OBJECTIVE To evaluate cognitive functioning in women with premenstrual syndrome (PMS) and controls during the follicular and luteal phases of the menstrual cycle. METHODS Thirty women with PMS and 31 controls were selected on the basis of psychiatric interview and prospective daily diary recordings. Subjects were tested on two occasions, follicular (days 8-10) and luteal (days 24-26), using complex tasks consisting of measures validated previously for the assessment of "executive" frontal-lobe functions. Tests were counterbalanced for order across subjects. RESULTS The Beck Depression Inventory scores were significantly different between the groups and across time (P < .001). Women with PMS had a mean luteal phase Beck score of 13.3 consistent with mild-to-moderate premenstrual depression. There were no statistically significant score differences in tests for attention, memory, cognitive flexibility, and overall mental agility. The evaluation of our preliminary data with 30 PMS subjects and 31 controls indicated a very small effect size (.02). To detect an effect size this small (if in fact one exists) with a power of .8 would require a sample of more than 1000 subjects per group. CONCLUSIONS Our sample of women with PMS failed to demonstrate objective evidence of diminished cognitive performance, despite subjective feelings of inadequacy.