An attempt to cross-validate two actuarial systems for neuropsychological test interpretation.

Abstract

The utility of the Halstead-Reitan Neuropsychological Test Battery in making neurological diagnostic inferences is well documented, but a persistent problem is the need to rely on intuitive and subjective methods of test interpretation. This study attempted cross-validation of 2 objective, computerized programs designed to determine the presence, location and process of brain lesions using scores from the Halstead-Reitan battery and the Wechsler Adult Intelligence Scale. Human subjects (150) with structural brain lesions and 100 normal controls participated. Both test interpretation programs classified subjects as brain damaged and not brain damaged with an acceptable degree of accuracy. The programs were much less accurate in making finer diagnostic inferences concerning the localization or process of structural lesions. Differences between the current findings and those of the original validation studies may be due to the sampling procedures and neurological criteria used. Ways in which these programs might be improved are discussed.