Wednesday, December 29, 2010

Dumb Science + Politics = Media Fascination!

The recipe is pretty simple: take some current topic of theoretical interest to many people (e.g. coffee drinking, organic food consumption, political views, amount of daydreaming, or some such), then pick some biological outcome to measure (e.g. death, prostate cancer, brain size, happiness, or some such) and try to see whether they correlate. Then, after you have made the fascinating finding (study finds that coffee drinkers less likely to develop constipation as octogenarians!), you get interviewed by "science journalists" and end up on TV and radio shows, even though most of these "journalists" appear to understand very little about science and have dismally low standards as journalists.

Case in point today comes from Salon as it references an article from the British paper The Telegraph about how different political views correlate with differences in the size of certain parts of the brain. The gist: "conservatives" appear to have larger sized amygdalas; the amygdala plays a central role in the processing of emotional reactions, which The Telegraph article describes as "often associated with anxiety and emotions." Moreover, conservatives have smaller anterior cingulates--a part of the brain the articles describe as "responsible for courage and optimism."

The reality? It's likely nonsense. The study hasn't been published yet (indeed, one very big red flag is the fact that the study findings are being released to the media before going through the theoretically rigorous process of peer review) so I can't comment with great precision on its validity. But it sure doesn't smell right: they took the "brain scans" (unclear from either story whether they were MRIs or CT scans) of two, yes, two Members of Parliament, one from Labour, the other a Conservative, and then reviewed the brain scans of 90 "students" (again, unclear if college-aged or not) and matched the sizes of the brain structures against a questionnaire designed to elicit political views. Again, the questions weren't made available, so it's really hard to know how they measured political values. Even so, the study design doesn't seem especially impressive. Students? Isn't the point of being a student to create one's worldview, meaning that some students will surely drift further toward the right in life, some further left, and some change not much at all? So how can this population be used to measure political attitudes? Is the distribution equal between men and women? Between whites and blacks? Immigrants and natives? Students and non-students? Well, we already know the answer to that question--and couldn't that bias the results all by itself?

Even more concerning is the whopper unloaded by the researcher himself when he said that he and his team were "very surprised to find that there was an area of the brain that...could predict political attitude." But his study predicted nothing at all! They took scans, administered a questionnaire, and then after the fact compared the two. Prediction would entail administering the scan and guessing what the responses would be before they were given. That's a freshman-level mistake and sure doesn't boost my overall confidence in the sophistication of this guy.

But the most odious part of this sorry little news flare is found in the first line of The Telegraph article, which begins, "Scientists have found that people with conservative views have brains with larger amygdalas." What's so singularly frustrating about this opener is that it quite possibly leaves less eagle-eyed readers with the impression that this is both a) unquestionably true because it was "found" like a little fact in the forest, and b) "scientists" are all agreed as to its truth--both of which seem dubious at best. At least in Salon, they change the passive verb formulation "scientists have found" to a more-cautious note of "a study to be published next year suggests."

It's still quite likely prattling nonsense, careful phrasing or no. And as the links supplied in the Salon piece make clear, political bloggers (apparently from both sides of the aisle) have seized on the findings and spun their own theories, each sounding as silly as the study probably is. People want scientific literacy? Find some journalists who have a sense of what constitutes good science and have them leave the junk to languish in throwaway journals, that's one place to start.


  1. The problem is that the journalists who know science don't write the story. They end up being "boring" and get fired. We only end up with the journalists who don't know science, because they don't have the scruples that get in the way of attracting eyeballs.

  2. In complete agreement. At NYT, WSJ, NPR, they do by and large an admirable job of providing context and mostly ignoring dumb-but-sexy stories like this. Problem is if you combine the total audience of these three outlets (even ignoring the fact that there's heavy overlap between them) and you've got less than 10 million or so people--not nearly the reach of CBS, NBC, CNN, and Fox.