The study was done by using the Iowa Driving Simulator, a computerized facility in which participants sit in a real car within a domed structure and experience realistic movement. It simulates driving conditions and measures driving performance. The system simulated a 45-mile drive in dry weather conditions on a two-lane rural highway. Each participant completed four driving sessions separated from each other by 1 week. Before each session, the driver received fexofenadine, diphenhydramine, alcohol, or placebo (a dummy pill containing no active ingredient). In addition to a pill, patients received a drink of soda before each drive; the drink either contained alcohol or just had a small amount of alcohol rubbed on the rim of the cup. Neither the driver nor the person who judged the driving performance was aware which of the four substances the driver had received. The researchers measured the ability of the participants to match the varying speed of a “virtual vehicle” they were following. They also measured drowsiness, the ability to stay in the lane, and the driver's response to a “virtual vehicle” that unexpectedly pulled out into the road and blocked the lane.