Triazolam (0,0.125, 0.25, 0.5, and 0.75 mg/70 kg) and lorazepam (0, 1, 2, 4 and 6 mg/70 kg) were compared in eight healthy men by using a double-blind, crossover design. Triazolam was chosen for study to examine experimentally whether it produces greater behavioral impairment than other benzodiazepines as reported previously. Lorazepam was chosen as the comparison drug because of its well documented behavioral effects and because it has a short metabolic half-life and no active metabolite, which eliminates the possible confounding effects of drug accumulation. Drug effects were assessed before and every 30 min for 8 hr after drug administration. Learning was chosen for study because it is a fundamental component of more complex behavioral processes such as recall, and benzodiazepines are known to disrupt learning. A psychomotor task and subject-rating scales, assessing drug effects and abuse potential, were also included to provide a more comprehensive comparison of these compounds. Triazolam and lorazepam dose-dependently disrupted learning and psychomotor performance and increased subject ratings of sedation to a comparable degree. These findings do not support allegations that triazolam produces greater behavioral impairment than other commonly used benzodiazepines.