Locus and extent of cerebral dysfunction in five closed head injured patients as inferred from positron emission tomography (PET) and computed tomography (CT) scan studies were compared with locus and extent inferences made in the same five patients using three additional independent sources of data. These data included: (1) clinical neurologic examination findings, (2) an extensive battery of neuropsychologic tests (including WAIS, Wechsler Memory, Halstead Reitan procedures), and (3) a comprehensive speech and language assessment battery (including the Boston Diagnostic Aphasia Examination (BDAE) and the Token Test). Neurologic, psychologic, and speech examinations were conducted within one week of the PET and CT studies. Three patients were studied at three months, one patient at six months, and one patient at two years after head injury. Follow-up CT scans were done 14 months later on four of five patients. This study supported the following major conclusions: the PET findings closely correspond with the site and extent of cerebral dysfunction inferences derived from the neurologic and behaviorial examinations but the CT findings did not; the follow-up CT scans, however, showed structural abnormalities (ie, encephalomalacia and atrophy) that were consistent with the PET, neurologic examination, and behavioral assessment findings; the study strongly supports the validity and predictive utility of neurobehavioral data which are directly pertinent to rehabilitation planning/programming in patients with closed head injury.