Virtual street crossing and scanning behavior in people with hemianopia: A step toward successful crossings.

Individuals with homonymous hemianopia (HH) may benefit from adopting compensatory crossing and scanning strategies to successfully cross streets. In this study, we explored the effect of HH on street crossing outcomes, crossing behavior and scanning behavior in a virtual environment. Individuals with real HH (N = 18), unimpaired vision (N = 18), and simulated HH (N = 18) crossed a virtual street displayed through a head-mounted display. Virtual cars approached from both directions, traveling at a speed of either 30 or 50 km/h. Participants' crossing and scanning behaviors were recorded and analyzed across groups and the two car speeds. Although individuals with real and simulated HH took more time to cross compared to individuals with unimpaired vision depending on the car speed, the number of collisions and time-to-contact after crossings did not differ between groups. We observed no differences in the selection of car gaps, crossing initiation, and scanning behavior between groups. Our findings suggest that individuals with real and simulated HH align their crossing behavior to their visuomotor capabilities by using varying compensatory strategies. HH did not alter scanning behavior before crossing a virtual street. Despite its current shortcomings, virtual reality holds promise for street crossing research and rehabilitation.