Neurology Center

Saturday saw the Guardian on fine form with a classic piece of bad neuro-journalism which made it all the way onto the front page:

What the research actually found was nothing to do with looking on the bright side of anything, and was nothing to do with depression either. In fact, it suggests that the gene in question doesn't cause mental health problems. So the headlines are a little misleading, then.

The study comes from Elaine Fox and colleagues from the University of Essex.* They took 111 people, presumably students, and got them to do a "dot-probe" task. Performance on this task was related to the genotype of the 5HTTLPR polymorphism, a variant in the gene which encodes for the serotonin transporter protein. Serotonin is "the brain's main feelgood chemical" as the Guardian put it... except it isn't, although it does have something to do with mood.

What's a "dot probe" task? It's a test which has become popular amongst all kinds of psychologists over the past 10 years or so, having first been used in 1986 by Colin MacLeod et. al. The task involves pressing a button whenever a "probe" - a little dot - appears on a screen. The goal is to press the button as quickly as possible, as soon as the dot appears.

The twist is that as well as the dots, there are other things on the screen. In the 1986 version of the test these were words, while in this experiment they were colour pictures. Some of the images were pleasant: smiling faces, flowers, and other nice things. Some were unpleasant - scary dogs, bloody injuries, etc. And some were neutral objects, like furniture.Pairs of these pictures appeared on the screen for a short time (half a second) immediately before each dot appeared, one on the left of the screen and one on the right. The key is that the dot appeared in the same place as one of the pictures.

The task operates under the assumption that if the viewer's attention is grabbed by one of the pictures, they are likely to be faster to respond to seeing the dot when it appears in the same place as that picture, because they will already be focused on that area of the screen. If, for example, people are on average faster to detect the dot when it appears in the same place as the nice pictures as opposed to the horrible ones, this is described as indicating a "positive attentional bias" i.e. an unconscious tendency to pay attention to pleasant pictures.

Unfortunately, now that you know what a dot probe task is, you can't take part in any psychology experiments which uses one, because once you know how it's supposed to work there's no point in doing it. Sorry. But on the bright side, you now officially know more about psychology than The Economist, whose write-up of this experiment managed to be even worse than the Guardian's. They not only sensationalized the results, but also misunderstand the whole point of the dot-probe task - it's not about "distraction", it's about selective attention-grabbing.

Anyway, that's the task, and the study found that carriers of two "long" variants of the 5HTTLPR gene showed a strong attention bias towards nice pictures and away from nasty ones, while other people showed no biases. Statistically, the result was highly significant, so let's assume it's true. What does it mean? You could take it to mean that carriers of two long variants were more optimistic in that they tend to pay attention to the good stuff. On the other hand you could equally well say they're so squeamish and wussy that they can't bear to look at the bad stuff and have to avert their eyes from it.

And what's this got to do with depression? Well to cut a very long story short the gene in question has been previously linked to depression and also to personality traits such as "neuroticism" - being anxious, worried and generally miserable (see this paper). But in this study they found no such association with neuroticism. Despite the fact that it was a report of this association which got everyone interested in the 5HTTLPR variant in the first place back in 1996! Brilliantly, they spin their negative finding as a good thing -

The fact that our genotype groups were matched on a range of self-report measures, including neuroticism can be seen as a major strength.

Hope springs eternal. Overall, while this paper is a fine contribution to the psychology literature on the dot-probe task (and the results genuinely do seem to be very significant - there's probably something going on here) it's got nothing to do with optimism and little to do with anything that the average newspaper reader cares about. Luckily, we have journalists to make science interesting on the cheap and on the quick - at the cost of accuracy. There's a lot of really interesting, really thought-provoking popular science writing to be done about the dot-probe, and about the 5HTTLRP gene. But none of it has yet made it into the British papers.

[BPSDB]

*Fox, my PubMed search reveals, also does work on so-called "electromagnetic sensitivity". The upshot of her work is that lots of people sincerely believe that signals from mobile phones and other sources make them feel unwell, but actually, it's all the placebo effect. Now that really is something that everyone should find fascinating - much more so than this study, anyway.

Elaine Fox, Anna Ridgewell and Chris Ashwin (2009). Looking on the bright side: biased attention and the human serotonin transporter gene Proc. R. Soc. B