Artikel
The effect of eye and head position on reading speed in a simulation of prosthetic vision
Suche in Medline nach
Autoren
Veröffentlicht: | 10. Dezember 2019 |
---|
Gliederung
Text
Objective: Visual scanning is important in perceiving the world, but is challenging for artificial vision. Two approaches have resulted, devices which adapt stimulation based on gaze position, so that the point of regard can be steered with eye motion, and those that do not, and are usually steered by head motion. As the visual pathway is largely retinotopic, we hypothesized that subjects would perform better on perceptual tasks under artificial vision with gaze-steering than with head-steering.
Materials and methods: We used a simulation of artificial vision that could switch between compensation modes. Normal, sighted subjects (n = 23) read MNREAD sentences under conditions of Full-gaze compensation (including eye position), and Head-only viewing (ignoring eye position) in a simulation with 2000 phosphenes.
Results: Under Full-gaze compensation, subjects demonstrated a mean acuity of logMAR 1.0, but were nearly unable to perform the task under Head-only compensation despite interventional coaching. Eye movements under Head-only were highly interfering, as phosphenes did not respond in an intuitive way. At the largest text size, logMAR 1.3, subjects read with 100% accuracy at 50% of normal reading speed under Full-gaze, but below 15% accuracy at less than 5% of normal speed under Head-only.
Discussion: Gaze-compensated prostheses are likely to produce substantially better outcomes than devices not sensitive to eye movements.
Acknowledgments: Fulbright Foundation, Haseotis Foundation, and William Wood Foundation.