Flumazenil discrimination by humans under a two-response and a novel-response procedure.

In this study we assessed the discriminative stimulus, self-reported, and performance effects of flumazenil in humans. The first group (n = 6) was trained to discriminate flumazenil (0.56 mg/70 kg i.v.) from saline and tested with flumazenil (0.10, 0.32, 0.56, and 1.0 mg/70 kg) under a two-response drug discrimination procedure. The second group (n = 8) was trained to discriminate flumazenil (0.56 mg/70 kg i.v.) from saline and tested with flumazenil (0.32, 0.56, and 1.0 mg/70 kg), midazolam (0.10, 0.56, and 1.0 mg/70 kg), and caffeine (75 mg/70 kg) under a novel-response drug discrimination procedure. In both groups, flumazenil was acquired and maintained as a discriminative stimulus. Flumazenil dose-dependently increased flumazenil-appropriate responding and ratings of strength of drug effect and sedation, and decreased ratings of stimulant effects and psychomotor performance. Under the novel-response procedure, midazolam produced dose-dependent increases in flumazenil-appropriate responding. However, midazolam produced 43 and 25% novel responding at the intermediate and highest test doses, respectively. Midazolam dose-dependently increased ratings of strength of drug effect and sedation, and decreased ratings of stimulant effects and psychomotor performance. The magnitude of effects on ratings of strength of drug effect and sedation were comparable after flumazenil and midazolam, but psychomotor performance effects were greater after midazolam than after flumazenil. Caffeine produced mostly saline-appropriate responding. The results indicate that flumazenil has agonist effects similar to those of midazolam; however, novel responding after midazolam, and the greater performance decrement after midazolam, suggest that flumazenil does not act as a traditional benzodiazepine agonist.