Rationale-Zolpidem is a hypnotic drug that binds to γ-aminobutyric acid type A (GABA A ) receptors but lacks consistently demonstrable anxiolytic efficacy.Methods-Rhesus monkeys (N=4) were trained under a multiple schedule in which foodmaintained responding was programmed (18-response, fixed ratio) for a 5-min period, followed by a 5-min period in which the food-maintained responding was suppressed by response-contingent electric shock (20-response fixed ratio). Doses of zolpidem (range= 0.03 to 1.0 mg/kg, i.v.) were administered 5 minutes before the session and responding was re-assessed at 3 additional 20 minute intervals. A similar experiment also was carried out with the non-selective benzodiazepine, triazolam, over a dose range of 0.001 to 0.1 mg/kg, i.v.Results-Zolpidem did not engender a significant increase in average rates of suppressed responding at earlier time points; however, rates of non-suppressed responding were robustly decreased. At 45-and 65-min post-injection, zolpidem treatment resulted in a dose-dependent increase in rates of suppressed responding. In contrast, the non-selective benzodiazepine triazolam increased rates of suppressed responding in a dose-dependent manner at all 4 time points, although decreases in non-suppressed responding were less at the later time points.Conclusions-These findings suggest that zolpidem has anxiolytic-like effects, but only >25 min after i.v. injection in this rhesus monkey conflict model. It was hypothesized that timedependent effects on the response rate suppressing properties of zolpidem become tolerant (i.e., acute tolerance). Because anxiolytic-like effects remain stable throughout the session, the absence of rate-decreasing effects may "unmask" anti-conflict effects.