Distinguishing between brain-damaged and schizophrenic patients remains a significant clinical problem. However, substantive and methodological difficulties have led to discouragement and consequent loss of interest in this area of differential diagnosis. Several models for making this distinction were proposed: cognitive, empirical-psychometric, attentional motivational, neuropsychological and information processing. In evaluating these models, three persistent methodological problems were identified: diagnostic inaccuracy, inadequate sampling, and difficulty in interpretation of the subject's performance. Various suggestions were made for resolving these difficulties through such means as equating samples for level of deterioration, utilizing modern, objective diagnostic techniques in the selection of samples, comparing subtypes of brain damage and schizophrenia, and using test procedures that have been shown to assess dispositional characteristics of the conditions under study, rather than their epiphenomena. It was suggested that recent substantive findings, mainly those involving neurological deficit in schizophrenia and detection of major behavioral differences among various subtypes of schizophrenia and brain damage, could be used in attempts to resolve the differential diagnostic problem under discussion. It was suggested that resolution of the methodological problems posed in combination with new substantive data concerning schizophrenic and brain-damaged patients may set the stage for a new start for investigating an old problem.