A learned avoidance response has been one of the hypotheses proposed to explain the 'one-trial tolerance' (OTT) phenomenon, which represents a drug's loss of anxiolytic-like effect in the elevated plus-maze (EPM) test in experienced rodents. Based on these facts, if some kind of learning occurs throughout Trial 1, then an impairment of its acquisition would maintain the drug's anxiolytic-like effect on Trial 2. Using male Wistar rats, the present study examined whether scopolamine (SCO; 0.5-1.5 mg/kg), a drug that impairs learning acquisition, given 30 min prior to Trial 1, actually prevents the OTT phenomenon to either the midazolam (MDZ; 0.5 mg/kg) or the memantine (MEM; 8.0 mg/kg) anxiolytic-like effect on the EPM Trial 2 (48 h later). According to the results, both MDZ and MEM increased open-arm exploration (indicating anxiolysis) on Trial 2 only in rats that had been treated previously with 1.5 mg/kg SCO. These results were observed in the absence of change in general exploratory activity. The present findings suggest that SCO impaired the acquisition of the behavioral strategy to cope with the subsequent EPM exposure that supposedly underlies the OTT phenomenon, thereby revealing the anxiolytic-like effects of MDZ and MEM on Trial 2.