It has been shown that infusion of certain neuropeptides into the ventral tegmental area (VTA) results in increased motor activity and enhanced dopamine turnover in the nucleus accumbens. In the present experiments, substance P (SP), neurotensin (NT), d-ala-metenkephalin (DALA) and morphine sulfate (MS) were injected bilaterally into the VTA and their effects on conditioned reinforcement were assessed. These effects were compared with infusion of amphetamine into the nucleus accumbens, which has previously been shown to strongly enhance responding for conditioned reinforcers. For these experiments, hungry rats were trained to associate a compound stimulus (light and click) with the presentation of food. In the test phase, responding on one lever (CR lever) resulted in the presentation of the stimulus but no food. Responding on the other (NCR lever) had no consequences. Different groups of animals received microinjections (0.5 microliter, bilaterally) of SP (0, 0.03, 0.3, 3.0 micrograms), NT (0, 0.025, 0.25, 0.5 microgram), DALA (0, 0.01, 0.1, 1.0 microgram) or morphine (0, 0.025, 0.25, 2.5 micrograms) into the VTA. SP infusion into the VTA resulted in a small increase in responding which was not selective for the CR lever. NT, DALA and morphine had no effect on responding for conditioned reward. In contrast, amphetamine (0, 0.2, 2.0, 20 micrograms) injected into the nucleus accumbens markedly enhanced responding for conditioned reward. These findings suggest that stimulation of the mesolimbic system at the level of the DA cell bodies, which induces a small increase in DA turnover, is not sufficient to potentiate responding for conditioned reward. On the other hand, an important requirement for potentiation may be excessive release of dopamine in the nucleus accumbens.