Social support is proposed to attenuate behavioral consequences of exposure to uncontrollable stressors. To test this possibility, we compared the effects of two post-stress housing conditions, in pairs or in groups of 10-12 animals per cage, on the behavior of rats tested in the elevated plus-maze (EPM) 24 hr after stress. We also included positive control groups to compare the effects of a standard anxiolytic, diazepam, with those of vehicle. Confirming previous results, diazepam increased the percentage of entries and time spent in the open arms (vehicle, % open entries: 37.0+/-2.7, % time spent in open arms: 17.6+/-1.9; diazepam, % open entries: 46.7+/-2.7, % time spent in open arms: 39.1+/-3.9). Group housing after restraint significantly prevented the anxiogenic effect of restraint (group housing, % open entries: 32.0+/-5.2, % time spent in open arms: 17.6+/-5.0; pair housing, % open entries: 18.7+/-2.2, % time spent in open arms: 6.5+/-1.0). These results suggest that housing conditions could be an important factor in the development of behavioral consequences of stress exposure.