Sex differences in brain connections found in a recent PNAS article have caused a media stir. The BBC reported that “men and women’s brains are wired differently” and that this might result in differences between what each sex does well. A backlash of comment debated whether this is a helpful position for gender equality, and whether the media story had covered the science accurately enough.
We seem to have endless appetite for knowledge on whether or how the sexes differ, with whole journals dedicated to studies on the subject. Yet we are also super sensitive when people do find differences. Perhaps this research feels too close to the provocative and limiting ‘men are from mars’ argument.
Maybe there is also something about physical brain data that feels more compelling or ‘true’ than behavioural data. Dr Michael Bloomfield, Clinical Research Fellow at the Medical Research Council Clinical Sciences Centre, and one of the experts who commented in the BBC article, reflected on the coverage that the PNAS paper had: “examining biological differences between the sexes can perhaps appear to some to be at odds with political movements of equality, as a mental paradox can be formed e.g. ‘If we are different, how can we be equal?’ or vice versa.” Bloomfield referred to wider research on how mental illnesses tend to be expressed differently in men and women, with hopes that this type of research into neuronal sex differences might help us understand this and help more.
The PNAS paper itself is not an easy read, largely due to the amount of statistical manipulation involved in using brain imaging data. Nonetheless the BBC article, in my view, did quite a good job of summarising its findings. The paper did say that they found statistically significant sex differences between men and women, and also looked at different age points and saw these differences developing from childhood. The paper’s authors also suggested that these differences might relate to differences in functional abilities between sexes.
Important caveats the media coverage could have emphasised more were that even though the study was reporting general statistically significant differences between two groups (men and women) there will still be overlaps between the two groups, so not all men will be different to all women. The PNAS paper also didn’t report the effect sizes i.e. the magnitude of the difference between the sexes or the degree to which the samples overlapped.
The study used 949 people, which it described as a “large sample”, but it compared both between sexes and between three age ranges, compromising the size of the differences it could find by looking at more variables. It also gave little information on the participants, other than sex and age, so other confounding variables could have been responsible for the differences seen between the sexes.
In addition, the technique used doesn’t take into account neurotransmitter activity, as Bloomfield pointed out. Further expert comments within the BBC article highlighted the potential for environment to influence brain development, the capacity for brain plasticity, and the gaps in our knowledge about what the connectivity studies imply.
The paper didn’t, despite what the BBC article suggested, directly look at associations between different patterns of brain connectivity between the sexes and different performance on cognitive tasks. Instead, the researchers obtained their participants from a larger sample who had undergone behavioural tests and extrapolated their findings to relate brain differences in their subgroup to functional differences in the wider sample. It also wouldn’t follow that all men’s behaviour will be different in all circumstances, because again we’re talking about general differences between samples,.
Without matching data for specific participants it is hard to really relate the brain data and behavioural data in this way. Even if there was an association between brain connectivity and behaviour we still wouldn’t be able to rule out the possibility of environment affecting brain structure through the way different sexes are treated or through more complex epigenetic effects.
Ultimately , this paper is a very small part of a bigger picture that we don’t fully understand. It is an interesting finding, partly because of the debate it stimulates, but it isn’t conclusive proof of a difference between the sexes at the behavioural level, or even enough information about how much of a difference is present at the physical level.
For more discussion on this see Dorothy Bishop’s ‘Storify’.