Knowing what not to report is just as important for journalists as knowing what to write. We’re inundated with handouts about some pioneering new scientific research or insightful new book. Should we write about it? It’s refreshing to hear experts who can dazzle you with their work but warn against falling for any hype about it. This “let’s not overdo it” approach has been a recurrent theme in the Neuroscience Boot Camp I’m attending at the University of Pennsylvania in Philadelphia. (Photo: The “official” boot camp T-shirt, 8 Aug 2009/Tom Heneghan)
Andrew Newberg‘s “no God spot” message to boot campers has already been noted here on FaithWorld. Other lecturers added similar reality checks to their presentations. Cognitive science has already begun to influence religion studies (as John Teehan explained here) and we’re bound to hear more in the future about what neuroscientific research has to say about faith, morals, altruism and other issues of interest to readers of this blog. Much of this will be fascinating. But before the next “gee-whiz” report comes out, here’s the advice the neuroscientists are giving us about speculative claims based on brain research.
After two days of explaining fMRI brain scanning, the sexiest procedure in current neurological research, Geoff Aguirre poured cold water on some of the exaggerated conclusions that researchers or journalists draw from it. When shown brain scan images, he said, “people immediately start thinking about trying to catch terrorists and being able to screen people as they pass through metal detectors.” This is “science fiction, science fantasy,” he said, but it comes up regularly. Why? Aguirre, who is an M.D and assistant professor of neurology at Penn, listed several reasons:
scientific awesomeness — “This is an incredible technology. Neuroimaging is not phrenology. It really is a scientific discipline that has reproducible results that makes valuable predictions that explain larges areas of cognition and cognitive neuroscience that previously had been inaccessible.”
image properties — “There’s definitely an esthetic in the presentation of this data. People see this as a natural aspect of the brain, not the result of tests. Some groups made a very wise investment in the display technology for how neuroimaging results were reported. Those were the images that got displayed on the covers of the top scientific journals and made a splash.”
thresholding — The brain images leave out data outside the main focus. “This contributes to the overly localised view of brain function. So we say, ‘ah this is the spot for love’ or whatever, because it’s all that we see.”
overinference — “It’s very easy to believe a lot of things about these images that might not be true… It’s also implied that when you’ve found activisation in a region, you’ve found the region ‘for’ something. But what does that mean?”
chicken versus egg problem — “Just because you find a difference between groups in some brain imaging measure does not mean that structural difference was genetically determined.” But the brain also develops according to its owner’s environment and experience, so this is too narrow a focus.
lurking Cartesian dualism — “In the way we think about people’s actions and describe the effect of diseases or drugs, there is frequently a lurking dualism there. We say, ‘oh it wasn’t his fault, his brain did that.’ Well, who else could it have been? Where else could those thoughts and feeling or plans have come from, except in the brain? This idea that the brain and the mind are separate is part of what makes these images so remarkable. Wow look! Here’s a part of the brain that’s more active when you’re feeling romantic love or not! That’s just astounding to folks who would have thought romantic love was outside the brain, in the heart or the soul and far away.”
(Photo: Near infrared spectroscopy imaging slide/GK Aguirre) illusion of inferential proximity — “It doesn’t automatically follow that a brain imaging technology is going to give you greater inferential leverage on a question than just talking to somebody. There’s an illusion that somehow you’re getting much closer to the behavior you want to measure, just because you’re measuring a brain image. That might not be the case.”
ease of imaging — Many hospitals have brain scanners and researchers can use them and free imaging software to create impressive images. “If you have an internet connection and a scanner, you can be a cognitive neuroscientist and publish a paper. Lots of the variance in the lousy scientific papers over these years can be explained this way. What will come out will be a well-formed brain image that will give the impression you must be a very good scientist because you created something that looks very polished.”
Aguirre said that brain scans might be able to identify pedophiles by showing they are excited by pictures of children. “Does having that response to seeing kids in underwear lead to an increased risk of you actually going out and molesting kids?” he asked. “It could be the case that this population of people now divides into two subgroups, one that can control that impulse and one that cannot.” It would be hard to base a policy on who to put in jail on the basis of such brain images, he said. (Photo: Reward responses slide/Joe Kable)
Another example would be a study into people who lose their temper. “So I do a study of people who are enraged and can find that activity within the right insula is associated with a sense of rage. I have explained the sense of rage,” he said. “But since we all strongly suspected that the sense of rage was derived from events taking place in our nervous system, what have we learned?” The study could say what happens in the brain during rage but still not explain why the person flew off the handle.