A case for the usefulness of laboratory social stressors