Relationship columnists and pop psychologists have long argued that men and women are wired differently, and a new study has proven them right.
Scientists developed an artificial intelligence model that could tell the difference between scans of men’s and women’s brain activity with more than 90 percent accuracy.
Most of these differences are in the default mode network, the striatum and the limbic network – areas involved in a wide range of processes, including daydreaming, remembering the past, planning for the future, making decisions and smelling.
With these results, scientists at Stanford Medicine add a new piece to the puzzle, supporting the idea that biological sex shapes the brain.
The researchers said they are optimistic that this work will shed light on brain disorders that affect men and women differently.
Scientists have long debated whether sex differences exist in the brain. This new study suggests this is possible, if scientists look in the right places
For example, autism and Parkinson’s are more common in men, while multiple sclerosis and depression are more common in women.
“A key motivation for this study is that sex plays a critical role in human brain development, aging, and the manifestation of psychiatric and neurological disorders,” said senior author Vinod Menon, professor of psychiatry and behavioral sciences at Stanford. , in a statement.
This image from the new study shows which parts of the brain are most important in distinguishing between men and women: the striatum and areas involved in the default mode network and the limbic network
“Identifying consistent and replicable sex differences in the healthy adult brain is a critical step toward a deeper understanding of sex-specific vulnerabilities in psychiatric and neurological disorders,” he added.
To investigate the issue of sex-specific brain differences, Menon and his team developed a deep neural network model that could learn to classify brain scans as male or female.
They started by showing the AI a series of functional magnetic resonance imaging (fMRI) scans and telling it whether it was looking at male or female brains.
Through this process, we began to understand which parts of the brain showed subtle differences depending on gender.
A new AI model was trained on images of the brains of men and women. Once trained, it could distinguish between males and females with about 90 percent accuracy
When the AI was given about 1,500 brain scans from a different set than the scans it was trained on, it successfully predicted the gender of the brain’s owner more than 90 percent of the time.
These brain scans were from men and women in the US and Europe, suggesting that the AI model was able to distinguish based on gender, even when there were other differences such as language, diet and culture.
“This is very strong evidence that sex is a robust determinant of human brain organization,” says Menon
A major difference between this team’s AI model and similar models is that it is “explainable.”
Scientists often criticize AI for being a ‘black box’: it can take in information and spit out results, but how it reaches its conclusions is often a mystery.
Not so with the Stanford team’s model.
The main differences the AI model identified were in the default mode network, the striatum and the limbic network
In this study, the team was able to conclude which parts of the brain were most important for the AI to determine a person’s gender.
The three areas the AI focused on were the default mode network, the striatum, and the limbic network.
The default mode network is active when a person is daydreaming, reminiscing, or otherwise thinking about themselves. The striatum is important for coordinating cognition, including planning, decision-making, and motivation. And the limbic network supports a range of brain functions, such as emotion, long-term memory and a person’s sense of smell.
The study appeared today in Proceedings of the National Academy of Sciences.
In addition to distinguishing between men’s and women’s brains, the scientists also tried to see if they could use the scans to predict how well someone would perform on a laboratory test of cognition.
What they discovered was that no AI model could predict everyone’s performance. You could predict men’s performance, and you could predict women’s, but neither could predict both.
This suggests that the characteristics that differ between men and women have different implications for behavior depending on gender.
“These models worked very well because we were able to successfully separate brain patterns between sexes,” Menon said.
“That tells me that overlooking sex differences in brain organization could cause us to miss key factors underlying neuropsychiatric disorders.”