The effect of estrogen and/or androgen replacement therapy on several aspects of cognitive functioning in surgically menopausal women was tested in a prospective, crossover design. Women who received either a combined estrogen-androgen preparation, estrogen alone, or androgen alone had scores on two tests of short-term memory, a test of long-term memory and a test of logical reasoning that were not different during the postoperative treatment phase compared to their preoperative performance. However, oophorectomized women who received placebo had lower scores on all four measures of cognitive functioning postoperatively, coincident with their significantly lower concentrations of plasma estradiol and testosterone. Patients who had a hysterectomy but whose ovaries were retained showed stability both in cognitive performance and in circulating sex steroid concentrations. These findings suggest that the drastic change in endocrine milieu following surgical menopause may have a direct, albeit modest, effect on aspects of cognitive functioning. Possible mechanisms of action of the sex hormones on cognitive functioning in women are discussed.