Enhancing Visual Exploration through Augmented Gaze: High Acceptance of Immersive Virtual Biking by Oldest Olds

Abstract

The diffusion of virtual reality applications dedicated to aging urges us to appraise its acceptance by target populations, especially the oldest olds. We investigated whether immersive virtual biking, and specifically a visuomotor manipulation aimed at improving visual exploration (augmented gaze), was well accepted by elders living in assisted residences. Twenty participants (mean age 89.8 years, five males) performed three 9 min virtual biking sessions pedalling on a cycle ergometer while wearing a Head-Mounted Display which immersed them inside a 360-degree pre-recorded biking video. In the second and third sessions, the relationship between horizontal head rotation and contingent visual shift was experimentally manipulated (augmented gaze), the visual shift being twice (gain = 2.0) or thrice (gain = 3.0) the amount of head rotation. User experience, motion sickness and visual exploration were measured. We found (i) very high user experience ratings, regardless of the gain; (ii) no effect of gain on motion sickness; and (iii) increased visual exploration (slope = +46%) and decreased head rotation (slope = -18%) with augmented gaze. The improvement in visual exploration capacity, coupled with the lack of intolerance signs, suggests that augmented gaze can be a valuable tool to improve the "visual usability" of certain virtual reality applications for elders, including the oldest olds.

Keywords: assisted living facility; augmented gaze; healthy aging; motion sickness; oldest old; technology acceptance; virtual biking; virtual reality; visual exploration.

Figure 5

Physical cyber-sickness symptoms as assessed through the SSQ differential score (post-session—pre-session) across the three gain levels. The dashed line indicates a lack of symptoms resulting from the virtual biking session. The maximal score is 235.62. Error bars are 99% confidence intervals of the mean.

Figure 6

Effective duration of virtual biking sessions (black data points) across the three gain levels. Also shown is the effective pedalling duration (gray data points). The dashed line indicates the maximal session duration. Error bars are 99% confidence intervals of the mean.

Figure 7

Visual exploratory activity across the three gain levels. Both the mean head rotation (absolute horizontal deviation from straight ahead, black data points) and the contingent mean visual rotation (grey data points) are illustrated. Error bars are 99% confidence intervals of the mean. The two dotted lines represent the condition in which visual exploration would be maximized, with no decrease in head rotation (grey), and the condition in which head rotation would be minimized, with no increase in visual exploration (black). Participants implemented a balanced, intermediate strategy. The dashed line indicates the natural condition (gain = 1.0), where the head and visual rotations coincide.

Figure 1

Augmented gaze. With the unitary gain, head movements and visual shifts are mapped 1:1 (i.e., the natural condition determined by head mechanics). When the visuomotor gain is higher, any given head movement produces a larger shift of the visual stimulus (visual amplification due to the augmented gaze condition). The red frame represents a hypothetical central region of the visual field, the solid arrow indicates the current head direction, and the dotted arrow represents the (larger) head movement that would have been required to produce that visual shift. Eye movements were not considered in this study; thus, the term augmented gaze refers to the visual amplification of head movements only.

Figure 2

A photograph of a virtual biking session. Used with permission.

Figure 3

Example of head position recordings (black traces) during virtual biking in the 3 gain conditions in one participant (age = 81). The grey traces show the corresponding visual rotation (when the gain is 1.0, the two traces coincide). Only horizontal rotation (yaw) is illustrated. LW = leftward, RW = rightward. To facilitate comparison, the visual traces have been depicted with inverted signs (normally, visual stimulus rotation moves in the opposite direction to head rotation.

Figure 4

User experience, as assessed through the global index, across the three gain levels. The global index was computed by averaging, for each participant, the satisfaction, motivation, sense of presence, and sense of safety ratings (open symbols, range 1–5). Gain 1 is the natural visuomotor gain, whereas gains 2 and 3 implement two-fold and three-fold augmented gaze, respectively. Error bars are 99% confidence intervals of the mean.