Researchers examining the stimulus properties of MDMA in rats conducted a 3-choice drug discrimination study with 1.5 mg/kg MDMA and 0.08 mg/kg LSD as training drug. Lower doses of MDMA and LSD, 0.25-2 mg/kg D-amphetamine, and (+)-0.25-2 mg/kg fenfluramine served as test drugs. Sessions with test drugs were only conducted after rats demonstrated at least 80% correct response for training drugs. Rats responded to lower doses of MDMA (0.375 and 0.75 mg/kg) as if they were MDMA, and they responded to lower doses of LSD (0.02 and 0.04 mg/kg) as if they were LSD. D-amphetamine produced dose-dependent increases in MDMA-lever response and decreases on saline-lever response without ever producing complete stimulus generalization. Fenfluramine produced a dose-dependent increase in MDMA-appropriate response at 0.25, 0.5, 1 and 1.5 mg/kg, and completely substituted for MDMA at 2 mg/kg. Only 1 mg/kg (+)-fenfluramine produced 20% LSD-appropriate responding, and all other doses failed to produce any response on the LSD-appropriate lever. When coadministered along with 1.5 mg/kg MDMA, the 5HT2A antagonist MDL100907 failed to abolish MDMA appropriate response, whereas MDL100907 abolished LSD-appropriate response after 0.08 mg/kg LSD. The dopamine D2 receptor antagonist haloperidol (0.1-0.4 mg/kg) failed to abolish MDMA-appropriate response after MDMA, but did reduce overall response rate. These results suggest that 5HT2A activity plays a far greater role in producing LSD stimulus cues in rats than in producing MDMA cues, and that D2 receptor activation has only a small role in the stimulus characteristics of both MDMA and LSD. The authors compare current findings with a previous 3-choice drug discrimination study with D-amphetamine and MDMA as training drugs (Goodwin and Baker 2000) wherein LSD substituted for MDMA. Taken together, these findings suggest that drug discrimination studies with stimulants, psychedelics and MDMA-like drugs are sensitive to the training method (3-choice versus 2-choice) and training drugs used. It should be noted that while no direct comparisons of these three compounds have been made in humans, anecdotal reports and examination of the literature suggests that humans do not consider racemic fenfluramine or (+)-fenfluramine to be MDMA-like. Study findings are only in partial agreement with human drug interaction studies (Liechti et al. 2000; Liechti and Vollenweider 2000). While these human studies failed to block most subjective effects of MDMA with the D2 receptor antagonist haloperidol or the 5HT2A antagonist ketanserin, both drugs reduced specific effects of MDMA, such as euphoria (with haloperidol) or altered perceptions (with ketanserin).