Sex differences in the emotional behavior of three strains of rats were investigated, using a variety of measures taken during the initial 2 min of open-field exposure -and after a loud bell had sounded. Animals were tested either in a clean open field or in one containing the odor of a same-sex conspecific. No general overall sex differences for a variety of measures associated with emotional reactivity were found, so the results did not support the theory that male rats are more "emotional" than females. Sex differences for particular measures associated with emotional responsiveness were found, but these varied with the strain, measure, and odor condition: The latter was found to be particularly important.Sex differences in the behavior of rodents in novel environments and to other test situations for "emotional" behavior have been frequently reported in the literature. Typically, females show more ambulation and less defecation in an open field (e.g., Broadhurst, 1957Broadhurst, , 1958Henderson, 1963; Gray, Levine, & Broadhurst, 1965), they emerge sooner from their home cage into a novel environment (e.g., Meyers, 1962;Woods, 1962; Pfaff & Zigmond, 1971), and they show faster acquisition of a shuttlebox avoidance learning task (studies reviewed by Gray, 1971;Beatty & Beatty, 1970) than do males. Gray (1971) interpreted these and other results as supporting the hypothesis that male rodents were more fearful than females.Gray's theory was criticized by the present author (Archer, 1971) on several grounds. One of these was that sex differences in open-field measures were more variable than Gray suggested, strain differences being a particularly important determinant of this variation. Another criticism was that the commonly used open-field measures (ambulation and defecation) were equivocal, since their hypothetical inverse relationship is found in only some, but not all, studies involving the appropriate correlation data. It was further suggested (Archer, 1973) that variables other than emotional reactivity could influence these measures, e.g., ambulation could, under different circumstances, reflect causally different types of behavior (e.g., lack of "freezing," presence of escape behavior, lack of lying or sitting, or lack of sniffing at localized stimuli). A fuller discussion of these and related issues is included in Archer (1973).Odors are particularly important in the social behavior of rodents (review by Cheal & Sprott, 1971). With a few recent exceptions (Whittier & McReynolds, 1965; Stevens & Koster, 1972; Davies & Bellamy, 1972), little attempt has been made in open-field studies to control *Supported by a grant from the Medical Research Council of the United Kingdom to R. 1. Andrew. I thank Professor Andrew for his helpful comments on this paper.
43for the possibility that a previously tested animal might influence the behavior of a currently tested one.The present paper describes an investigation of sex differences in the emotional behavior of rats in an open-field test, modified to overcome some...