Citation

Hills BL, Burg A. Transp. Res. Rec. 1978; 681: 47-50.

Copyright

(Copyright © 1978, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Vision standards for driver licensing require not only the selection of valid visual characteristics to be tested but also the establishment of valid cutoff scores as criteria for passing or failing. To date there has been virtually no research into the latter problem, and this study was conducted with this need in mind. This paper summarizes the major findings of the study and is taken from a more detailed report. The study explored in depth the implications of Burg's data for driver-vision standards and concentrated on determining whether certain subgroups of the driving population displayed stronger relations between vision and driving than did others. Preliminary work suggested that analysis of older drivers rather than of those with poor vision was most likely to show these stronger relations. Therefore, in the main analyses, the sample was divided into four age groups: under age 25, ages 25 to 39, ages 40 to 54, and over age 54. In summary, it must be said that, as a basis for vision standards that are valid in terms of potential accidents saved, the tests studied must be regarded as disappointing. The failure to find a direct relation between poor visual performance and high accident rate for young and middle-aged drivers has been consistent throughout the study, and, for the over 54 age group, the relations obtained are significant but weak. The ability of these tests to identify drivers likely to have accidents--without paying an unacceptably high penalty in the rejection of good drivers--remains questionable. These findings lend support to current attempts to find perceptual tests of visual performance that are much better accident predictors than the largely classical sensory tests of vision studied here. (Tests of contrast sensitivity, movement perception, and hazard perception are among those currently being examined, and it is recommended that investigation of other stimulus conditions for the promising glare-recovery test be carried out.) It should be stressed that the significant relations found for older drivers may not be causal. A driver's visual performance in this age range may merely reflect his or her "effective" (or phenomenal) age, and some other factor such as deterioration of the brain's central processing capacity may be the fundamental cause of increased accident rates. Thus, improving a driver's visual performance may not improve his or her accident rate; however, even if a measure of visual performance is not causally related to accident rate, and predictive power it may have could still be valuable for the purposes of screening or visual standards.