Researchers are able to classify male and female brains with >90% accuracy.

New study suggests that sex differences in functional brain dynamics are highly replicable, generalizable, and behaviorally relevant.

A new comprehensive study published in the prestigious journal PNAS provides some fascinating insights into how female and male brains are intrinsically organized. Employing an innovative deep neural network model, male and females brains were accurately classified with greater than 90% accuracy. This accuracy was achieved with high replicability and generalizability across multiple sessions within the same individuals and three independent cohorts of young adults. According to the researchers, their results refute the “continuum hypothesis” of sex differences by firmly demonstrating binary (male/female) sex differences in the functional organization of the human brain.

The study also uncovered both sex-independent and sex-specific differences in the relationship between functional brain organization and cognition. The most striking finding was that the default mode brain network (DMN) most reliably distinguished between male and female brains. The finding was consistent at both regional and network levels with VERY LARGE effect sizes (d > 2.0). In particular, the researchers identified the posterior cingulate cortex, precuneus, and ventromedial prefrontal cortex nodes of the default mode network as the most consistent discriminators between the sexes.

The default mode network plays a crucial role in integrating self-referential information processing and monitoring the internal stream of consciousness, including introspection, mind-wandering, imagination, the mental projection of one’s self in the future, and autobiographical memory retrieval. Sex differences in the DMN may influence how female and males recall past experiences, form self-concepts, or engage in perspective-taking. The findings may also help explain the higher levels of intrusive rumination and neuroticism see among females on average.

Large differences were also found in the striatum and limbic networks, particularly the orbitofrontal cortex (d > 1.5). The striatum is important for learning cue associations, habit formation, reinforcement learning, and reward sensitivity. The orbitofrontal cortex is involved in learning and reversal of stimulus-reinforcement associations, and correction of behavioral responses when they are no longer appropriate because previous reinforcement contingencies have changed.

The human orbitofrontal cortex is also implicated in representing the reward value, expected reward value, and subjective pleasantness of reinforcers. According to the researchers, this link to subjective pleasantness could provide a basis for investigating the limbic networks’ role in sex differences in hedonic experiences. Prior research has found that women on average report more negative emotions and also more positive emotions than men. So these brain findings may be relevant to that difference.

The study also looked to see whether sex differences in functional brain organization predict cognitive profiles differently in females and males. The profiles were derived from a principal component analysis of a comprehensive cognitive assessment. Their findings reveled three key components: general intelligence, response inhibition and processing speed, and delay discounting and reward sensitivity. While sex differences in general intelligence don’t seem to exist, they found that the brain correlates of general intelligence and these other cognitive variables are different in males and females.

These findings have significant clinical implications and offer important information for researchers of sex differences who investigate vulnerability to individual psychiatric and neurological disorders. Clinical studies in neurodevelopment and psychiatric disorders have consistently pointed out that males display more externalizing problems while females tend to exhibit more internalizing problems. Females are consistently more likely than males to experience depression, anxiety, and eating disorders. On the other hand, males are more likely to develop autism, ADHD, and schizophrenia. Therefore, knowledge of these brain sex differences is critical for understanding both normative behavior as well as psychopathology.

This study has some limitations of course. Most critically, the researchers didn’t investigate the developmental pathways of the brain differences. It must be noted that all brain networks develop as a result of nature interacting with nurture. Socialization processes are certainly at play at an individual level of development. That wasn’t the purpose of the study, however. Their findings suggest that at the group level male brains and female brains can be distinguished with a high level of accuracy and there are particular features of intrinsic organization of these brains that are significant and most likely evolved over the course of human evolution for adaptive purposes. All brains don’t have to be equal for there to be equality. Brains can be complementary with equal opportunities for contribution.

Also of note is that the researchers didn’t distinguish between biological sex and gender identity. If the researchers had measured gender identity (and had enough trans/non-binary participants), then they could have looked at some interesting questions like whether having a brain that was more like that of the opposite sex makes it more likely to have an atypical gender identity. I would certainly like to see more research along those lines. While the large majority of humans feel as though their gender identity matches their biological sex, there are a sizable number of humans who do not feel that way, and I think understanding their experiences and neurodivergent brains is an important and understudied area of research.

Nevertheless, these findings are significant at many levels and highlight the critical role of sex as a biological determinant in human brain organization, with significant clinical implications for developing personalized sex-specific biomarkers for psychiatric and neurological disorders. The research also provides innovative artificial intelligence methods for future research on sex differences.