Psychologists and psychiatrists are regularly used during civil commitment and other hearings to prognosticate dangerousness for the courts. In this process, the judiciary has shown almost complete deference to the recommendations of forensic specialists despite a number of early studies that did not support a clinician's ability to predict violence. This study examined if the Cognitive Control Battery (CCB) can contribute to discrimination in a model that includes known predictor variables and if the CCB can discriminate violent juveniles based upon actual episodes of violence. A stepwise discriminant analysis was used on a calibration sample (N=175) of dangerous and non-dangerous juvenile inpatients. Forty-five demographic, psychosocial, and cognitive variables were employed. The resulting statistical model was cross-validated on the remainder of the sample (N=175). Results indicated that 58.6% of the cross-validation sample was correctly classified. Moreover, the CCB demonstrated that dangerous youths processed aggressive/neutral stimuli differently than non-dangerous youths. Results did not suggest that cognition (or type of cognitive style) could be viewed as a cause of violence. Recent advances in cognitive psychology (e.g., the CCB) may assist policymakers, judges, and forensic clinicians to make more informed choices concerning dangerous juvenile inpatients. (ABL)