Objective: The purpose of this study was to assess the functional significance of ventricle-brain ratio (VBR) in terms of how it might affect sensitivity to cocaine, an indirect dopamine agonist. Method: Relationships between VBR and subjective responses to acute intravenous cocaine hydrochloride were examined in 20 male poly drug abusers. Tests were performed in conjunction with positron emission tomography scans to measure cerebral glucose metabolism. Results: Subjective measures of effects of cocaine, including self-report ratings of intensity of the drug effect, scores on the morphine-benzedrine scale of the Addiction Research Center Inventory, and several items on visual analogue scales, correlated negatively with VBR. VBR also differed significantly among subjects who were grouped according to scores on items ("rush" and "crash") of the Cocaine-Sensitive Scale (larger VBR in subjects with weaker responses). VBR was not correlated with cocaine-induced changes in cerebral metabolic rates for glucose. Conclusions: Relative insensitivity to the subjective effects of cocaine in poly drug abusers with ventricle enlargement suggests that ventriculomegaly may reflect changes in periventricular brain regions that mediate these effects of cocaine.