Several lines of evidence indicate that placebos produce analgesia through the activation of endogenous opioid systems. Recently, we showed that placebos may also produce respiratory depressant responses, a typical side-effect of narcotics, when a subject had a prior experience of respiratory depression in the course of narcotic treatment. In the present study, we report that the placebo respiratory depression can be induced after repeated administrations of the partial opioid agonist buprenorphine. The placebo respiratory depressant effect that resulted from the buprenorphine conditioning was completely blocked by a dose of 10 mg of naloxone, indicating that it was mediated by endogenous opioids. These findings show that placebos act, via the activation of opioid receptors, not only on pain mechanisms but on the respiratory centres as well, thus mimicking a typical side-effect of narcotics. In addition, the experimental procedure we used did not produce any expectation of respiratory depression and, similarly, the subjects did not notice any sign of respiratory discomfort. Thus, the placebo respiratory depression elicited in the present study cannot be explained on the basis of cognitive or motivational mechanisms. Rather, it appears to be a sequence effect due to learning, thus suggesting a conditioning mechanism mediated by endogenous opioids.