Findings are inconsistent regarding whether women's cognitive performance fluctuates across phases of the menstrual cycle, but differences in methodology and the use of reported cycle phase rather than precise hormonal measures may underlie these disparities. Studies in monkeys may help resolve these discrepant findings, since hormonal status can be reliably determined. We tested four young (5-7 years old) female rhesus monkeys daily during one entire menstrual cycle on three cognitive tasks displayed on a computerized touch-screen system: a Matching to Sample task with a 30 s delay (MTS-30s), a Matching to Sample task without delay (MTS-no delay) and the spatial condition of the Delayed Recognition Span Test (spatial-DRST). Blood samples were collected at specific time intervals throughout the cycle and assayed for estradiol and progesterone in order to identify hormonal status. There was a nonsignificant trend for the MTS-30s scores to be better during the follicular and luteal phases, when estradiol levels were low, than during the peri-ovulatory phase, when estradiol levels were at their highest. MTS-no delay performance did not vary as a function of hormonal status. Spatial-DRST scores were significantly better during the follicular and luteal phases than during the peri-ovulatory phase of the cycle. These data in the female rhesus monkey support the hypothesis that spatial memory performance is sensitive to estradiol variations across the menstrual cycle, with better performance associated with low estradiol levels.