Last Sunday a harsh review of Jonah Lehrer’s new book on the science of creativity, Imagine, appeared in The New York Times. That was followed by a lengthy response from Lehrer and an even lengthier response to that response by the author of the review, Christopher Chabris.
In one sense this is just a spat between an author and a reviewer. But it’s worth looking at closely because it’s also about how science gets communicated and translated, summarized and (possibly) dumbed-down.
Here’s a brief dissection of the back-and-forth (the quotes are from Chabris’s review):
- “Visual information from the left eye does not go only to the brain’s right hemisphere; information from the left visual field does.”
Chabris points out an error. Lehrer acknowledges it. Point Chabris.
- “The different electrodes in an EEG don’t record brain waves of different frequencies; they record from different locations on the scalp.”
Another error that Lehrer admits. Point Chabris.
- “And the enzyme COMT is not involved in producing dopamine; it breaks it down.”
I could find only a single reference to COMT in Imagine, and it’s in a footnote. Lehrer describes it as a gene that “underlies dopamine production.” This seems like unfortunate wording. He’s talking about how DNA-coding differences can lead to “variations in attentional abilities,” and presumably it still holds that variations in COMT (which does indeed break down dopamine, according to multiple sources) would play a role in these variations. Is that a flub worthy of a vigorous lashing in a high-profile review? Nah. Point Lehrer.
- “He uncritically accepts studies whose results support his argument, rarely bothering to discuss whether or how often they have been replicated. On the basis of one experiment, for instance, he claims that ‘being surrounded by blue walls makes us more creative.’”
I e-mailed one of the authors of the study in question, Ravi P. Mehta. Here is what he said:
Note that part of our research is indeed replications of past research on the effect of color on performance. In addition, in this paper, we also tried to use different tasks across various studies to theoretically replicate our results. Of course, like any new findings, replication and extension of our research in future endeavors are much needed and will enrich our understanding of how our contexts, such as color, can affect the way our mind works. That being said, you cannot negate importance of new findings.
I read the Lehrer section in question and Mehta’s paper. It seemed like a fair summary to me, and Mehta thinks extrapolating the effect to include wall color is not unreasonable. Point Lehrer.
I could go on and on because, in his response to Lehrer’s response, Chabris points to more supposed errors. Are they nitpicky? Maybe. But there’s value in someone like Chabris, a serious researcher who also co-wrote the popular book The Invisible Gorilla, scrutinizing a book like Imagine. He’s right that details matter. And Lehrer wouldn’t deny that (in fact, he acknowledges as much in his response).
The heart of Chabris’s critique, though, is that Lehrer doesn’t sufficiently acknowledge the provisional nature of science, that for effect (blue walls make you creative!) he breezes past dull notions like replication. Lehrer’s defense is that he is fully aware that, as he puts it, “our current science is very much a first draft,” but that you can’t weigh down every example with a page full of caveats and expect a normal person to read it on an airplane. This is a perennial tension between scientists and science writers, and a debate takes place every time someone like Lehrer or Malcolm Gladwell publishes a blockbuster.
Speaking of Gladwell, probably the biggest problem with Lehrer’s book isn’t something Lehrer wrote. Instead, it’s a blurb from Gladwell that says Lehrer “knows more about science than a lot of scientists.” That’s a ridiculous thing to say, even by blurb standards, and I can’t imagine Lehrer concurs. It’s the kind of hubris that could get a science writer into real trouble.