The Human Brain Is Best Described as Being on a Female/Male Continuum: Evidence from a Neuroimaging Connectivity Study

The Human Brain Is Best Described as Being on a Female/Male Continuum: Evidence from a Neuroimaging Connectivity Study. Yi Zhang et al. Cerebral Cortex, bhaa408; January 20 2021, https://doi.org/10.1093/cercor/bhaa408

Abstract: Psychological androgyny has long been associated with greater cognitive flexibility, adaptive behavior, and better mental health, but whether a similar concept can be defined using neural features remains unknown. Using the neuroimaging data from 9620 participants, we found that global functional connectivity was stronger in the male brain before middle age but became weaker after that, when compared with the female brain, after systematic testing of potentially confounding effects. We defined a brain gender continuum by estimating the likelihood of an observed functional connectivity matrix to represent a male brain. We found that participants mapped at the center of this continuum had fewer internalizing symptoms compared with those at the 2 extreme ends. These findings suggest a novel hypothesis proposing that there exists a neuroimaging concept of androgyny using the brain gender continuum, which may be associated with better mental health in a similar way to psychological androgyny.

Keywords: androgyny, brain functional network, sex difference

Discussion

In the present study, we identified an age-dependent pattern of sex differences in the brain functional architecture using the fMRI data of nearly 10 000 participants from teenagers to older adults, and systematically examined the potentially confounding effects on these findings. Based on the identified sex differences, we trained an SVM classifier that achieved a 77.75% accuracy in an independent test sample. Using the continuous output of this SVM, we constructed a brain gender continuum and defined an androgynous brain to be at the middle of this continuum. Indeed, we showed that the patterns of functional connectivity, at the 2 extreme ends of this brain gender continuum, represented predominantly either more female or male features as compared with the center of the continuum. Finally, we used this brain gender continuum to uncover a U-shaped relationship between the neuroimaging-defined brain gender and mental health, particularly the participants with an androgynous brain indeed had fewer internalizing symptoms.

The age-dependency of the sex differences may be associated with a number of factors such as the behavior, genetics, and hormones. Research has shown that different environmental contexts, experiences, and behaviors, throughout the lifespan may alter the structural and functional architecture of the brain, in addition to modulation by neurotransmitters (Kolb and Gibb 2011). Genetic factors may also have differential expression across the lifespan, for example Deary et al. (2006) have shown different rates of heritability of intelligence across age. In addition, the sex hormones have nonlinear developmental trajectories (Haimov-Kochman and Berger 2014; Mcewen and Milner 2017) which increase during childhood and adolescence (Nottelmann et al. 1987) but decrease during aging (Rosario et al. 2004; Cui et al. 2013). Particularly, testosterone, a sex hormone, has been implicated in the developmental change of the DMN (Nota et al. 2016), and in our study we found that 3 brain regions (i.e., the cingulate cortex, angular cortex, and precuneus) with the most differences in their functional connectivity were all identified within the DMN and these differences were also supported by previous studies using smaller samples at different age groups (Lombardo et al. 2018; Ritchie et al. 2018; Ernst et al. 2019). Furthermore, in the trained SVM, a multivariate classifier, we also found that the DMN contributed the most to the classification accuracy of this model. Our findings suggest that the patterns of functional connectivity in the brain are unlikely to be entirely determined by the sex hormone levels. In the UKB sample, we showed that the greater the number of years since menopause, presumably reflecting decreased estrogen levels, the larger the gender brain continuum score, suggesting a shift towards the male end. However, the effect size of this association was small (r = 0.048). Therefore, while sex hormones influence the brain’s functional connectivity many other factors, including those discussed above, also have an impact.

After systematically testing the potential confounders, we confirmed the findings of sex differences in the brain’s functional connectivity. Based on the differences identified, we trained an SVM classifier and mapped each brain onto a brain gender continuum by using the continuous output of the SVM classifier. Some previous studies using cross-validation within the training samples achieved a high classification accuracy (~90%) (Wang et al. 2012; Luo et al. 2019). However, applying such classifiers to the independent test samples, only moderate classification accuracies could be achieved (~75%) (Satterthwaite et al. 2014; Weis et al. 2019), which were comparable with the classification accuracy of 77.75% achieved in the current study. Compared with the low classification accuracy (i.e., 65.7%) in a previous study using a test sample from a different age group compared with the training sample (Weis et al. 2019), our classifier achieved a better accuracy after regressing out age and its higher order terms from the functional connectivity matrix (77.75%). This result was in support of the finding that the sex difference in brain functional connectivity was age dependent.

The moderate classification accuracy of the multivariate classifier indicated that the brain functional architecture was unlikely to be conceptualized as binary, as is the case with biological sex, but was more likely to be continuously represented on a brain gender spectrum. At the behavioral level, Bem had hypothesized that an androgynous gender role would lead to higher self-esteem and better mental health (Bem 1974), since individuals identifying with androgyny are free to act in both masculine and feminine ways without many constraints of gender appropriateness (Bem 1977). In particular, the androgynous group reported having fewer internalizing symptoms (Pauletti et al. 2017). However, previous studies provided only the behavioral observations, therefore there was a need to understand the neural mechanism of such observations. Our results demonstrated that the participants whose brain functional connectivity mapped onto the androgynous segment of the brain gender continuum had fewer internalizing problems, which is advantageous for mental health. This U-shaped association was seen for both males and females, although it was most prominent in males. These findings may indicate that being more compassionate and sociable (traditionally female traits) could potentially improve self-esteem of men, thereby potentially reducing internalizing problems; but being more aggressive and confrontational (traditionally male traits) might not boost self-esteem of women (Pauletti et al. 2017). Future research should include self-report data on male/female behavioral traits within different contexts, for example work, home and social settings, which could further elucidate the relationship between psychological androgyny and the concept of brain androgyny.

However, the current study also has several limitations. First, no single large dataset exists that contains samples covering the entire lifespan, from infancy to old age. In our study, we first analyzed the large-scale multicenter samples from different age groups, and then validated the findings using a single-center sample covering a wider age range but with a smaller sample size. Across this age range, there will inevitably be many environmental factors which will have changed and may have some influence. Second, although the sex hormones have been implicated in the sex dimorphism of the brain’s functional architecture (Bao and Swaab 2011), we need the lifespan measurements of the sex hormones to further investigate the molecular mechanisms underlying the brain gender continuum.