Effects of gonadal steroids on conspecific odor preference for either (1) sexually active male or active female, (2) sexually active or gonadectomized (gdx) males, (3) sexually active or gdx females, and (4) gdx males or gdx females were determined in male and female rats in a three-chamber apparatus. For the first test, gdx females were made sexually active by treatments with estradiol benzoate (EB) and progesterone (P), and sexually active males were selected by prior screening. Sexually active males and females preferred opposite-sex odor over same-sex odor. Odor of sexually active opposite-sex conspecifics was preferred over that of inactive ones. Immediately after the completion of the first test, sexually active males were gdx and females were left without hormonal treatment. Second and third tests were carried out 2 and 5 weeks after the first test. In the second test, gdx males preferred odor of sexually active males rather than that of receptive females (male-directed preference); in the third test, both males and females showed no preference when tested with four stimulus pairs. The final tests were carried out in gdx males with EB and P, and gdx females with 2-week exposure to testosterone (T). Males with EB and P showed a male-directed preference again, whereas T-treated females kept their own female preference. Injection of EB alone to gdx males did not induce any preference. The present study clearly demonstrated sex difference in conspecific odor preference. Although both male and female preferences depend on their circulating sex steroids, the direction of male preference is more susceptible to their hormonal states, compared to that of females.