Theoretical reliability of visual evoked response-based acuity determinations.

The process of determining acuities from visual evoked responses (VER's) was computer modeled. Based on amplitude variability data from earlier studies, the reliability and validity of the acuity determinations for normal subjects were assessed. For subjects who had poor quality VER data [low signal-to-noise (S/N) ratios], considerable variability and an artifactual shift toward elevated (better) acuity determinations were found.