Recent evidence suggests that the placebo effect is mediated by the dopaminergic reward mechanisms in the human brain and that it is related to the expectation of clinical benefit. On the basis of this theory, we propose some criteria for the proper investigation of the placebo effect, and review the evidence for a placebo effect in Parkinson's disease, depression, pain, and other neurological disorders. We also discuss the evidence for the use of placebos in long-term substitution programmes for the treatment of drug addiction.