The Curse of Knowledge
June 13, 2012 7:38 AM   Subscribe

Isaac Chotiner reviews Jonah Lehrer's Imagine: How Creativity Works. Imagine is really a pop-science book, which these days usually means that it is an exercise in laboratory-approved self-help. Like Malcolm Gladwell and David Brooks, Lehrer writes self-help for people who would be embarrassed to be seen reading it. For this reason, their chestnuts must be roasted in “studies” and given a scientific gloss. The surrender to brain science is particularly zeitgeisty.
posted by shivohum (29 comments total) 17 users marked this as a favorite
 
This blog post convinced me that Jonah Lehrer's presentation of data can't be trusted.
posted by John Cohen at 7:52 AM on June 13, 2012 [4 favorites]


A new incarnation or three of this book comes around every year. I almost never read them, but I usually get a kick out of the detailed angry takedown essays.

This almost certainly makes me a snob and a bad person, but it does leave me with somewhat more time on my hands than everyone else I know, who are actually reading the books instead of just being pleasurably disgusted by them at one remove.
posted by brennen at 8:07 AM on June 13, 2012 [4 favorites]


So he's like the Thomas Friedman of neuroscience?

I always have mixed feelings about this type of stuff. On the one hand, I can say without qualification that people like Lehrer, whose greatest skill seems to be being wrong about science in a way that is tremendously compelling to lay-readers, drive me up the fucking wall. Even worse is that that ability is usually mistaken for penetrating insight by those same assholes, and who doesn't start feeling a little stabby when presented with somebody who's as glib as they are wrong? On the other hand, it's tempting to extend that resentment to the people that read those books, and I think that's wrong, because I think even the most tirelessly skeptical among us has probably eaten some tasty-sounding bullshit, and in the end I don't think there's really much harm done by a bunch of people putting stock in an elegant but inaccurate idea like left/right-brain thinking. Hell, it might even be a useful tool for introspection. So I guess what I want is for that anger I feel to be a righteous anger, but that seems harder and harder to justify the more I think about it.
posted by invitapriore at 8:25 AM on June 13, 2012 [2 favorites]


This blog post convinced me that Jonah Lehrer's presentation of data can't be trusted.


Whether Jonah Lehrer can be trusted or not is debatable but the person who wrote that article doesn't really understand statistics when he goes on and on about the median not being a group measure. I quote:

But now that I realized he really meant median, and that maybe he didn’t know what median meant. Because median guesses are not guesses by a crowd, as Lehrer states. They are guesses by a single person.

and later:

In his WSJ article Lehrer is using a number [the median] generated by one person, to one question, instead of by 144 people to 6 questions.

posted by peacheater at 8:33 AM on June 13, 2012 [7 favorites]


I guess I reserve my righteous anger for pop science publishers. There are many actual working scientists who are nominally interested in writing pop-science books, but don't have the name recognition of 'journalists and public intellectuals' like Lehrer. Maybe, at worst, Lehrer could ghost-write for the actual scientists who actually did the research?

Whether Jonah Lehrer can be trusted or not is debatable but the person who wrote that article doesn't really understand statistics when he goes on and on about the median not being a group measure.

He actually addresses that debate in future blog posts, if you're actually interested in his opinion. It's actually an incredibly incidental part of his argument, which is that Lehrer seriously mis-read the table and cherry-picked the one bit that proved his point, while ignoring the rest of the paper.
posted by muddgirl at 8:39 AM on June 13, 2012 [2 favorites]


Lehrer, whose greatest skill seems to be being wrong about science in a way that is tremendously compelling to lay-readers

Maybe if he tried using more than ten percent of his brain, he could do better!
posted by thelonius at 8:42 AM on June 13, 2012 [4 favorites]


Maybe if he tried using more than ten percent of his brain, he could do better!

He's probably also not drinking 8 full glasses of water a day! Room for improvement!
posted by kiltedtaco at 8:45 AM on June 13, 2012 [2 favorites]


Whether Jonah Lehrer can be trusted or not is debatable but the person who wrote that article doesn't really understand statistics when he goes on and on about the median not being a group measure.

I was bothered by this too. He seems to think understanding what "median" means requires some profound mathematical genius that is far beyond WSJ readers to comprehend. It's metaphysics, so he has to diagram it out. His point could have been explained in a paragraph. Two, if you didn't already know what a median was.
posted by justkevin at 8:45 AM on June 13, 2012 [1 favorite]


I think the presentation of these as self-help is apt; like self-help, they're attractive, fun to read, and briefly make you feel like you have achieved some profound insight. Like self-help, the feeling fades quickly when reality remains complex.

My experience with them is that after this happened a few times, I learned to spot them and stop being sucked in by them.
posted by emjaybee at 8:47 AM on June 13, 2012


He seems to think understanding what "median" means requires some profound mathematical genius that is far beyond WSJ readers to comprehend.

Doesn't he concisely explain median to lay-readers in the original critique? It's the middle answer of all your answers.

The meaning of a median is, of course, much more complex.
posted by muddgirl at 8:49 AM on June 13, 2012


Well, phraseological idiosyncrasies and metaphysical implications aside, I think his argument gets into shaky territory when he tries to go beyond "Lehrer's hypothesis isn't supported by these data because the error is too high" to "Lehrer's hypothesis isn't supported by these data because the median isn't a meaningful statistic in this context." The first argument is strong enough, and the second one isn't air-tight the way he thinks it is.
posted by invitapriore at 8:53 AM on June 13, 2012 [1 favorite]


OK, I'm going to pull a Godwin II here: MALCOLM GLADWELL.

Good takedown of yet another mashup of creativity and neuroscience (can't somebody stop these people? I'll give Oliver Sacks an exception to this particular category of awfulness.). Worth reading to the last sentence: "this book is a failure of the imagination."
posted by kozad at 8:55 AM on June 13, 2012 [1 favorite]


I think this is the gist of his argument:
Not only had he cherry picked a single value out of a column of values, but he had chosen a column of values the authors did not use as their dependent measure; the authors explicitly say that it is the geometric mean that should be used to judge crowd wisdom.
The rest of his critique is more a philosophical look at pop-science journalism, and less about this particular error.
posted by muddgirl at 8:56 AM on June 13, 2012 [1 favorite]


Yeah, agreed, and I think his case against Lehrer's presentation there is totally solid. It just makes me sad to see the median being shat upon unnecessarily. The median is my friend!
posted by invitapriore at 9:02 AM on June 13, 2012 [3 favorites]


I haven't read Imagine, but I have enjoyed some of Lehrer's other writing. And the linked review really annoyed me with its too-easy critique of pop science. This part especially rubbed me the wrong way: IMAGINE is a collection of stories—all pop-science these days must be translated into stories, as if readers, like children, cannot absorb the material any other way;

Isn't the whole point of pop science to try to lightly explain the basic concepts of esoteric fields to non-experts? And is there any argument that anecdotes and stories are far more likely to reach a wide audience than 'any other way'? (I'm guessing this is a reference to 'hard science' writing).

The criticism of Lehrer and Gladwell is completely misplaced. Of course they are going to rely on anecdotes. And of course that means that nuances are going to be missed. And of course experts in the field (and snobs generally) are going to have problems with the work.

But if you actually bother to read and consider works like Outliers or How We Decide, and imagine doing so from the perspective of someone who knows nothing about those fields (and make no mistake, those are generally the audience for these books), you can see both their appeal and their usefulness.

For that audience, these books provide an excellent primer to fields of study they'd otherwise never learn about. The readers won't become experts, and sure, they're going to walk away with an oversimplified idea of those fields, but they will be exposed to new ways of thinking about their world.

Chotiner might be spot-on about Imagine, and the post that John Cohen referenced earlier may be totally true. But the snobby dismissal of pop science and pop sociology just comes across to me as smug and elitist.

This doesn't relate specifically to Lehrer, but it's worth noting that the American Sociological Association has formally honored both Gladwell and Brooks for their reporting.
posted by graphnerd at 9:06 AM on June 13, 2012 [3 favorites]


Isn't the whole point of pop science to try to lightly explain the basic concepts of esoteric fields to non-experts?

Yes. I don't think anyone reads these things and goes off to open a neuroscience clinic. This kind of criticism, and its fans, is a standard "I liked this band before they were cool". It's the kind of snobbism nerds complained about in high school but can't see when they do it. Everybody needs to start somewhere and if the worst problem in your life is people reading and having their curiosity piqued by books that don't stand up to graduate school-level criticism, well, I'm envious.
posted by yerfatma at 9:10 AM on June 13, 2012 [1 favorite]


I would have much less of a problem with 'just-so science stories' if at least the underlying science was correct.

Otherwise we are back to the argument that I despise: That it's OK to lie to or mislead people as long as we're 'making it a teachable moment' or 'changing hearts and minds.' If I am elitist because I reject that concept, then I am proud to be elitist.
posted by muddgirl at 9:13 AM on June 13, 2012 [5 favorites]


For that audience, these books provide an excellent primer to fields of study they'd otherwise never learn about.

I periodically binge on pop-science writing. It's one of my favorite genres. I'll even ignore the standard hate-on and say that I've gotten some good stuff out of, say, some of Gladwell's work. I think the argument here is that the specific book at hand isn't so much an excellent primer.

Again, I probably am a bad person and a snob, but I think that has more to do with how much I enjoy snarky reviews than it does with how I have a strong preference for coherence, consistency, and approximate correctness in popularizing authors.
posted by brennen at 9:18 AM on June 13, 2012


I attended a conference where he was a plenary speaker. I really enjoyed his talk, which was based on this book. I admit I haven't read the book yet as it's not my normal genre and the reviews have been just awful but at the same time, the talk about innovation and creative thinking was really inspiring and interesting. I'd like to read the book one day. Nothing was truly groundbreaking but it was a great reminder (esp for those who could be my boss in the future) that it's okay NOT to multitask, that sometimes you need to stare out the window and think or do something completely different to shift your creative gears and let your subconscious munch on a creative problem for awhile. It's why so many people have shower epiphanies.
posted by 1000monkeys at 9:35 AM on June 13, 2012


I don't mind storytelling in science writing.

I do mind when authors use stories like this as parables, if that makes any sense.

I mean, the parable thing is not new. Every undergrad in the world has learned about the Stanford Prison Experiment since ... well, since it happened, I guess. When it's taught well, it's an opportunity to talk about methodology ("Was there a control group? Could there have been? Does it matter?") and experimental ethics ("Should someone have stopped it? Is it bad that the same experiment couldn't be done today?") and to see whether there are any real conclusions you can tease out of that mess of an "experiment." When it's taught badly, it's just a parable about authority, an opportunity to pontificate or shake your head and go "Well, it just goes to show." It's no longer a conclusion that you're trying to extract from the story, but a moral.

I think the important question is, would your audience feel differently about this story if they discovered that it was fictional? If your mention of the Stanford Prison Experiment could be replaced by a story from Kafka — or your mention of Bob Dylan by a story from oh let's say Goethe — without changing your point, then you're no longer doing science writing. You're just telling tales with morals tacked on. Which is a fine and noble activity, but let's be honest about what it is.
posted by nebulawindphone at 9:48 AM on June 13, 2012 [6 favorites]


graphnerd: And of course experts in the field (and snobs generally) are going to have problems with the work.

yerfatma: Everybody needs to start somewhere and if the worst problem in your life is people reading and having their curiosity piqued by books that don't stand up to graduate school-level criticism, well, I'm envious.

It's really not evident to me that major factual errors, as opposed to mere over-simplifications*, are a necessary component of pop-science writing. And, you know, going back on my first comment here, there is damage to be done by writing pop-sci primers that are wrong in the glib way that Lehrer likes being wrong, because I don't think people have their "curiosity piqued" by these books. I think they read them and put them down feeling (and not wrongly) that they're now conversant in basic neuroscience, and, hey, wasn't that easy, which I think primes people to be amenable to simple solutions for complex problems in general**. I don't think that's a healthy mindset.

* Applying a succession of over-simplifications that are later overturned by more sophisticated concepts is the default pedagogical model for teaching a lot of science, after all.

** There's a chicken-and-egg question here about whether these books encourage such a mindset or whether the already-existing presence of that mindset encourages the hunger for these books, but regardless of their origins it isn't too much of a stretch to surmise that the two phenomena feed on each other.

posted by invitapriore at 9:51 AM on June 13, 2012


If you have a shower epiphany you're probably wasting water.

And yes, I am not only a faddish eco-nut (ever since reading Silent Spring as a kid fifty years ago), but an elitist and a snob. In defense of these qualities, though, they extend only to my personal tastes, which I generally inflict only semi-anonymously, on forums such as this.

Some of my best friends say "My kid could do that" when they look at a Mark Rothko painting. I'm incredibly opinionated when it comes to art/music/writing, but I don't think that my opinions are better than everybody else's. And yet, I could never be a local art critic, for example. I wouldn't want to hurt anyone's feelings. I may be an elitist in my tastes, but my elitism just extends to the artwork itself. I might have a good friend who likes Thomas Kinkade...but come to think of it, probably not. Maybe I am kind of a snob. What can I say? I was raised that way?
posted by kozad at 9:52 AM on June 13, 2012


I'd point to Oliver Sacks and V.S. Ramachandran as people who do this sort of pseudoneuroscientific or pop-neuroscientific storytelling well. They're still simplifying things and I gather if you're an expert their stuff isn't totally unproblematic. But they've still got a basic commitment to telling true stories — which means for instance that if they don't have any evidence about what was happening inside their protagonist's skull, they generally don't just go and make something up.
posted by nebulawindphone at 9:52 AM on June 13, 2012 [3 favorites]


I'm confused as to why Chotner refers to Aspergers as a 'disease'.

. Of a surfing expert with Asperger’s, Lehrer writes, “Clay’s ability to innovate in surfing is rooted in a defining feature of his mental disorder.” Is Lehrer saying that Clay’s surfing expertise is the result of his disease, or merely that certain properties of the disease may lead to success in fields like surfing?

I may be being pedantic but conflating disease and disorder is the sort of mistake that he no doubt would have lambasted Jonah Lehrer for.
posted by mippy at 10:10 AM on June 13, 2012


(Unless there's a meaning in American English or a way of evaluating Asperger's of which I'm unaware.)
posted by mippy at 10:11 AM on June 13, 2012


I mean, the parable thing is not new. Every undergrad in the world has learned about the Stanford Prison Experiment since ... well, since it happened, I guess. When it's taught well, it's an opportunity to talk about methodology ("Was there a control group? Could there have been? Does it matter?") and experimental ethics ("Should someone have stopped it? Is it bad that the same experiment couldn't be done today?") and to see whether there are any real conclusions you can tease out of that mess of an "experiment." When it's taught badly, it's just a parable about authority, an opportunity to pontificate or shake your head and go "Well, it just goes to show." It's no longer a conclusion that you're trying to extract from the story, but a moral.

Zimbardo has taught the Stanford Prison Experiment exactly that way in every Psych 101 class at Stanford he's taught for the past 40 years, so draw your own conclusions.
posted by blucevalo at 10:24 AM on June 13, 2012


And, as someone who ended their hard science education at 16 yet went on to study the social sciences and wishes she had more scientific knowledge to help her in her daily work, I quite like a bit of the pop science. I've tried reading New Scientist but even a fairly mainstream magazine assumes, in my view, considerable knowledge in the reader. If you can explain something to me using examples I understand, then I feel more comfortable going on to read further on the topic.

Pop sociology does this same job for people who haven't spent enough time with the theory to understand the more scholarly works. I read an article on Bust today which touched on concepts of 'classist privilege' and gendering of fashion - this is on a magazine website, not a journal, but if you've never really encountered feminist theory before you can only guess on what this means and the context isn't really explaining it properly. But if someone writes a piece aimed at the mainstream that picks on something zeitgeisty - People of Walmart, say, or the ostentatious mansions of Russian oligarchs, or a Premier League footballer's wedding - and explains why people might choose to react to it the way they do, then the idea of people's opinions being influenced by a) our perceptions of the working class and their habits and why we might hold these b) our perceptions of 'acceptible' dress and how that might change according to your own class or income level, and why c) how the concept of taste is also forged in, and linked to class backgrounds d) how reactions to fat women might be different from those of photos of fat men e) how lifestyle is as much of a class signifier as income or family background....then it makes a lot more sense than sending the novice off with a copy of Bourdieu and a pair of corduroy pants. Anything that makes subjects seen as mostly academic or 'hard' and makes them more inclusive is a good thing.
posted by mippy at 10:25 AM on June 13, 2012


Zimbardo has taught the Stanford Prison Experiment exactly that way in every Psych 101 class at Stanford he's taught for the past 40 years, so draw your own conclusions.

Wait, in which way? As a serious case study for talking about experiment design and research ethics? Or as a parable meant to illustrate some secular sermon about the nature of evil?

If it's the former, that's great. Lots of people do teach it well, and I'd really expect him to be one of them, since he's clearly spent a lot of time thinking about it.

If it's the latter, then that's pretty lame, and the fact that he's a big shot doesn't make it any less lame.
posted by nebulawindphone at 10:30 AM on June 13, 2012 [2 favorites]


I'd point to Oliver Sacks and V.S. Ramachandran as people who do this sort of pseudoneuroscientific or pop-neuroscientific storytelling well. They're still simplifying things and I gather if you're an expert their stuff isn't totally unproblematic. But they've still got a basic commitment to telling true stories — which means for instance that if they don't have any evidence about what was happening inside their protagonist's skull, they generally don't just go and make something up.

It's worth pointing out that Oliver Sacks and V.S. Ramachandran are practitioners as well as writers. Sacks is an MD, and has been a clinical professor and consulting neurologist. V. S. Ramachandran is a professor of psychology.

Jonah Lehrer has a bachelor's degree in neuroscience, but has no higher degrees and is neither a clinician nor a research scientist.
posted by BrashTech at 10:50 AM on June 13, 2012 [1 favorite]


« Older I'm guessing it took a lot of work to make it look...   |   Prime Martian-Science Real Estate Newer »


This thread has been archived and is closed to new comments