Twenty-one (three groups of seven), men and women, 25–50 years of age were studied to determine whether or not rebound insomnia would increase the likelihood of self administering a benzodiazepine (triazolam 0.25 mg) hypnotic. The groups compared were patients with insomnia and disturbed sleep, insomnia and normal sleep, and healthy normals. Rebound insomnia, by both subjective and polysomnographic assessment, was induced. The experience of rebound insomnia did not increase the likelihood of self administering a benzodiazepine hypnotic in any of the groups. There were clear group differences in pill self administration with normals rarely and insomnia patients frequently, but not differentially (placebo versus active drug) self administering pills.