The present study used an intensive longitudinal design to examine whether mental rotation performance varies according to a monthly cycle in both males and females and whether these variations are related to variations in progesterone, estradiol, and testosterone levels. We collected reaction time and accuracy data for 10 males and seven females each workday over eight weeks using 136 pairs of mental rotation stimuli/ day, and measured sexual hormones concentrations in the saliva twice a week. A mixed linear model statistical analysis revealed that all females and seven males showed significant cycle effects in mental rotation performance. The female cycle showed an amplitude that was twice as large compared with the amplitude found in males. For males and females, estradiol and testosterone were significantly linearly and quadratically related to interindividual variation in performance at the beginning of the study (progesterone was linearly related to performance for females). The association between testosterone and performance differed across sexes: for males, it had an inverse U-shape, for females it was U-shaped. Towards the end of the study, none of the hormones were significantly related to performance anymore. Thus, the relationship between hormones and mental rotation performance disappeared with repeated testing. Only estradiol levels were significantly elevated at the lowest point of the cycle in mental rotation performance in females. In conclusion, in this intensive longitudinal study spanning two months, a monthly cycle in mental rotation performance was found among both males and females, with a larger cycle's amplitude for females.
Available at: http://works.bepress.com/christian-geiser/46/