The influence of vision on computerized neurobehavioral test scores: a proposal for improving test protocols.

Computerized tests of neurobehavioral function are frequently administered in neurotoxicological studies with little attention given to the optical properties of test stimuli or to the vision of subjects. Yet many test stimuli are small or briefly presented, and test endpoints often involve short reaction times. Stimulus detection and reaction time are known to be strongly dependent upon stimulus luminance, contrast, and size, as well as on the subject's visual abilities. The current study assessed the influence of visual contrast sensitivity on Neurobehavioral Evaluation System 2 (NES2) test results in three data sets. Analyses indicated that vision was associated with up to 24% of the variance (Hand Eye Coordination test) in NES2 scores, even when visual acuity was normal, and that vision often influenced the significance of group differences. It is suggested that researchers measure the luminance, contrast, and size of test stimuli, the distance from the subject's eyes to the monitor, and the subject's visual contrast sensitivity. The measurement and control of stimulus parameters and the inclusion of visual function scores in analysis models could reduce the variability among computerized test scores both within and between studies. Models that assess the influence of vision on computerized test results may help to identify the CNS domains and specialized functions adversely affected by neurotoxicant exposures.