Rats implanted with a monopolar electrode in the medial mesencephalon and trained to press a lever to self-administer trains of stimulating pulses were tested under a continuous (CRF) and a fixed interval (FI-1s) schedule of reinforcement before and after systemic pimozide injection (0.35 mg/kg). As previously shown, this dose of pimozide caused reductions in maximal response rates and in the rewarding effectiveness of the stimulation. The magnitude of the attenuation in rewarding effectiveness was significantly greater under CRF than under FI conditions, an effect that can be attributed to the decrease in the number of earned rewards concomitant to a decrease in response rates. The present results show that the tighter control over reward density provided by the use of a Fl schedule of reinforcement reduces the probability of artifactually measuring increases in reward thresholds following treatments that suppress response rates.