Abstract
Synopsis Three problems with the dopamine hypothesis of major psychosis are pointed out: the long time-course of neuroleptic therapy; the absence of tolerance to the antipsychotic effects of neuroleptic drugs, or of a supersensitivity psychosis on drug withdrawal; and the absence of potent psychotogenic properties in the direct dopamine agonists. A resolution of these paradoxes is suggested relying on a role for dopamine in learning processes at a relatively high (cognitive) functional level. The hypothesis proposed is also used to explain the origin of some of the more distinctive psychotic symptoms.