By Tom Stafford
The headlines
The Guardian: Male and female brains wired differently, scans reveal
The Atlantic: Male and female brains really are built differently
The Independent: The hardwired difference between male and female brains could explain why men are ‘better at map reading
The Story
An analysis of 949 brain scans shows significant sex differences in the connections between different brain areas.
What they actually did
Researchers from Philadelphia took data from 949 brain scans and divided them into three age groups and by gender. They then analysed the connections between 95 separate divisions of each brain using a technique called Diffusion Tensor Imaging.
With this data they constructed “connectome” maps, which show the network of the strength of connection between those brain regions.
Statistical testing of this showed significant differences between these networks according to sex – the average men’s network was more connected within each side of the brain, and the average women’s network was better connected between the two hemispheres. These differences emerged most strongly after the age of 13 (so weren’t as striking for the youngest group they tested).
How plausible is this?
Everybody knows that men are women have some biological differences – different sizes of brains and different hormones. It wouldn’t be too surprising if there were some neurological differences too. The thing is, we also know that we treat men and women differently from the moment they’re born, in almost all areas of life. Brains respond to the demands we make of them, and men and women have different demands placed on them.
Although a study of brain scans has an air of biological purity, it doesn’t escape from the reality that the people having their brains scanned are the product of social and cultural forces as well as biological ones.
The research itself is a technical tour-de-force which really needs a specialist to properly critique. I am not that specialist. But a few things seem odd about it: they report finding significant differences between the sexes, but don’t show the statistics that allow the reader to evaluate the size of any sex difference against other factors such as age or individual variability. This matters because you can have a statistically significant difference which isn’t practically meaningful. Relative size of effect might be very important.
For example, a significant sex difference could be tiny compared to the differences between people of different ages, or compared to the normal differences between individuals. The question of age differences is also relevant because we know the brain continues to develop after the oldest age tested in the study (22 years).
Any sex difference could plausibly be due to difference in the time-course of development between men and women. But, in general, it isn’t the technical details which I am equipped to critique. It’s a fair assumption to believe what the researchers have found, so let’s turn instead to how it is being interpreted.
Tom’s take
One of the authors of this research, as reported in The Guardian, said “the greatest surprise was how much the findings supported old stereotypes”. That, for me, should be a warning sign. Time and time again we find, as we see here, that highly technical and advanced neuroscience is used to support tired old generalisations.
Here, the research assumes the difference it seeks to prove. The data is analysed for sex differences with other categories receiving less or no attention (age, education, training and so on). From this biased lens on the data, a story about fundamental differences is also told. Part of our psychological make-up seems to be to want to assign essences to things – and differences between genders is a prime example of something people want to be true.
Even if we assume this research is reliable it doesn’t tell us about actual psychological differences between men and women. The brain scan doesn’t tell us about behaviour (and, indeed, most of us manage to behave in very similar ways despite large differences in brain structure and connectivity). Bizarrely, the authors seem also to want to use their analysis to support a myth about left brain vs right brain thinking. The “rational” left brain vs the intuitive’ right brain is a distinction that even Michael Gazzaniga, one of the founding fathers of “split brain” studies doesn’t believe any more.
Perhaps more importantly, analysis of how men and women are doesn’t tell you how men and women could be if brought up differently.
When the headlines talk about “hardwiring” and “proof that men and women are different” we can see the role this research is playing in cementing an assumption that people have already made. In fact, the data is silent on how men and women’s brains would be connected if society put different expectations on them.
Given the surprising ways in which brains do adapt to different experiences, it is completely plausible that even these significant “biological” differences could be due to cultural factors.
And even reliable differences between men and women can be reversed by psychological manipulations, which suggests that any underling biological differences isn’t as fundamental as researchers like to claim.
As Shakespeare has Ophelia say in Hamlet: “Lord, we know what we are, but know not what we may be.”
Read more
The original paper: Sex differences in the structural connectome of the human brain
Sophie Scott of UCL has some technical queries about the research – one possibility is that movements made during the scanning could have been different between the sexes and generated the apparent differences in the resulting connectome networks.
Another large study, cited by this current paper, found no differences according to sex.
Cordelia Fine’s book, Delusions of gender: how our minds, society, and neuro-sexism create difference provides essential context for looking at this kind of research.
UPDATE: Cordelia Fine provides her own critique of the paper
Tom Stafford does not work for, consult to, own shares in or receive funding from any company or organisation that would benefit from this article, and has no relevant affiliations.
This article was originally published at The Conversation.
Read the original article.
Mind Hacks 
I would really like to see a study which examines if the brains of neuroscientists who believe there is a fundamental difference between men and women are wired differently than those of neuroscientists who don’t.
I have a few comments on your comment. I will say I haven’t read the original paper – it’s behind the PNAS paywall, and I don’t buy papers. I’m a retired software developer, not a current academic whose university library has access to all that stuff.
First, the number of scans is, by itself, significant. Most studies I’m aware of have n somewhere in the vicinity of a dozen – this is a very common criticism of brain scan results. 949 scans suggests that they took the raw data from scans done for other projects around the university and analyzed them, rather than doing the scans themselves for their project.
The age cutoff – 22 – also suggests they used existing data from university undergraduates and younger collected for other projects. In fact, I feel reasonably confident that the biases in the data are WIERD – White, etc, which goes to address your point about not having social status data to filter against.
As far as plausibility goes, my bias is exactly the opposite of what yours seems to be. I’ve been an onlooker to the transgender community for a long time, so I can say with some confidence that the notion that gender-related differences are entirely socially constructed is a crock of you know what – that notion is more likely to be socially constructed.
Criticizing possible effect size issues, and then turning around and saying that it’s rational to believe what the researchers have found, is a cheap shot. Either you have reasons to believe there are problems with the effect size or you don’t.
You say that sex differences could be plausibly assigned to the time course of development. Of course! Puberty and adolescence are when major rewiring of the brain happens. It’s also when sexual characteristics show up and most (not all) gender differences show up behaviorally. The very plausibility is why that should be the null hypothesis – deviations from it should be the unexpected result. The fact that even the researchers say they were surprised says, at least to me, how badly the priors have been messed up by the completely unjustified assumption that there are no inherent gender differences.
For the research to “assume the difference it wants to prove,” the assumption has to be up front. According to the researchers, this isn’t the case. The entire rest of this passage is, in my opinion, another cheap shot to try to make people who think there are differences think they may be wrong.
I agree that the reported differences don’t say anything in detail about specific behavior. Nobody who knows the debates going on in the field, or the way brain scan studies are commonly misinterpreted in the press, would think it’s possible. That doesn’t mean that there can’t be a pervasive influence on behavior, though.
The Sophie Scott comment appears to be reaching for any possible stick to beat it with – it uses a common criticism of brain scan technology – movement in the scanner – to launch a completely contrived hypothesized sex-related difference to try to deny sex-related difference.
As far as the Cordellia Fine book is concerned, as the Guardian article you reference itself points out, there are significant differences of opinion in the field. The book appears to lack any pretense of objectivity.
Back up to the title. On the map reading thing, there are well-known differences between the sexes in mental rotation, which can be overcome by concentrated practice. While I’m not aware of studies on map reading, I find it entirely plausible that differences in the ability to do mental rotations could affect the ability to read maps, at least for some tasks.
I’m only qualified to talk about the acquisition part of the methods, but there is sufficient cause here for me to want to wait for new information before considering the results further:
https://pubpeer.com/publications/3CFCCE950D22E7560E9B07C8B63979
Insistence on gender factualism in the face of compelling evidence to the contrary is, frankly, anti-scientific (to the tune of fundamentalism). _Certainly_ *some* aspects of gender differences *are* socially constructed. This is a far cry, however, from asserting that *all* such differences are due to social conditioning.
The paper to which you refer is but the tip of the iceberg as regards physically measurable differences between genders (and, interestingly enough, the evidence supports the APA, AMA, and BPS, to name a few, as regards the reality of Transsexuality on a *physical* level, even prior to hormone treatment). The following page supplies links to a number of articles in a number of scholarly, peer-reviewed, scientific and medical journals which indicate, rather convincingly, that while *some* gender differences are indeed due to social conditioning, others are, quite the contrary, due to physiology:
http://aebrain.blogspot.com/p/transsexual-and-intersex-gender-identity.html
You might also like to consider this:
http://www.newscientist.com/article/dn20032-transsexual-differences-caught-on-brain-scan.html#.UpzD0OKFfis
Ciao.
@John Roth – thank you for the detailed comment with many reasonable points. My issue is less to weigh in on the topic of whether there are inevitable sex differences, but to say that this study doesn’t provide information on this either way (and yet is reported as such).
We could, if we wanted, get into the sorts of evidence that would inform this debate: developmental, cross-species, etc. For now, I just wanted to note the overclaim between a study of encultured brains and the authors’ and press’s claims about hardwired psychological differences between males and females
Thanks, Tom.
Since I don’t have access to the paper, I can’t comment on the adequacy of their statistical workup. I’ve since seen a couple of other comments from people who do have access who raise the same questions. This isn’t helped by PNAS’s reputation for rushing papers by senior members into print without adequate peer review.
I thoroughly agree that press reports tend to be grossly over-simplified and give entirely the wrong impression of the actual situation, at least as far as it’s known. Some of the reported comments by members of the research team make me wince, too.
I suspect my own position gets lost in the muddle. It’s rather simple, if not simplistic. Gender affiliation is as innate as sexual preference, at least to the first approximation. That is, most people affiliate with the sexually appropriate gender, but there’s enough slop going from genotype via epigenetics, brain structure, brain plasticity and cultural imprinting to actual behavior that it’s hardly 1 to 1.
What gets up my nose is the position that it’s all cultural imprinting, but discussing it tends to polarize and ignore the all-important nuances. Also, much of the commentary and argumentation tends to not address what I consider the rather narrow point of whether gender affiliation is innate, rather it emphasizes areas that are indisputably cultural imprinting.
“That is, most people affiliate with the sexually appropriate gender, but there’s enough slop going from genotype via epigenetics, brain structure, brain plasticity and cultural imprinting to actual behavior that it’s hardly 1 to 1.”
Probably that language isn’t helpful language to discuss this issue. Slop? Appropriate? I’d check that normativity.
The “Another large study” link is.
Anecdotal evidence is also inconclusive: I can’t tell if men are better wired to read something they refuse to use.
At the risk of wading into a quagmire, it’s hard to take this study’s findings seriously, that there are new and significant hardwired differences in human male and female brains in additional to what we already knew, because the authors do not explain all the factors involved in why they found what they did.
For example, can we raise kids in our culture along typical gender roles and biases, then at ages 12-14, say that the differences in their brains are solely due to their genders?
To do so would be to ignore what is known about epigenetic and environmental influences in shaping the brain.
http://surfaceyourrealself.com/2015/02/15/problematic-research-on-hardwired-differences-in-human-male-and-female-brains-surfaceyourrealself/