The latest front in the gender wars

The research is carried out using functional magnetic resonance imaging, fMRI. That means that scans are taken of brain activity to see how different parts of the brain are connected: their “functional connectivity”. It found that, as the foetuses’ brains grew, different parts of the brain became more strongly connected in male foetuses than in female foetuses.

There are some things to note. First, fMRI studies are tricky. (I imagine they’re even trickier in foetuses.) They’re expensive, which means that the sample sizes are usually small, and the data is messy: those neat ‘brain area lights up’ pictures you see are usually an oversimplification.

This study used 118 subjects, of which 48 were female and 70 male. It’s quite a big sample by fMRI standards, but by the standards of, say, a genome-wide association study looking at genetic links to physical traits, which would probably look at tens or hundreds of thousands of people, it’s tiny.

Correspondingly, its results are shakier. The solidity of scientific results are measured in “p-values”. If you want to understand them in a bit more detail, read this, but essentially they’re a measure of how likely it is that you’d see a given result by random chance. The lower the p-value, the less likely it is that a finding is pure fluke.

By convention, findings with a p-value of less than 0.05 are considered “statistically significant”, i.e. probably real. You just won’t get a paper published with a higher p-value than that. And of the 18 p-values quoted in the paper, all but two are “p<0.05”, which means they were just about significant. This isn’t the fault of the researchers, because, as mentioned, it’s really hard to do good research with fMRI, but it does mean we have to be wary.

The other thing worth noting is that “statistically significant” is a really bad phrase, because it doesn’t actually tell us whether a finding is significant in the sense that you or I would use it. An effect can be real and statistically detectable, but so small as to be unimportant. This paper doesn’t tell us how big the effects are, as far as I can tell. It just quotes p-values. If the “effect size” is tiny, as I mention here in another context, then we probably don’t care about it very much.

But most importantly, in one sense, we don’t care about these results anyway. I don’t care what your brain connectivity is like, whether its posterior cingulate-temporal pole and its frontal cerebellum are particularly strongly connected, except insofar as it will tell me about how you behave. This study (obviously) doesn’t tell us anything about how the owners of these brains behave, because they’re all foetuses.

Male and female bodies are different, and different bodies need different brains to control them. So it’s possible that these different connections aren’t connected to behaviour. That’s probably not the case, but we don’t know. And we already knew that male and female brains are innately different: for a start, the average adult male brain is about 115 cubic centimetres bigger by volume, which is, let’s face it, unlikely to be the product of societal prejudice.

I said earlier that this is a weapon in a proxy war. What I meant by that was: no one cares about whether female brains have an innately different pattern of connections to male brains. What they care about is whether innate differences between male and female brains are behind different male and female participation in the labour force, the greater burden of unpaid care that falls to women, male violence, the gender pay gap, and so on. This study doesn’t, directly, tell us anything about that, although, if it is replicated, then it may be a useful hint that there are real differences even at that early stage.

But it will be used by one side of the war as a nice easy quotable quasi-fact. ‘Look, male brains and female brains are different in the womb! This study says so! Therefore the gender pay gap isn’t real.’

This goes on quite a lot. There was a piece in The Sunday Times this week about a physicist who’d been fired for saying that women are less likely to become physicists because they’re less interested in physics. A thread went viral on Twitter about it: he’d said in the piece that people weren’t “building walls” to keep women out of physics, and the thread listed various studies which suggested that there were in fact such walls.

It included, among many others, studies that found that faculty members rated female-named applicants lower than male-named ones; that women are rated less competent for similar-quality work; and female researchers are less likely to be invited to speak at conferences than male ones

You should read the thread: it’s startling. And I have little doubt that most if not all of the results are real. But it would be easy to put together another thread which would, to an untrained observer, suggest the exact opposite. The more gender-equal a country, the fewer women go into STEM careers, suggesting that women, given greater freedom, tend not to choose science as often as men. There is a “very large” gender difference in interest in people vs interest in things, which would also explain those choices. There are “moderate” or “large” gender differences in computer skills, spatial visualisation and perception, the ability to mentally rotate an image, assertiveness, and “tendermindedness”. (All those last ones are from a meta-analysis which, ironically enough, went viral in an article that claimed the differences between men and women are “vastly exaggerated”.)

If I wanted to say that the reason only 20% of A-level physics students are female is discrimination, I’d pick a few from thread A. If I wanted to say that it’s because of innate gender differences, I’d pick thread B.

But: none of these things contradict each other. Women could (on average) be less likely to choose physics because they’re more interested in working with people than with things, and yet those many women who do choose physics could face real discrimination. It is perfectly possible – hugely likely, I’d say – that it’s both!

Obviously the question is how much discrimination affects things. I don’t know, although I would say that however much it does affect things, it probably affects things fairly early on, because you see boys dominating the physics-maths-computer science stuff at school level and university. So it seems likely that it’s a society-wide problem rather than something we can specifically blame on sexist Reddit nerds or tech bros in Google.

Probably female and male foetuses do have different brains. And probably men and women do have innate differences in behaviour, just as chimps do and rhesus monkeys do. But working out how much those differences affect the real-world differences in social outcomes that we see is really hard. And knowing that the somatomotor area might be more strongly connected to the posterior parietal lobe in female foetuses doesn’t help us do that.

---

Tom Chivers is a science writer. His second book, How to Read Numbers, is out now.

TomChivers