Abstract
Purpose :
Drivers with homonymous hemianopia (HH) may compensate for their hemifield loss by scanning. Prior research has mostly quantified scanning in terms of overall gaze (combination of eye and head movements) or just head movements, failing to account for potential differences between eye-only scans (scans without any head movement) and eye-with-head scans (scans with head movement). We evaluated whether drivers with HH demonstrated compensatory behaviors in eye-only and/or eye-with-head scans when approaching intersections and examined driver and environmental factors affecting scanning behaviors.
Methods :
26 drivers (7 with left HH (LHH), 6 with right HH (RHH), and 13 age-matched with normal vision (NV)) drove in a high-fidelity driving simulator while eye and head movements were recorded. Scanning was analyzed on approach to 36 intersections in an urban environment with cross-traffic. The scans were classified into eye-only and eye-with-head scans. The effects of driver attributes and intersection characteristics were examined using linear mixed models and analysis of variance.
Results :
Per intersection, drivers made 5.4 eye-only scans of magnitude 11.4°, and 2.5 eye-with-head scans of magnitude 42.1°. Scan magnitudes increased as drivers came closer to the intersection (p < 0.001) and decreased with increasing age of the driver (p = 0.023). RHH drivers made more eye-only scans than NV drivers (6.64 vs 4.58, p = 0.06) while LHH drivers tended to make more eye-with-head scans than NV drivers (2.82 vs 2.42, p = 0.11). HH drivers made significantly more scans to the blind than seeing side (eye-only: 2.62 vs 2.27, p = 0.001; eye-with-head: 1.26 vs 1.09, p < 0.001), suggesting compensatory behavior. Overall, there was no evidence of compensatory behaviors in scan magnitudes for either eye-only or eye-with-head scans: HH drivers made smaller (not larger) scans to the blind than seeing side (eye-only: 10.4° vs 11.5°, p < 0.001; eye-with-head: 37.7° vs. 40.7°, p = 0.03).
Conclusions :
HH drivers showed evidence of compensatory behaviors in scan numbers, making more scans than NV drivers and more eye-only and eye-with-head scans to the blind than seeing side, but no evidence of compensatory behaviors in scan magnitudes. Smaller eye-only and eye-with-head scans to the blind side may be due to lack of guidance from peripheral vision on that side.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.