RATIONALE: Experimental research has shown that 3,4-methylenedioxymethamphetamine (MDMA) can improve some psychomotor driving skills when administered during the day. In real life, however, MDMA is taken during the night, and driving may likely occur early in the morning after a night of "raving" and sleep loss. OBJECTIVES: The present study assessed the effects of MDMA on road-tracking and car-following performance in on-the-road driving tests in normal traffic. METHODS: Sixteen recreational MDMA users participated in a randomized double-blind placebo-controlled four-way cross-over design. They received single, evening doses of 0, 25, 50, and 100 mg MDMA on separate occasions. Actual driving tests were conducted in the evening when MDMA serum concentrations were maximal and in the morning after a night of sleep loss. RESULTS: The primary measure of driving, i.e., standard deviation of lateral position (SDLP, a measure of weaving) was significantly increased during driving tests in the morning in all treatment conditions, irrespective of MDMA dose and concentration. The increments in SDLP were of high clinical relevance and comparable to those observed for alcohol at blood alcohol concentrations >0.8 mg/mL. These impairments were primarily caused by sleep loss. CONCLUSIONS: In general, MDMA did not affect driving performance nor did it change the impairing effects of sleep loss. It is concluded that MDMA cannot compensate for the impairing effects of sleep loss and that drivers who are under the influence of MDMA and sleep deprived are unfit to drive.