Abstract
Background: In this study we sought to determine whether visual acuity (VA) or Pelli-Robson contrast sensitivity (CS) measured under a range of illumination conditions could better predict night-time driving performance than standard measures of VA. Experimental Design: 24 participants divided into 3 age groups drove around a closed road circuit under day and nighttime conditions. At night, headlight intensity was varied over 1.5 log units using ND filters. Subjects drove twice around the circuit at each light level: once for normal and once for a commentary condition where drivers had to report relevant targets including road signs, road obstacles, and pedestrians who wore retroreflective markings on either the torso or the limb-joints to create “biological motion.” Recorded measures included average speed and target recognition. Visual performance measures included VA and Pelli-Robson CS under 4 illumination conditions and dark-adapted pupil size. Results: Recognition performance of all participants was degraded in low light and these changes were more strongly predicted (r=0.18 to 0.63) by Pelli-Robson CS under standard photopic or reduced illumination conditions than by VA (r=0.09 to 0.24). Pedestrian recognition was greatly enhanced by marking limb-joints, a benefit that increased with driver age. Average speeds decreased with age and reduced light, but the latter effect was small. Conclusions: These findings confirm that visibility is seriously degraded during night driving and that, especially with pedestrians, the problem is greater for older drivers. This degradation in night-time visibility was better predicted by Pelli-Robson CS than VA.
Supported by Queensland University of Technology, Franklin & Marshall College and Queensland Transport.