Are male and female brains that different? A new way to investigate this question has led us to the conclusion that they exist, but we need artificial intelligence (AI) to tell them apart.

The question of whether we can measure differences between male and female brains has long been debated, and previous studies have yielded conflicting results.

One problem is that men’s brains tend to be slightly larger than women’s. This is likely due to the fact that men are generally larger, and some previous studies have compared the size of various small areas of the brain. Unable to adjust whole brain volume. However, no clear findings have been made so far. “When you correct for brain size, the results change quite a bit,” he says. Vinod Menon at Stanford University in California.

To tackle this problem in a different way, Menon’s team used a relatively new method called dynamic functional connectivity fMRI. This involves recording the brain activity of people lying in a functional MRI scanner and tracking changes in how activity in different areas changes in sync with each other.

The researchers designed an AI to analyze these brain scans and trained it on the results of about 1,000 young people from an existing database in the United States called the Human Connectome Project, identifying which individuals are male and which individuals. told the AI whether the person was female. In this analysis, the brain was divided into 246 different regions.

After this training process, the AI was able to differentiate between a second set of brain scan data from the same 1000 men and women with approximately 90% accuracy.

More importantly, the AI was equally effective at differentiating male and female brain scans from two different, never-before-seen brain scan datasets. Both consisted of about 200 people of similar age, ranging in age from 20 to 35, from the United States and Germany.

“What we bring to the table is a more rigorous study with replication and generalization to other samples,” Menon says. None of the people in the training or testing data were transgender.

“Replication with a completely independent sample from the Human Connectome Project gives us even more confidence in our results,” he says. Camille Williams At the University of Texas at Austin.

The next question is whether the AI will be just as accurate when tested on an additional, larger set of brain scan results. “Time will tell what results we get with other datasets,” he says Menon.

If confirmed, the findings could help us understand why some medical conditions and forms of neurodiversity, such as depression, anxiety, and attention-deficit hyperactivity disorder, differ by gender. No, says Menon.

“If we don’t develop these gender-specific models, we will miss important aspects of differentiating factors.” [for example]”An autistic man and a control man, and an autistic woman and a control woman,” Menon said.