Shopping with sight loss: Using a virtual reality shopping task to systematically quantify the impact of a scotoma size and eccentricity on everyday quality of life

Abstract

Purpose : To objectively quantify the impact of scotoma size and location on vision-related Quality of Life (vrQoL). To this end, we have developed a virtual reality (VR) simulator that allows participants to perform everyday tasks while experiencing gaze-dependent scotomas of variable size and eccentricity.

Methods : Normally sighted subjects (n=8 at time of writing; additional data collection ongoing) were asked to locate various products in a virtual supermarket. Participants physically walked through the supermarket (body tracking). They attempted to locate various items listed on a virtual shopping list and used their hands to pick up products and place them in a virtual trolly. Each participant experienced five different gaze-contingent scotomas of varying size and eccentricity (within-subject design), simulated using the VARID open-source toolkit (varid.co.uk). Outcome measures included time spent searching for items and reading the shopping list; number of head, eye, and body movements; and numbers of errors (e.g., items knocked over).

Results : Participants spent longer reading the shopping list in the presence of a 9° radius scotoma at 9° eccentricity, versus no impairment (mean {95%CI} increase = 64 {5 – 122} secs; paired t-test: t₇=2.6, P=.038). Participants also spent longer searching for items in the aisle (mean {95%CI} increase = 66 {12 – 120} secs; t₇=2.9, P=.023). Reducing the scotoma to 5° radius reduced this deficit by approximately 50% (mean {95%CI} increase vs. no impairment = 30 {5 – 56} secs; t₇=2.8, P=.026). Further analyses will also be presented, including the effects of more gradual changes in scotoma size and location, and outcome measures relating to head, eye, and body movements.

Conclusions : Relatively modest changes in the size and location of a scotoma have a measurable effect on a person’s ability to perform an everyday visual guided task. More generally, this work demonstrates the ability of VR environments to objectively quantify vrQoL. Details of the VR environment and simulated scotomas, including how to obtain them, will also be presented.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

 

Photograph of one of the authors (L) holding two products and the view from the right monitor inside the headset (R). The scotoma is located over the label of the orange item held by the author, making the label unreadable but still showing the shape of the object.