Abstract
Ishihara pseudoisochromatic plates are one of the best screening tools for red-green colour vision deficiencies. Although a majority of persons with normal colour vision read all plates correctly, a significant proportion makes mistakes. The purpose of this study was to obtain results for normal trichromats reading the Ishihara plates and analyse the misreading responses to seek clinical implications. A sample of 249 (161 female) was tested with the Ishihara pseudoisochromatic plates. The number and nature of errors were recorded and typical errors, those that observers with abnormal colour vision were expected to make, were distinguished from other kinds of error. Out of 249 normal trichromats, 111 individuals (45 per cent) misread at least one plate. Females made up to six total errors and males up to five total errors. When only typical errors were counted, all the normal trichromats made two or fewer errors. There was no significant gender difference for either total or typical errors. It is suggested that clinicians count only typical errors when administering the Ishihara test. Using a criterion of no more than two typical errors for a diagnosis of normal colour vision could improve the specificity and sensitivity of the test.