Rats with unilateral depletions of neostriatal dopamine display increased sensitivity to dopamine agonists estimated to be 30 to 100 x in the 6-hydroxydopamine (6-OHDA) rotational model. Given that mild striatal dopamine D-2 receptor proliferation occurs (20-40%), it is difficult to explain the extent of behavioral supersensitivity by a simple increase in receptor density. This study was designed to investigate the quantitative aspects of the rotational behavior model utilizing constrained non-linear curve fitting routines. A dose-response curve for the rotational response arising from apomorphine stimulation of the normosensitive striatum was obtained in animals bearing unilateral lesions of striatal efferents (predominantly the striato-nigral pathway as previously described). After the control dose-response experiment, rats received a dopamine- (DA) depleting lesion in the contralateral hemisphere. In one group, 6-OHDA was infused into the medial forebrain bundle (MFB), a placement which is common in the literature and is known to deplete DA in both the striatum and nucleus accumbens. In a second group of rats, 6-OHDA was infused into the globus pallidus at a site which depletes caudate DA, but leaves n. accumbens DA relatively intact. The two experimental groups were tested in identical apomorphine-induced rotation dose-response experiments. The ED50's of the MFB- and caudate-lesioned rats were reduced by 36 and 5.8 fold, respectively, as compared to the control dose-response curve. The MFB and caudate lesions depleted striatal DA and produced a 30 and 36% increase in striatal D-2 binding sites, respectively. Modeling the behavioral and biochemical data with the null model for receptor occlusion indicated that increased striatal D-2 receptor density could account for the magnitude of behavioral supersensitivity in neither the MFB-lesioned group, nor even in the caudate-lesioned group. Thus simple up-regulation or D-2 receptors is unlikely to account for supersensitization as measured in the rotational model. Further, we suggest that quantitative modeling of such hypotheses is a valuable experimental technique for assessing relationships between biochemical and behavioral variables.