Why MIT Students Can't Write and Harvard Students Can't Count

An MIT PhD engineer dad was recounting an old saw about how MIT students can't write and Harvard students can't count and it made me chuckle because I am a Harvard grad who counts on her fingers.

Like the old MIT-Harvard rivalry, there's often a cortical battle for resources between spatial and verbal / visual "picture" thinking. In studies of spatial experts, high levels of spatial expertise were correlated with lower levels of verbal fluency, auditory verbal memory, and visual memory (for more, read here. But these studies, if you look at mathematicians and physicists talking about their thought processes (see Hadamard's Psychology of invention. From the mathematician Hadamard: "I insist that words are totally absent from mind when I really think...even after reading or hearing a question, every word disappears at the very moment I am beginning to think it over; words do not reappear in my consciousness before I have accomplished or given up the research...I fully agree with Schopenhauer when he writes, "Thoughts die the moment they are embodied into words." Well no wonder MIT students can't write. In fact, they may take solace in the words of polymath Francis Galton: "It is a serious drawback to me in writing...that I do not so easily think in words as otherwise. It often happens that after being hard at work, and having arrived at results that are perfectly clear and satisfactory to myself, when I try to express them in language I feel that I must begin by putting myself upon quite another intellectual plane. I have to translate my thoughts into a language that does not run very evenly with them. I therefore waste a vast deal of time in seeking for appropriate words and phrases, and am conscious, when required to speak on a sudden, of being often very obscure..." If you look at the SAT subtest scores of MIT and Harvard students (25th percentile listed here - because 75th percentile was clustered at 800), MIT students are indeed weakest at reading and writing (not surprising you find many dyslexic engineers, mathematicians, physicists at MIT).

If you're an MIT student, Harvard students really may seem bad at math. The more uniform differences between reading, math, and science suggest more verbal (fewer spatial) left hemispheric types at Harvard.

There aren't any studies yet comparing the math abilities of highly verbal thinkers but if you superimpose test subjects doing verbal reasoning tasks (green) with estimations of nummber (red) - the areas really are distinct. And there are certainly plenty of famous highly verbal thinkers (for instance the polyglot Max Muller) who have gone on record saying that they didn't think that thinking could exist separate from words - definitely from the descriptions of spatial mathematicians and scientists above. We don't know how Muller did with Math, but certainly the multi-talented author, linguist, and etymologist C.S. Lewis was notoriously bad at math and simple calculations. He failed the mathematics part of college entrance exams twice, and was only allowed into college without passing math because he had served in WWI.

Spatial Expertise Gray Matter pdf
Causal / Verbal Reasoning pdf
fMRI of Dyscalculia pdf
C.S. Lewis and Math


  1. Anonymous3:51 AM

    Fun post. I'm reminded of another old saw that Life Magazine is for those who can't see and Time is for those who can't think.

  2. Anonymous3:45 PM

    This is interesting, but what do you make of Ann Roe's study (1952) of eminent scientists? It seems that verbal ability is more important than spatial ability (noted by the much higher verbal IQ's than spatial IQ's).

    I found a chart here.

  3. Thanks for the link. The details may be revealing. Certainly certain types of science are more deductive than inductive, same in mathematics. Physics is a classic 'spatial' discipline. In Sparks of Genius, the Root-Bernsteins also spoke of the different classes of scientific researchers - the deductive scientists were more likely to extend a field, whereas the inductive ones were more likely to create a new one. I would wonder about how they defined their 'eminent' group.

  4. Very interesting post. As a graphic designer, I never think in words when doing my work despite interacting with a computer that has a menu system mostly driven by language. I cannot even explain where I go in my mind when I am working. Often if I am interrupted by somebody else, it takes me 20-30 seconds to even comprehend what they're saying because I'm so far removed from verbal thinking. Their words are like a foreign language until I tune back in. Thanks for the interesting train of thought!

  5. Anonymous8:02 AM

    I found the following links about Roe's study of eminent scientists: one and two.

    In short, there were 64 of the top scientists in the early 50's, broadly split into three groups of biologists, physical scientists, and social scientists. They were given IQ tests—testing verbal, mathematical, and spatial—developed specifically for them by ETS. Theoretical physicists had much higher verbal scores that experimental physicists. Biologists and social scientists had intermediate verbal scores. The physical scientists had the highest spatial ability while biologists and social scientists were rather low.

    This has confirmed my opinion that to do well in mathematical logic (e.g., completeness theorems in the predicate calculus), pure mathematics (e.g., real analysis), and theoretical physics, one needs high verbal ability with good spatial ability. Those with weaker verbal ability seem to go towards engineering, applied mathematics, and experimental physics. Of course, I'm merely talking about group properties, so some individuals would certainly not fit this picture.

  6. This is very interesting. Thank you for the added detail. To some extent, the results should not be surprising - successful theoretical physicists should have strong enough verbal abilities to communicate their ideas - but experimental physicists and engineers make or do - and do not have to talk about it (or like the Orville Wright, have a brother who can do your talking for you).

    The issue of logic - we'd have to think about. We have seen some very strong logicians who seem relatively weak verbally - they explain that mathematics is a different language - and we have not probed to see more specifically how they do what they do. There are certainly very strong computer programming types who also tell us mathematical and computer languages are fine, but English - forget about it!

  7. Anonymous8:56 AM

    This is fascinating. And as a Harvard type who thinks in words dating an MIT type who probably doesn't, I can relate.

    Are you familiar with the recent Northwestern study relating poor reading to difficulties picking out certain speech sounds in background noise? It looked like it might have interesting implications for education. Links are at: http://www.northwestern.edu/newscenter/stories/2009/07/ReadingNoise.htm (for the digest on the Northwestern website) and http://www.pnas.org/papbyrecent.shtml (for the complete study).

    Also, just wanted to let you know, I have been reading this blog for at least 5 years. I started reading it while I was in high school. It introduced me to neuroimaging, and I am now majoring in cognitive science so I can do neuroimaging research for the rest of my life. Thank you both!

  8. Hi tdq, Thanks for the link. Not familiar with the study, but it certainly makes sense.

    Cool you're majoring in cognitive science. It's a fantastic field. Look forward to see your brain blog in the future!

    F & B

  9. It's really funny. The psychologists who've tested me say that my right frontal lobe is less well-developed and less efficent and yet when I write stories, not only are they praised by my creative writing teachers as being vivid and creative, but I notice that sometimes when I write, I'm able to picture things far more vividly than I am in an everyday situation.

    As far as everyday life goes, I think in words mainly, but when I do something like design an idea for a magazine advertisement, then my attention becomes diffuse and I have this big movie screen in my mind playing out the ad in motion. I wish I could access this inner silver screen every day at any time in my life but I can't. I can only access it in very specific situations.

    I sort of hate it when people refer to Harvard English Students as "left brained" because it belittles the crucialness of having an intact right hemisphere to understand the complex literature they have to read.

    On the other hand, I wonder if, due to the ridiculous over-emphasis on right-left brain, we are underestimating the importance of the left brain in creative activities.

    I know I'm going all over the place here, but basically it breaks down like this.

    Visual thinking is thought of as the creative superior to verbal thinking just because there's just so many more ways that images can be manipulated than words.

    However, I'd like to believe that it takes as much creative energy to extend a field as it does to start an entirely new one.

  10. New is new - whether you're extending a field or creating a new one, that is certainly true.

    The MIT-Harvard differences are over-simplified that is true, but some of the traits are interesting. And they are fun / funny if you are an MIT or Harvard alum and know the rivalry. The 75%tile scores for MIT and Harvard students are essentially the same because they are at the top (800) - so there are plenty of strong or strong-enough whole-brain learners out there.

    I would still think there are more extreme asynchronies on the MIT - perhaps because of the "MIT disease" (aka dyslexia) that Negroponte refers to - as well as other twice exceptional brain patterns.