluster was found for a two-way interaction between optic flow and posture in each radial and lateral condition.

FIGURE 3

Vection results: (A) mean vection ratings and (B) histograms of perceived vection direction for each posture and optic flow type (left panels: radial optic flow [expansion/contraction], right panels: lateral optic flow [rightward/leftward]). Error bars show standard errors of the mean across participants.