Effects of the menstrual cycle on emotions and abilities

Read Also

 There is a phase of her menstrual cycle (ovulatory) when a woman is fertile and a phase (menstrual) when she is not.  Consequently, if anything in mental life and behavior could be predicted to vary as a function of the menstrual cycle,  it ought to be sexual desire.   Some species of mammals have zero sexual desire when they are not fertile.  Of course,  everybody knows that humans are not that way.  However, there is a subtle variation in humans,  in the expected direction:   women's libido is maximal at ovulation when they are most fertile,  and minimal at menstruation when they are not fertile.  One team of investigators examined prospective data from 1,066 women (aged 19-44 years) for evidence of covariation between the timing of sexual desire and menstrual cycle characteristics. In any given menstrual cycle, sexual desire was usually first experienced a few days before the basal body temperature (BBT) shift, around the expected ovulation date. Positive correlations were found (1) between the day of the BBT shift and the day of sexual desire onset and (2) between the length of the menstrual cycle and the temporal lag between the onset of sexual desire and the BBT shift.   These results are obviously consistent with a model in which sexual desire is affected by the same process that regulates the menstrual cycle, suggesting that hormonal factors may contribute to sexual desire in a biologically meaningful manner:  sexual desire increases with reproductive fitness thus economizing on expenditure of time, energy and risk of infection.

Another outstanding psychological correlate of menstrual cycling variations is mood.  Dozens of studies have demonstrated that mood is at its best around ovulation and at its worst around menstruation (especially just before they start) and this effect is often found to be statistically significant.  This particular psychological effect of the menstrual cycle is one of the most robust.   It is also the most likely to be mediated by the neurotransmitter serotonin,  which is the neurotransmitter most affected by menstrual variations in estrogen.  Simple as this account may seem,  one study has managed to impose a constraint on it.   One research team tested mood in two cohorts of normal-cycling women.  One group was tested before being asked about their menstrual status and had not been informed of the purpose of the study.  The other group had been informed of the purpose of the study,  and was asked about menstrual status before completing the mood test.   The authors found that the effect of the menstrual cycle was not significant in the uninformed group.  They interpreted their findings in the following manner:   women attribute a negative mood to themselves as a manner of accounting for the more specific displeasures of menstruation.   In other words,  the classical finding of menstrual cycling of mood scores on mood tests seems to consist,  in large part,  of a cognitive construal (an overextended interpretation) more than of a truly specific variation of mood.  My own belief is that there is a very minor cycling of mood,  but that this particular study was not methodologically sound enough to pick it up.   However,  I do agree that much of the effect consistently reported to date has consisted of cognitive construal.  

Student’s tribune:  Mood and the menstrual cycle,  is the variation objective or subjective ?

In a recent experiment carried out by my student Loïc Villeneuve,   we found that mood was very significantly lower in women tested at menstruation than in the same women tested at ovulation.  However,  the women knew that the investigation was designed to study the menstrual cycle.   And in fact,  salivary estrogen and testosterone levels and  body temperature did not correlate with mood at all.  

Another  finding commonly reported is that whenever a significant cycling of cognitive performance is observed (which is not always the case),  it is at ovulation that the performance is at its best,  and just before menstruation that it is at its worst.   This has been reported for basic perceptual and motor functions,  but also for higher order functions in the cognitive domain            -including such things as abstract reasoning,  planning,  mental calculation,  and several performances on neuropsychological tests.  In fact, it seems that of the two basic levels of cognitive functioning,  the simple and complex,  it is the complex which is more often found to be modulated by the menstrual cycle.  Simple reaction time has most often been found not to relate to the menstrual cycle,  whereas complex reaction time effects do seem to follow the menstrual cycle on occasion.  Likewise,  in evoked potential experiments,  the same general trend is observed.  Evoked potentials are a special form of electroencephalography.   Brain electrical activity is collected from the scalp in normal people in response to repeated stimulation,  involving simple processing such as detection or complex processing such as discrimination, classification, etc.   One complex evoked potential is the third (late) positive wave (P3).  This wave is one of the few which has been found to follow the menstrual cycle.  It’s latency is shorter at ovulation,  indicating that the brain is functioning a bit better at this moment of the cycle.  How substantial are these variations ?   Not very.   Most women cannot even notice it in themselves.   The vast majority of these tasks are performed just as well by women as by men.  Gouchie and Kimura (1991) presented findings to the effect  that women do better in female-typical tasks at mid-luteal (high- estrogen) phase, and better at male-typical tasks at the menstrual (low estrogen) phase.  

Now here is the most interesting finding from a neuropsychological perspective.  As I explained in chapter 3, experimental psychologists have devised ingenious techniques to understand how one hemisphere of the brain can be more efficient than the other.    The two main such techniques are tachistoscopy for vision and dichotic listening for audition.  I am aware of eight studies that have investigated the relative efficiency of the two hemispheres by comparing perceptual or cognitive processing of stimulation of the two sides of the body (i.e., by inference, of the two hemispheres) with these techniques.   The results have always been basically the same.     The left-sided stimulations gain a significant relative advantage at menstruation,  whether the material to be processed is of a verbal nature or not !  (see Bibawi et al, 1995 and Hesiter et al, 1980).  This is a remarkably consistent finding in several senses.  First, it is remarkable because the techniques used are notoriously capricious (the effects of laterality depend on many technical details).  Second, it is remarkable because it fits well with the rest of what is known about sex differences in cognitive abilities and brain function.   Indeed,   women's cognitive advantages over men are in the verbal domain and the verbal domain is left hemisphere-dependent.  It seems that increased hemispheric specialization,  in women,  results from a drop in steroid hormones rather than from  an increase.   This actually fits logically with the literature I reviewed in chapter 3 in one sense.   Overall, men manifest more hemispheric specialization on tasks such as these (tachistoscopy or dichotic listening) than do women (see Halpern, 1992;  Kimura, Levy & Heller, 1992; 1993; McGlone, 1980,  for reviews of the relevant literature).   Women present a more male-like pattern (with regard to hemispheric specialization) when their hormonal status is more male-like.    This account is incomplete however.   The specialization of the women,  occurring at menstruation,  is always in the direction of a relative right hemisphere advantage.    Recall that men manifest more left hemisphere specialization for verbal material and right hemisphere specialization for spatial processing  (see chapter 11 for more reflections on this theme).

Finally,  the cutting edge in the field of neuropsychology of the menstrual cycle comes from two research teams who have both very recently reported data compatible with an effect of the menstrual cycle on interhemispheric relay (Nicole Weekes and her colleagues,  and Elisabeth Hampson and her colleagues).  They proposed that based on their results,  interhemispheric relay is less efficient (less accurate) around ovulation than at menstruation.  There is also evidence that the same effect is observed in rats. 

Student’s tribune:  Does interhemispheric physiology vary as a function of the menstrual cycle ?

To follow up on these exciting research findings,  my student, Loïc Villeneuve, tested 28 normal regularly cycling women at mid-menstruation and as close as he could to ovulation.   To estimate interhemispheric relay time, he used simple visual reaction time,  arranged in a paradigm called the “Poffenberger paradigm”.   The  right visual field projects to the left hemisphere,  and the left to the right.   The left hemisphere produces a right hand response and the right a left.  When a subject responds to a lateral stimulus with the contralateral hand (the hand opposite to the stimulated visual field),  it is assumed that the neural relay must cross the brain commissures.  However,  when the subject responds with the hand ipsilateral to the stimulus (on the same side),   interhemispheric relay is not required.  The difference in reaction time between these two conditions is believed to be an estimate of interhemispheric relay time.  Loïc found that interhemispheric relay time was not different in the two menstrual stages,  but he did find that accuracy (smaller prevalence of omission errors) of interhemispheric relay was indeed significantly lower at ovulation than at menstruation.  Now the neurotransmitter used by most interhemispheric neurons is glutamate.   Glutamate is an excitatory neurotransmitter.   An intriguing in vitro study of slabs of rat brain recently found that serotonin has a selective inhibitory influence on interhemispheric neurons of the corpus callosum  -which is the main interhemispheric commissure.  Central nervous system serotonin reaches its peak around ovulation in normal cycling women. So one speculation could be that the menstrual cycle influences interhemispheric relay by a hormonal influence on serotonin.   However,  it should be noted that the variation we observed of so-called interhemispheric relay accuracy (which cannot plausibly be construed to consist of an attitudinal artifact)  did not correlate with our salivary measures of estradiol or of testosterone  -a finding which is typical of reports of menstrual cycling of mentation or behavior.  

I conclude that the evidence suggests that sex differences in cognitive performance depend, among other things, on the activating on-line effect of circulating hormones -in a complex manner that remains to be clarified.   Now nobody would suggest that cognitive performance is directly produced by steroid hormones.  Hormones are stupid. To explain these putative correlations between cognitive performances and hormone levels,  one has to imagine some sort of interaction between at least one (probably several) of these hormones and selective neuronal processes which alone are directly responsible for the cognitive performances.