This is a combustible topic, so approach it with caution. A team of researchers from Stanford University (California) examined the brain and asked the question of possible differences between men and women. You found some. Their study, published in the journal PNAS from February 20thwill not fail to provoke a reaction.

Not long ago, mainstream researchers claimed that men were superior to women because their brain volume was larger. Until we understand that the volume of the skull is proportional to height … However, a man with a height of 1.95 meters is not more intelligent than another just because he measures 20 centimeters more. QED.

This time, researchers from the University of California’s Departments of Behavioral Sciences, Neuroscience and Human-Centered Artificial Intelligence didn’t take such a crude approach. No more morphological comparisons! They were interested in the functional organization of our gray matter, not the shape or size of its various parts. Your conclusion is clear: “Our study provides compelling evidence for the existence of reproducible and generalizable sex differences in the functional organization of the human brain. » According to the authors “These brain features predict unique cognitive profiles in women and men, demonstrating their importance for behavior.”.

Differences noted

How did they make such claims? Their basic material was functional MRI images (fMRI) of around 1,500 people of both sexes between the ages of 20 and 35. This imaging technique, which allows for indirect visualization of brain activity, was used in the default mode network, the regions that are active when we are doing nothing and our attention is not needed. “This is what the brain does when the mind wanders back and forth between associations of ideas and memories, for example. This activity at rest is the signature of our brain.”explains Sylvie Chokron, research director at the CNRS and the Laboratory of Psychology of Perception (University of Paris-Descartes), regular collaborator of the World.

To study this material, the American team developed a spatio-temporal deep neural network (stDNN) model that describes it as an explainable artificial intelligence (AI), as opposed to the black boxes that make up some AI. This artificial intelligence was trained on around a thousand cases and then compared with other fMRIs. Result ? In more than 90% of cases, the model was able to distinguish men’s brains from women’s. The differences found in several areas of the brain (default mode network, striatum, limbic network, orbitofrontal cortex) were sufficiently precise to make it a reliable classification tool.

You still have 51.17% of this article left to read. The rest is reserved for subscribers.