Research Project
10298110
Project Summary Hemianopia (the loss of half the field of vision on the same side in both eyes) is a severe visual consequence of stroke and traumatic brain injury. Resuming driving after the onset of hemianopia is an important goal to maintain independence and quality of life. People with hemianopia are permitted to drive in about 18 states. However, in a retrospective analysis of crash records, drivers with hemianopia were found to have a higher risk for crashes than drivers with a full field of vision. Intersections are especially challenging for drivers with hemianopia as a wide field (up to 180°) has to be checked for potential hazards, requiring head as well as eye movements. If drivers with hemianopia do not scan (look) toward the side of the hemianopia, or do not scan far enough, then a hazard that appears within a field loss area might never be seen or might be seen too late to make a safe driving response. Although there is strong evidence of inadequate scanning by people with hemianopia in simulated driving, little is known about their scanning behaviors and the extent to which inadequate scanning contributes to collision risk in real-world driving. The overall goals of this research are to (1) Investigate inadequate scanning and factors contributing to near-collision incidents (a proxy for actual collisions) in naturalistic driving; (2) Evaluate the efficacy of a novel approach to scanning training to address intersection scanning deficits; and (3) Evaluate whether scanning training in the simulator transfers to real- world driving. To address the first goal, in Aim 1, drivers with hemianopia and drivers with normal vision will complete an extended period of driving with a video-recording system in their own vehicle. This will provide important data to better understand factors contributing to inadequate scanning in naturalistic driving and the role of inadequate scanning as compared to other at-risk behaviors in near-collision incidents. To address the second and third goals, randomized controlled clinical trials will be conducted to evaluate the efficacy of the scanning training for former drivers with hemianopia (Aim 2) and current drivers with hemianopia (Aim 3). Scanning training will be given in a simulator. In Aim 2 the effects of the training on intersection scanning and responses to hazards will be evaluated in the driving simulator. In Aim 3, the effects of the training will be evaluated through recordings of scanning behaviors in naturalistic driving before and after training, which will enable an evaluation of whether training in the simulator generalizes to improvements in scanning in real-world driving. The program of research will also examine whether scanning in the virtual environment of a driving simulator is representative of scanning in real-world driving as a first step toward validating the driving simulator as a tool for both training and evaluation of scanning of at-risk drivers. The proposed research has the potential to make a significant contribution toward improving road safety by improving scanning of drivers with hemianopia, thereby improving detection of hazards and reducing their collision risk at intersections.
