Viewpoint oscillation improves the perception of distance travelled in static observers but not during treadmill walking

Martin Bossard,Cédric Goulon,Daniel Mestre

doi:10.1007/s00221-020-05786-y

Abstract

Optic flow has been found to be a significant cue for static observers’ perception of distance travelled. In previous research conducted in a large-scale immersive display (CAVE), adding viewpoint oscillations to a radial optic flow simulating forward self-motion was found to modulate this perception. In the present two experiments, we investigated (1) whether the improved distance travelled perceptions observed with an oscillating viewpoint in a CAVE were also obtained when the subjects were wearing a head mounted display (HMD, an Oculus Rift) and (2) whether the absence of viewpoint oscillations during treadmill walking was liable to affect the subjects’ perception of self-motion. In Experiment 1, static observers performed a distance travelled estimation task while facing either a purely linear visual simulation of self-motion (in depth) or the same flow in addition to viewpoint oscillations based on the subjects’ own head oscillations previously recorded during treadmill walking. Results show that the benefits of viewpoint oscillations observed in a CAVE persisted when the participants were wearing an HMD. In Experiment 2, participants had to carry out the same task while walking on a treadmill under two different visual conditions simulating self-motion in depth: the one with and the other without the visual consequences of their head translations. Results showed that viewpoint oscillations did not improve the accuracy of subjects’ distance travelled estimations. A comparison between the two experiments showed that adding internal dynamic information about actual self-motion to visual information did not allow participants better estimates.

Highlights

During locomotion, spatio-temporal information is delivered by the pattern of optic flow
The results showed that the main effects involved were those of the block factor (F (7, 133) = 4.95, p < 0.001), the distance factor (F (4, 76) = 182.43, p < 0.001), and the optic flow factor (F (1, 19) = 22.87, p < 0.001)
No statistical differences were found between our two optic flow conditions, in terms of either the distance travelled estimates or the leaky path integrator model parameters. These results argue against an ecological explanation according to which a self-motion perception advantage might be expected to occur under oscillatory conditions—when visual, proprioceptive, efference copy, and vestibular information all match, while the visual information available under non-oscillatory conditions is in conflict with other dynamic information generated by body and head movements

Summary

Introduction

Spatio-temporal information is delivered by the pattern of optic flow (dynamic visual information, Gibson 1950). Optic flow information is known to play a key role in many aspects of self-motion perception, including distance travelled (Campos et al 2010; Frenz and Lappe 2005; Lappe et al 2007; Redlick et al 2001), speed of selfmotion (Banton et al 2005; Larish and Flach 1990), heading (Warren and Hannon 1988), and time-to-collision (Lee and Lishman 1975) By dissociating these cues from dynamic information of other kinds, many authors have focused on static observers’ ability to use optic flow alone. The results of these studies clearly showed that humans are able to a previously seen distance accurately by walking the same distance, even when they are deprived of vision (Elliott 1986; Fukusima et al 1997; Loomis et al 1992; Mittelstaedt and Mittelstaedt 2001; Rieser et al 1990; Sun et al 2004; Thomson 1983)

Methods

Results

Conclusion