Seeing What They Want to See

January 6, 2009

Sometimes it takes a fresh eye (or a bleary-eyed grad student) to catch what others have missed. Jonah Lehrer, Mind Hacks, and Neurocritic all blog about a paper by Edward Vul and others revealing some troubling exaggerations of correlations in social neuroscience imaging. Hopefully this paper will get some traction and cause the methodology in question to be reexamined–that’s how science progresses, after all.

The paper is available online (PDF) and the abstract is here:

The newly emerging field of Social Neuroscience has drawn much attention in recent years, with high-profile studies frequently reporting extremely high (e.g., >.8)
correlations between behavioral and self-report measures of personality or emotion and
measures of brain activation obtained using fMRI. We show that these correlations often
exceed what is statistically possible assuming the (evidently rather limited) reliability of
both fMRI and personality/emotion measures. The implausibly high correlations are all
the more puzzling because social-neuroscience method sections rarely contain sufficient
detail to ascertain how these correlations were obtained. We surveyed authors of 54
articles that reported findings of this kind to determine the details of their analyses. More
than half acknowledged using a strategy that computes separate correlations for
individual voxels, and reports means of just the subset of voxels exceeding chosen
thresholds. We show how this non-independent analysis grossly inflates correlations,
while yielding reassuring-looking scattergrams. This analysis technique was used to
obtain the vast majority of the implausibly high correlations in our survey sample. In
addition, we argue that other analysis problems likely created entirely spurious correlations in some cases. We outline how the data from these studies could be
reanalyzed with unbiased methods to provide the field with accurate estimates of the
correlations in question. We urge authors to perform such reanalyses and to correct the
scientific record.


DC Hurts My Brain

January 6, 2009

Jonah Lehrer (of the Frontal Cortex, author of Proust Was a Neuroscientist) writes How the City Hurts Your Brain in the Boston Globe:

Related research has demonstrated that increased “cognitive load” — like the mental demands of being in a city — makes people more likely to choose chocolate cake instead of fruit salad, or indulge in a unhealthy snack. This is the one-two punch of city life: It subverts our ability to resist temptation even as it surrounds us with it, from fast-food outlets to fancy clothing stores. The end result is too many calories and too much credit card debt.

Don’t worry, there are benefits too.


Jonah Lehrer, Kanye West, and John Mayer

March 20, 2008

lehrerkanyemayer

Jonah Lehrer of The Frontal Cortex spoke at the Clinton School last night, and it was great. I just reviewed Lehrer’s first book Proust was a Neuroscientist and was excited to hear him speak on the purposefully outrageous title “Kanye West Was a Neuroscientist.” I also had the privilege of introducing him. I had written out the introduction, but then cut it down quite a bit, so here’s the longer version I didn’t actually get to give:

I am a student of science. Science appeals to my practical side; it took us to the Moon and back, it let us map the human genome, and science helped us eradicate smallpox, all within my parents’ life spans. Science appeals to me because the reductionist technique is an incredible tool for fighting disease and gaining knowledge about the world we inhabit.

I started my day this morning with a lab for Advanced Genetics, where we’re studying Arabidopsis thaliana, a plant species so important that only genetics students have ever heard of it. This semester we undergrads have been using the elegant techniques of science to create new, if extremely specialized, knowledge about the world. The techniques of science and the philosophy of reductionism are so powerful that it’s tempting to apply them to every field of knowledge. But we need balance.

After my genetics lab I go to the daily Chorus rehearsal. Walking into the music building for Chorus is like crossing an invisible but impenetrable barrier between entire realms of knowledge and ways of knowing. Few of my friends in the science building are active in the arts, so I have a hard time explaining the irreducible experience of singing Brahms’ German Requiem with a group of talented musicians. And hardly anyone in Chorus would stay awake if I started talking about the routine glory I see in the polymerase chain reaction.

Jonah Lehrer’s first book, Proust Was a Neuroscientist is an exuberant demonstration that these barriers between science and art can and should be broken down. By taking test cases of artists, authors, musicians, and cooks who anticipated the discoveries of today’s neuroscience, Lehrer makes a strong case for the importance of the arts. And when Lehrer writes about the limitations of science, it is all the more convincing because he also writes knowledgeably about its groundbreaking successes.

Lehrer studied neuroscience at Columbia University, where he worked in the lab of Nobel Prize-winning neuroscientist Eric Kandel. Lehrer was a writer for the Columbia Spectator, a Gospel music DJ for Columbia’s radio station, and Editor-in-Chief of The Columbia Review, the university poetry magazine. He also worked as a line cook at Le Cirque 2000 and Le Bernardin.

Lehrer then studied at Oxford University as a Rhodes Scholar, where he did much of the research that would become his first book. Today he is Editor at Large for Seed, a magazine about science and culture. He has written for the Boston Globe, Nature, and NPR, and writes a blog about neuroscience called the Frontal Cortex. Join me in welcoming Jonah Lehrer.

Lehrer’s talk was largely based on content from his book. He talked about Escoffier and Stravinsky, and how our love of music is rooted in our ability to recognize patterns. Great artists make good music by setting up a pattern and intentionally withholding it from us. Because we want the pattern again and don’t quite get what we want, our ears and our brain stay fascinated. This, Lehrer argued, is a key similarity between Beethoven and Kanye West. It’s almost like Stravinsky and Kanye West are evolutionarily conserved hox developmental genes in fruit flies and mammals (I’ll let you pick which is which) that have been recently discovered to be homologous. From the outside they look completely different, but what makes them tick is basically the same.

Lehrer’s speech also reminded me of the new John Mayer song I heard for the first time last week. “Say” is pretty simple–it’s the high guitar (or ukulele?) hook at the beginning that first caught my ear. But listen to the chorus too; it’s one six-word phrase repeated over and over:

This simple pattern is very catchy–it sticks in your head. I heard it once on the radio, and then once on the PA system before Bill Clinton spoke (random mix of music I’m sure) and then I had it stuck in my head the whole drive back from New Orleans. I literally flipped through stations hoping to hear the song. I thought about stopping to buy it, but never passed any music stores. At that point I didn’t know who it was or the name of the song, but I was hooked.

So I got home, found out what the song was, and bought in on iTunes. I must say, I’m already kind of tired of it. The Chorus is just too repetitive. While Mayer has some excellent other music, his singles tend to be of lower quality. The difference between a good song and an ok song, or between mere pop music and real music seems to be in these patterns. “Say” is simply too predictable. You know exactly what Mayer will say before he says it, because it’s the same thing over and over again. Really great music flirts with that repetition but never quite gives you satisfaction. And that’s why Brahms takes more effort to listen to, but will likely stand the test of time longer than John Mayer. Or at least his radio singles.


Moth Memory

March 11, 2008

Retention of Memory through Metamorphosis: Can a Moth Remember What It Learned As a Caterpillar?

Yes. Awesome.

moth

Via Frontal Cortex.


Proust Really Was a Neuroscientist

March 9, 2008

I just finished Jonah Lehrer’s first book, Proust Was a Neuroscientist. Lehrer is an editor for Seed magazine, blogs at the Frontal Cortex, and did a lot of the research for his book while on a Rhodes Scholarship at Oxford. Proust Was a Neuroscientist was just published last November, and one of the most impressive things about it is that Lehrer is only 25.

Proust was a neuroscientist Sandwiched between an introduction called “Prelude” and a postcript titled “Coda,” Proust presents eight chapters on 19th century figures who anticipated a discovery of modern neuroscience in their art. Authors Walt Whitman, George Eliot, Marcel Proust, Gertrude Stein, and Virginia Woolf are profiled along with chef Auguste Escoffier, painter Paul Cezanne, and composer Igor Stravinsky.

Proust is an intellectual tour de force; you get excellent descriptions of the techniques and discoveries of neuroscience alongside biography of fascinating cultural figures and accessible criticism of their works. Given my background studying biochemistry, the science wasn’t that in-depth, but the literary history is something I really needed. My dad is an art professor, so I’m generally more familiar with the great art of history than great literature.

And that’s exactly the point. My friend Jimmy recommend CP Snow’s The Two Cultures a couple years ago. Snow’s classic little text is a call for more communication between the humanities and the sciences. Lehrer summarizes the philosophy behind his book in its Coda by beginning with Snow’s argument for a Third Culture of artists and scientists enjoying knowledge from both spheres. Lehrer’s critique:

Snow turned out to be prophetic, at least in part. The third culture is a now a genuine cultural movement. However, while this new third culture borrows Snow’s phrase, it strays from his project… The third culture today refers to scientists who communicate directly with the general public. They are translating their truths for the masses… From Richard Dawkins to Brian Greene, from Steven Pinker to E.O. Wilson, these scientists do important scientific research and write in elegant prose. Because of their work, black holes, memes, and selfish genes are now part of our cultural lexicon.

He then summarizes the theme of Wilson’s Consilience: The Unity of Knowledge–a book I enjoyed immensely but felt ultimately fell short–with this quote from Consilience:

The central idea of the consilience world view is that all tangible phenomena, from the birth of stars to the workings of social institutions, are based on material processes that are ultimately reducible, however long and tortuous the sequences, to the laws of physics.

This is a statement with which I can’t argue, but also find dissatisfying. Jonah’s response?

Wilson’s ideology is technically true but, in the end, rather meaningless…When some things are broken apart, they are just broken. What the artists in this book reveal is that there are many different ways of describing reality, each of which is capable of generating truth.

And on a derogatory reference to Virginia Woolf by Steven Pinker:

But if Pinker is wrong to thoguhtless attack Virginia Woolf (seeing an enemy when he should see an ally), he is right to admonish what he calls “the priests of postmodernism.” Too often, postmodernism…indulges in cheap disavowals of science and the scientific method. There is no truth, postmodernists say, only differing descriptions, all of which are equally invalid. Obviously, this idea very quickly exhausts itself. No truth is perfect, but that doesn’t mean all truths are equally imperfect. We will always need some way to distinguish among our claims.

The reception for Lehrer’s first offering hasn’t been completely positive (see a rather critical review at Slate) but it seems that most of the complaints I’ve read center around small quibbles with Lehrer’s points or a misunderstanding of the overall thesis. To be sure, I was skeptical going in; it seems that every critique of the epistemology of the likes of Wilson I’ve read has been cover to a religious or postmodern agenda (or both).

Proust Was a Neuroscientist was so refreshing because it was written by someone with an obvious knowledge of and respect for science on a level commensurate with its accomplishments. Because Lehrer can write compelling prose about neuroscience while admitting that our mind really is just the machine and not the ghost, he can also write convincingly about the value of art in giving us knowledge about our human experience.