Visual, Perceptual, and Cognitive Measures as Predictors of Collision Detection in Older Drivers

Abstract

This experiment examined whether visual, perceptual, or cognitive measures predicted the ability to detect vehicle collisions in intersections. Sixty subjects, comprised of three age groups balanced by gender, were presented dynamic intersection approaches in a part-task driving simulator. The subjects were asked to project the forward progress of crossing traffic and to indicate whether any of the crossing vehicles would conflict with their vehicle. Independent variables included visual, perceptual, and cognitive test batteries. Dependent variables included accuracy in collision detection and error type. Results showed that all three batteries predicted accuracy, but that the perceptual battery was the most predictive for each age group.