reporting on science

I like reading the popular science press, especially popular science blogs. I love the groovy magazine Feed (currently on hiatus?), and I really like the wide variety of blogs at Science Blogs (my favorite is the Culture Wars, on the evolution debate). But what I don’t like is when some random blogger or random journalist picks up on a study, assigns it a tagline, attributes causality where there is only correlation to be found, and then uses it to justify random social factors:

ScienceDaily (perfectly fine source) writes about a new study:

The finding led to an unexpected discovery: Many brain areas communicating with the amygdala in men are engaged with and responding to the external environment. For example, the visual cortex is responsible for vision, while the striatum coordinates motor actions. Conversely, many regions connected to the left-hemisphere amygdala in women control aspects of the environment within the body. Both the insular cortex and the hypothalamus, for example, receive strong input from the sensors inside the body.

Sounds interesting, right? Well, sort of. As a non-human biologist, I’m not really sure what this amygdala stuff is all about. But no worries, Wired magazine’s Sex Drive Daily blog is perfectly clear:

It’s why men don’t know they have to eat, they just get cranky until a woman feeds them. It’s why women are generally more in tune with our health, while men are more aware of what kind of motorcycle just roared by.

But because my mind is on sex, I think this could also help account for the truth behind the cliche that men look at porn and women read romance novels.

Uh.. what?

Here’s my response:

This is a very interesting study, but I disagree with your commentary. The researchers found that at rest, men’s brains receive information mostly from the external environment, while women’s brains receive information mostly from the internal environment. That’s interesting, but your conclusions seem very exaggerated and extrapolated.

First, correlation isn’t causality. The researchers never said what this difference actually means in terms of what data people’s brains are processing, what effects this might have on the senses, and so forth. It might be very small.

Second, the effects that you’re attributing, especially romance novels vs. porn, are really socially determined. Not only are you generalizing about men = porn and women = erotica (which I’m not even sure is really accurate anymore), but you’re also assuming that erotica somehow correlates to internal body regulation while porn correlates to external data. While I understand the latter, I don’t understand the former.

Third, it’s very dangerous to take a single study and use it to explain away parts of society that are differently gendered. There are a huge host of factors contributing to gender patterns in society. That is why gender tends to play out differently in different cultures. “It is why it is because it’s natural” is not only usually not scientifically true, but leads us down a slippery slope to justifying social difference based on biological factors, which I think we can agree is a Bad Idea.

I’m not even going to touch your “men get cranky until a woman feeds them..” do they really? Do you assume your readers can’t feed themselves? My guy friends seem to be able to judge their own hunger perfectly well. That’s a fairly condescending view of the opposite sex.

As a social scientist myself, I think it’s fairly important that we look closely at scientific studies that are reported in the popular press and discuss them accurately. It’s all too easy for journalists to pick up a tag line and report it far and wide, while the actual study may have a small sample size, a flawed methodology, unrealistic generalizability or simply does not mean what you think it means.

Deanna Zandt at Alternet also wrote a nice response to this. So here’s my second action item: Don’t let bad interpretation of science stand. Write a comment, make it articulate, and make it clear what the problem is. I’m a big fan of science, and especially ethical, well-researched studies. But I’m not a big fan of them being used to justify gender stereotypes.