Recent research using genetically modified mice has pointed to the specific contribution of individual receptor subtypes to the various effects of benzodiazepines. The aim of this study was to examine the relative significance of alpha(1)-containing GABA(A) receptors in the effects of modulators at the benzodiazepine site in the elevated plus-maze (EPM) under dim red light in rats. We tested the effects of the non-selective antagonist flumazenil (0-20.0 mg/kg), the preferential alpha(1)-subunit selective antagonist beta-carboline-3-carboxylate-t-butyl ester (beta-CCt, 0-30.0 mg/kg), the non-selective agonist midazolam (0-2.0 mg/kg), the preferential alpha(1)-subunit selective agonist zolpidem (0-2.0 mg/kg) and the non-selective inverse agonist methyl 6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate (DMCM, 0-2.0 mg/kg). The influence of flumazenil (10.0 mg/kg) and beta-CCt (30.0 mg/kg) on the effects of both kinds of agonists were also examined. The standard spatio-temporal parameters reflecting anxiety (percentage of open arm entries and time) and locomotion (closed and total arm entries) were analyzed. beta-CCt did not affect behavior, while flumazenil at the highest dose (20.0 mg/kg) decreased indices of open arm activity and total arm entries. Midazolam at the dose of 1.0 mg/kg significantly increased the percentage of open arm time, whereas at 2.0 mg/kg both anxiety-related parameters were increased. In contrast to the open arm entries, the open arm time was independent of the decreased closed arm entries, observed at 2.0 mg/kg. Flumazenil abolished these effects, whereas beta-CCt partially potentiated the anxiolytic actions of midazolam. Zolpidem significantly increased both open-arm indices at 1.0 mg/kg, but the effect was dependent on the decreased closed arm entries. The selectivity of the anxiolytic-like effects of zolpidem was further checked under brighter white illumination. In these settings, the influence on anxiety-related, but not activity-related parameters, was absent. All of the activity-related effects of midazolam and zolpidem were mainly counteracted by both antagonists. DMCM produced significant anxiogenic effects at 1.0 mg/kg (open arm time) and 2.0 mg/kg (both parameters). beta-CCt (30.0 mg/kg) and flumazenil at higher dose (20.0 mg/kg) antagonized the effects of DMCM. The results indicate the anxiolytic effects of a non-selective benzodiazepine site agonist involve a predominant role of subunits other than alpha(1), whereas the behavioral indices of the anxiolytic-like properties of an alpha(1)-selective ligand, if observed, depend on the experimental settings and the changes in locomotor activity, and hence were behaviorally non-specific. The present results generally correspond well to the behavioral findings with the genetically modified mice. On the other hand, the relative significance of the alpha(1)-subunit in the anxiogenic effects of DMCM could not be clearly deduced.