Based upon the ethological element "stretched attend" to a conspecific, which reflects ambivalence between approach and avoidance in a social context, a simple, non-social behavioral method for studying conflict behavior in mice was investigated. Thus, stretched attend posture (SAP) and other behavioral acts were measured in untreated or drug-treated mice which were individually placed on a perforated platform previously rubbed with foreign male urine and boluses. In naive, untreated mice, the occurrence of SAP was partly housing- and lighting-dependent. After repeated exposure to the test situation, untreated mice showed less SAP, whereas static behavior (immobility and activity at rest) was increased. After single oral treatment with diazepam, clobazam or phenobarbital, SAP was reduced, whereas static behavior or "going forwards in SAP" was increased. Chlorpromazine and imipramine did not influence SAP. In naive mice, single IP injection of pentylenetetrazol did not significantly increase SAP. The similarity in the behavioral response between experienced, untreated mice and naive animals treated with diazepam, clobazam, or phenobarbital suggests that drug-treated mice behaved as if they were already familiar with the test situation. The latter drug-induced changes are consistent with data obtained in animal models which are based upon the measurement of behavioral inhibition. Under the present test conditions, pentylenetetrazol did not show "anxiogenic" properties. Nevertheless, the SAP test, based primarily on the measurement of the ambivalence element, offers a simple procedure for examining the conflict-reducing properties of drugs, dispensing with the need of noxious stimuli or prior training.