The present study demonstrates marijuana-and alcohol-induced impairment of a driving-relevant measure in a driving simulator task at (estimated) blood alcohol and THC concentrations that are below the per se cutoff for impaired driving in several states. The subject was an adult male with a history of occasional alcohol use (2-3 times/week for the past 6 months) and past but very infrequent use of marijuana, i.e., less than once/month for the past 6 months. The testing procedure was a crash avoidance test using a fi xed base driving simulator. In this procedure, while driving at 55 mph, the subject was required to make an 'emergency' steering maneuver to avoid crashing into a 'stalled car' that appeared on the roadway immediately (40 meters) ahead. In the absence of any drug treatment, after training the subject effectively made this avoidance maneuver in >98% of trials (20 trials/session), with a crash avoidance response latency of approximately 450-475 msec from the onset of the car ahead until an abrupt crash avoidance steering response of >10 degrees. On two test days separated by 14 days, the subject was tested for crash avoidance reaction time following oral alcohol (beer consumption) use and, on another occasion, following oral marijuana use (approximately 10 mg in a 'candy'). The testing involved a predrug test and several post-treatment tests, and one fi nal post-treatment test 24 hours later. On both test days, blood samples were collected at various times after drug administration and throughout behavioral testing. Both alcohol and marijuana treatment signifi cantly increased crash avoidance reaction time (from approximately 475 msec to > 550 msec). Plasma alcohol concentrations of 42 mg/dl, 79 mg/dl and up to 99 mg/dl (corresponding to BAC values of 39, 67 and 86 mg/dl, respectively) were associated with alcohol impaired driving over the time period from 45 minutes to 175 minutes after the onset of drinking. On the marijuana test day, plasma concentrations of THC were 4.7 and 2.4 ng/ml (corresponding to blood THC concentrations of 2.9 and 1.5 ng/ml, respectively) at times when signifi cant impairment of driving was observed. This is the fi rst study to demonstrate dramatic driving simulator performance impairment at a THC blood concentration less than 3 ng/ml, which is the below the 5 ng/ml cutoff for marijuana-impaired driving in several states (e.g., Washington, Colorado). These data further suggest that the crash avoidance reaction task might be useful in further studies on the effects of marijuana, alcohol and other drugs (prescription, non-prescription and also drugs of abuse) on driving performance.