Subscriber OnlyScience

How scientists detected differences in the brains of men and women for the first time

Stanford researchers’ breakthrough has not been universally welcomed

Although descriptions of purported differences between male and female human brains have been around for a long time, scientific studies have failed to correlate brain differences with sex – that is, until now.

A research group at Stanford University has just published a paper by Ryali and others in the Proceedings of National Academy of Sciences unveiling a new artificial intelligence (AI) model that can identify whether magnetic resonance imaging scans (fMRI) of brain activity come from a man or a woman.

This breakthrough has not been universally welcomed, particularly on the basis that it can/will be used to bolster biological determinism, the idea that human behaviour is determined by our genes with the environment playing little or no role. I understand this fear, but I think it is exaggerated.

Neural networks used in AI are inspired by the structure and function of the human brain. The Stanford researchers created a deep neural network model and trained it to classify brain imaging fMRI data by showing it very many brain scans, each pre-identified as male or female. The model learned to notice subtle differences between men’s and women’s brains. The model was then tested on 1,500 brain scans where it was not told the sex of the individual scans in advance. In almost all cases (more than 90 per cent) the model correctly identified the sex of the brain scans.


Prof Vinod Menon, leader of the research group and director of Stanford Cognitive and Systems Neuroscience Laboratory, said: “This is a very strong piece of evidence that sex is a robust determinant of human brain organisation.” But Menon emphasised their work is silent on whether sex-related brain differences are driven by hormonal differences or different social circumstances that either sex are more likely to encounter.

Parts of the brain that particularly help the model to distinguish male from female brains include the default mode network, the system that helps the brain to process self-referential information and the striatum and limbic networks that are involved in learning and responding to rewards.

Gina Rippon, emeritus professor of cognitive neuroimaging, Aston University, reacted to the Stanford research in the Guardian, noting the reported brain differences could be due, at least partially and possibly exclusively, to the fact that brains are malleable and changeable throughout our lives.

She illustrated her point by describing how brain scans of musicians can tell whether they are keyboard players or stringed-instrument players – your brain reflects the life you have lived. And, since there is no doubt society treats men and women differently, starting with “pink for a girl, blue for a boy”, this could explain differences in brain organisation.

Rippon acknowledges the Stanford research does not claim to explain how sex-associated differences in brain organisation arise. We must await results of further research to elucidate whether these sex-associated differences are “hard-wired” into brains from birth or explained, partly or wholly, by subsequent environmental moulding.

Antipathy towards “biological determinism”, the proposal that an individual’s characteristics and behaviour are determined exclusively by biological factors, mainly our genes, not by the environment, greatly colours much reaction to research such as the Stanford study. But early crude versions of determinism have been drastically revised in the light of research. We now know that although genes powerfully influence behaviour this is always under modification by the environment.

The brain is so dominantly part of our higher selves that we understandably would hope to be able to wilfully influence its development and capacity over time and not be presented at birth with a completely hard-wired organ whose performance is unresponsive to our best efforts for improvement.

Take the very personal matter of intelligence – is it hereditary? This matter has been studied intensely. Dr Robert Plomin, deputy director of the MRC Social, Genetic and Developmental Psychiatry Centre at King’s College, London, succinctly answered this question in a Scientific American article from May 2016: “Genes make a substantial difference, but they are not the whole story. They account for about half of all differences in intelligence among people, so half isn’t caused by genetic differences, which provides strong support for the importance of environmental factors.”

Finally, and with a sigh of relief, I can report that no differences have been found between men’s and women’s IQs!

William Reville is an emeritus professor of biochemistry at UCC