The effects of 5-HT(1A) receptor agonist buspirone, a nonselective (diazepam), and a selective (zolpidem) GABA(A) receptor agonist were compared in the open field test of neophobia. Unhabituated rats were pretreated with the drugs once, prior to a first exposure to the open field, and their behavior was recorded both during this test and during a second trial 24 h later. It has been hypothesized that the decrease in exploratory activity observed during the second test session may be considered an adaptive reaction to the first day aversive experience (neophobia). If so, a selective modulation of 5-HT and GABA systems activity during the test could bring about significant changes in animal behavior on the retest. Buspirone at the lowest dose of 0.3 mg/kg revealed anxiolytic-like properties on the first day, whereas the action of diazepam and zolpidem was modulated by the dose-related sedative effect. At the dose of 2.4 mg/kg buspirone elicited delayed in time anxiolytic-like action, i.e., produced the antithigmotactic effect during the retrial 24 h later. Diazepam and zolpidem failed to exhibit similar profile of action. Autoradiography of [3H]muscimol binding after pretreatment of rats with buspirone showed a significant increase in the selective radioligand binding within the frontal cortex and a similar, near-significant tendency in the dentate gyrus of the hippocampus. The behavioral data validate buspirone as important drug for the treatment of anxiety disorders, devoid of disruptive influence on motor and cognitive processes. The open field test, as modified by us, appeared sensitive in distinguishing the behavioral profiles of action of different anxiolytic compounds, including 5-HT(1A) receptor agonist. The present results support the assumption that reduced turnover of 5-HT due to stimulation of 5-HT(1A) autoreceptors, may bring about changes in GABA(A) receptor system activity, in some brain structures, leading to the anxiolytic effect.