In a new study, researchers studied dozens of women through two menstrual cycles. They say they eliminated ‘false positives’ found in previous research.
Does menstruation affect a woman’s cognition?
According to new research findings, the answer is “No.”
A longitudinal study published this week in Frontiers in Behavioral Neuroscience found no consistent and meaningful link between prefrontal cognitive functioning and changes in female hormone levels across two menstrual cycles.
The investigators studied 88 participants, ages 18 to 40. The research subjects included 58 women with no health issues and 30 women diagnosed with endocrine disorders.
Researchers assessed each participant’s hormone levels, visuospatial working memory, attention, and cognitive bias at multiple points throughout a full menstrual cycle.
Then 68 participants were reassessed over the course of a second menstrual cycle to rule out practice effects and false-positive chance findings.
They found no relationship between estrogen, progesterone, and testosterone levels, or differences in cognitive functioning between different participants.
They also found no consistent and meaningful link between changing hormone levels and cognitive functioning across individual participant’s menstrual cycles.
The researchers did find an association between changes in hormone levels and test scores for working memory and cognitive bias in the first menstrual cycle they assessed.
However, those associations didn’t replicate during the second cycle.
The findings suggest that associations found in the first cycle were false-positives.
“Research findings are fraught with uncertainty, especially in probabilistic sciences such as medicine and psychology, where there is a large variability in how persons react to given stimuli,” co-author Michael P. Hengartner, PhD, a senior researcher at the Zurich University of Applied Sciences, told Healthline.
“Due to the very complex nature of women’s responses to hormonal fluctuations, neuroendocrinological research is particularly prone to false-positive findings,” he added.
In this study, the researchers took the novel approach of using data from a second menstrual cycle to validate their findings from the first cycle.
As far as they know, they are the first research team to use this study design.
“With this approach, we can rule out chance findings, as true and meaningful associations should be consistent across both cycles,” Hengartner explained.
Rather than a true relationship between cognitive functioning and hormone levels, he believes the associations in the first cycle reflect random variations and systematic biases.
For example, practice effects might play a part.
Practice effects are improvements in test scores that can result when subjects are exposed to the same test materials multiple times.
In other words, participants’ changing test scores may reflect their growing familiarity with the test materials, rather than changes in their cognitive abilities.
Given the potential for false results, positive findings from other studies should also be interpreted with caution.
“There are many inconsistencies in published findings,” Hengartner noted. “That is, some [past] studies found an association and others didn’t. However, we believe that many positive findings could be false-positive chance findings, as no previous study used data from a second cycle to verify their own results and test for chance findings.”
“Hormones certainly have an effect on mood and behavior, but thus far we don’t know why and how,” he added. “What we can tell, however, is that hormone levels do not have uniform and universal effects.”
According to the authors of a 2014 review article, research findings on this topic have been mixed, and positive findings have been “notoriously difficult to replicate.”
While some studies have found changes in cognitive functioning associated with changes in hormones across the menstrual cycle, others have found no such link.
Many investigations that have found associations have relied on cross-sectional research designs, which are susceptible to selection bias.
“There are problems with the longitudinal design as well,” Dr. Inger Sundström Poromaa, lead author of the review article, told Healthline. “But I think it’s still much better [to do it longitudinally] than to do it cross-sectionally.”
“Ideally, I think the best way to study this would be to do various types of hormonal interventions,” she continued. “There’s a German group right now who are randomizing women to relatively high doses of estrogen or placebo — and that’s an even better design.”
When it comes to Hengartner and her colleagues’ findings, Sundström Poromaa didn’t seem surprised by the lack of consistent association they found.
“From a clinical perspective, I don’t expect that you would see any cognitive changes across the menstrual cycle,” she said. “Because if you did, I would see women who would be complaining that ‘I can’t do math when I’m in this phase,’ or ‘I failed my test because I wasn’t in that phase.’ And I’ve never come across any of those complaints.”