A study involving 88 women, some of whom had endocrinological disorders, has found that, while some hormones were associated with changes across one menstrual cycle in some of the women taking part, these effects didn't repeat in the following cycle. In other words, there was no consistent effect of hormonal changes on cognition. This is not to say that some individuals might not be consistently affected, just that it doesn’t appear to be a general rule.
While the number of participants isn’t huge, it is considerably larger than is common in these sort of studies. The replication across two cycles is particularly important, since if the researchers had settled for just looking at one cycle, they would have concluded that there was an effect on cognition — as several studies have previously concluded. This more rigorous study suggests that earlier findings should be regarded with caution.
The study followed the women through two menstrual cycles. For the first cycle, 88 women participated; 68 women were re-assessed for a second cycle, to rule out practice effects and false-positive chance findings. Visuospatial working memory, attention, cognitive bias and hormone levels were assessed at four consecutive time-points across both cycles.
Of the initial 88, 58 had no endocrinological problems, 13 were diagnosed with endometriosis, 16 with polycystic ovary syndrome (PCOS) and one woman with hyperprolactinemia. Additionally, 12 women presented with obesity. Women were excluded if they were using oral contraceptives, had been pregnant or breastfeeding within the past 6 months, were using medication or had surgery which might interfere with endocrine parameters, had severe psychiatric or general diseases, worked irregular shifts, had menstrual or ovulation disorders except those investigated in the study, or showed any additional abnormality in hormonal parameters. Mean age was 30. Data from the subset of healthy women were also analyzed separately, confirming no difference in the findings. I would have liked the researchers to mention how the 68 women in the replication were selected, but I assume, after all their emphasis on methodological rigor, that they would have been careful to make sure there was no bias in that selection.
It should be noted, however, that the cognitive testing wasn’t exhaustive by any means — it’s possible that other cognitive aspects might be affected by hormonal changes. However, attention and working memory are the areas generally accused, and most likely to be noticed by an individual.
Of course, that’s the thing about attention and working memory — they’re very sensitive to a host of factors, including sleep quality and stress. So, we often notice that we’re not working at top gear, and we’re likely to look around for reasons. If we’re women, and it’s our period or just before it, we’re quite likely to attribute the reason to that. And it may be true in an indirect way — if we have pain, or sleeplessness, or are stressed, for example. What this study tells us, is that the changes in hormonal levels don't seem to consistently affect cognition.
 Lack of Associations between Female Hormone Levels and Visuospatial Working Memory, Divided Attention and Cognitive Bias across Two Consecutive Menstrual Cycles. Frontiers in Behavioral Neuroscience [Internet]. 2017 ;11. Available from: http://journal.frontiersin.org/article/10.3389/fnbeh.2017.00120/full