Abstract
Purpose :
Multiperiscopic peripheral prisms (MPP) are a novel, high-power field expansion device for homonymous hemianopia (HH). In preparation for a randomized controlled multisite clinical trial, we conducted a pilot study comparing MPPs to commercially available Fresnel peripheral prisms (FPPs). We tested the hypothesis that detection rates for colliding pedestrians approaching on a 40° bearing angle (greatest collision risk when walking) would be higher with 100Δ oblique MPPs (≈42° expansion) than 65Δ oblique FPPs (≈30° expansion).
Methods :
Subjects with HH (n=5) completed a virtual reality pedestrian collision detection test, without prisms, with FPPs, and with MPPs. The test, presented as a video on a large screen (104° H × 82° V), simulated walking through a busy shopping mall with multiple pedestrians moving in various directions. In each trial, one of the pedestrians (the target) could be on a collision course with the subject (constant eccentricity with looming), approaching from either the left or the right on a bearing angle of either 20° or 40°, or there could be no colliding pedestrian. A secondary task (calling out numbers in a string of letters) was used to keep gaze and attention generally toward the forward path but still allow scanning. Subjects attended multiple sessions. Rates for correct detection of colliding pedestrians are reported for data aggregated across sessions.
Results :
For 20° blind-side pedestrians, median (range) detection rates were 3% (0 to 75%) without prisms but improved to 100% (50 to 100%) with FPPs and 84% (75 to 100%) with MPPs. For 40° blind-side pedestrians, detection rates were 0% (0 to 56%) without prisms, only 25% (0 to 59%) with FPPs, but improved to 88% (50 to 100%) with MPPs (significantly higher than FPP and no prisms, p=0.04). Seeing side detection rates were 100% in all conditions.
Conclusions :
Consistent with our hypothesis, detection rates for 40° blind-side colliding pedestrians were significantly higher for MPPs than FPPs but did not differ for 20° pedestrians. Although the secondary task kept gaze in the forward direction, participants could scan, accounting for blind-side detection without prisms and limited detection of 40° blind-side pedestrians with FPPs. The results are encouraging and will be used to confirm sample size calculations for the multisite clinical trial.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.