This study examined whether preference for a drug (diazepam or placebo) could be switched using conditioning procedures. During the first 4 sessions of Phase 1, 6 participants received 5 mg of diazepam or placebo under double-blind conditions. During the remaining 5 sessions of Phase 1, participants selected the drug they wished to receive. The first 4 sessions of Phase 2 were a replication of Phase 1, except that following ingestion of the drug, participants completed a computer task for which they could earn money. Payment for the computer task was lowest following ingestion of the drug they preferred in Phase 1 and highest following the drug they had avoided. Preference was reassessed during the last 5 sessions of Phase 2. Five of the participants preferred placebo in Phase 1 but diazepam in Phase 2. Subjective responses to the drugs also changed across the 2 phases.