Diagnosis of Schizophrenia by Rorschach Patterns

In the belief that the diagnostic validity of Rorschach methods would gain from formalizing the judgmental processes of clinicians, 14 score patterns were constructed from eight Rorschach variables, each pattern designed to characterize patients in one of three broad diagnostic categories, Schizophrenic, Borderline Psychotic, or “Other.” A random sample of 196 patients of a VA mental hygiene clinic was drawn from clinic files. The “pattern diagnoses” obtained by applying the pattern system to their Rorschach scores were found significantly related to diagnoses formulated by the examining psychologists and by the clinic intake staff. More patients diagnosed schizophrenic by the pattern method engaged in autistic behavior than patients with other pattern diagnoses. The diagnostic efficiency of the pattern method, measured in terms of ability to reach a definite diagnosis, proportion of correct diagnoses, and proportion of false positive errors, did not differ significantly from the diagnostic efficiency of the psychological examiners, and differed significantly from the psychiatrists' diagnostic efficiency only in producing more “indeterminate” diagnoses. The diagnostic validity of the patterns was considered to be established, but their usefulness as a clinical tool depends heavily on population base rates and the relative costs of false positive and false negative diagnostic errors.