Bullshitting with Einstein

People who are smart, skeptic-minded, and knowledgeable about science at times believe in unfounded, unscientifically-tested things. One such example is brain training, specifically, memory training. Memory training was popularized by the 2011 best-seller Moonwalking with Einstein, by journalist and ‘memory champion’ Joshua Foer. The idea is, by using certain techniques and tricks, such as mnemonics and ‘memory palaces’, it is possible to achieve impressive feats of recall, such as memorizing and recalling the order of a deck of cards quickly and with a high degree of accuracy.

I don’t deny photographic/eidetic memory is possible. I don’t doubt that some people, such as the late super-savant Kim Peek, can do this. However, I am much more skeptical that such a skill can be taught, and if so, that is not just a function of IQ (or some other innate, biological trait).

Why I’m skeptical:

–Using a mnemonic device (such as a ‘memory palace’) still requires one memorize the mnemonic. If I ask you to memorize ten historical dates, a trick may be to associate these dates with a mental visualization, but you must still remember ten associations, which is still not easy.

–Memory training has never been replicated under a controlled environment with multiple subjects, and then the results published in a reputable journal (and then the results successfully replicated by other researchers). The very few studies that exist only have one or two subjects.

–The role of IQ is always ignored, as far as I know. Joshua Foer, being that he is a hugely successful author and attended Yale, likely has a very high IQ (at least 140, which is higher than 99.5% of people). As further evidence of the role of biology, Foer has two brothers who are also very successful authors. Digit recall, which tests for short-term memory, is highly g-loaded meaning it is highly correlated with IQ, so we would expect higher-IQ people to excel at memorization, with or without mnemonics. Without controlling for IQ is is impossible to know if feats of memory are really just feats of IQ, or a specific skill that can be learned (I am very certain it is the former).

Failing to take into account IQ for a task as g-loaded as memory is like measuring vertical jumps without accounting for height and weight, which are important variables for vertical jump distance.

Now of course, it’s possible that everyone regardless of IQ benefits from memory training, but failing to control for IQ is still shoddy science.

–Some people say they can do this, but we still have to control for IQ and other factors in order for it to be considered ‘science’ and not just anecdotal evidence. This means gathering a lot of data by having a large sample size and then controlling for important variables.

–If not IQ, there may be some other innate, biological trait that accounts for superior memory. Maybe memory champions have some anomaly that allows them to link data with visualizations, that if ordinary people were to try they would make little to no progress.

–One study showed that that training is non-transferable. This means if one learns a sequence of numbers, the skill fails when one tries to recite them backwards or a new set of numbers. Furthermore, I have yet to see a study that demonstrates a so-called ‘meta technique’ that could allow anyone to quickly memorize and recall anything, such as a list of complicated legal or medical terminology. So for example, a technique that works for number recall, and then afterwards, legal jargon.

–If the above were possible (if it were actually possible, regardless of IQ, to develop a meta-technique to retain and recall information quickly), the economic consequences would be huge. Productivity would surge. Students would be able to learn material as fast as they can read it, without having to spend hours re-reading, making notes, and highlighting. It would be like that scene in The Matrix where Neo instantly learns how to fly a helicopter. Give how many people attend law school and medical school, and the competitive, high-stakes environment of graduate school, there is a huge market for this, yet no one has been able to create an app or a program that can allow an average person to nearly effortlessly memorize and recall information.

As an addendum, I decided to do some further research as to the IQs of memory champions, and not surprisingly, my hypothesis was confirmed that memory champions aren’t just slightly smarter than average in terms of IQ, but many, many magnitudes so. It’s not even close to average IQ.

In an article from Psychology Today, memory champion Boris Nikolai Konrad is asked about the role of IQ (because that is the most obvious variable):

According to the psychologist K. Anders Ericsson, IQ is irrelevant in acquiring expertise. To him, it’s all about deliberate practice. Have you ever had your IQ tested?

About ten years ago I did the Mensa IQ test and passed it. Barely.

How many of the memory experts at these competitions do you think have a high IQ?

Over the last years we had a large number of memory athletes come to our lab in Munich at the Max Planck Institute of Psychiatry. We found that most of them had a pretty high IQ on standard tests. This does not answer the question whether IQ is a necessity to perform at world-class level in memory sports or if it is just because intelligent people are more likely to enjoy constantly challenging themselves in cognitive tasks. But it was quite apparent that nearly all of them had really high (>130) IQ scores.

Ben Pridmore, whom Foer mentions in his book, is a prominent memory expert whom you know. Does he have a high IQ?

Oh yes. He once won the “World Intelligence Championships”, a competition for difficult IQ-test like tasks and he is really good in many cognitive demanding games and activities.

So that is strait from an expert. An IQ of 130 seems to be the cutoff for acquiring exceptional memory. Having 30 extra IQ points is the ‘secret’ to exceptional memory, but that does not make for a best-selling book or sold-out TED talks.

Regarding Ben Pridmore again, he has an IQ of 160:

He held the official world record for memorizing the order of a randomly shuffled 52-card deck, and has memorised a pack in a time of 24.68 seconds on television. We talk about how he does this and he gives out a technique that anyone can use. I’m currently practicing for fun as part of my monthly challenges. Besides memory sports he is famous for his mental calculation skills and took part in the Mental Calculation World Cup in 2004, 2006 and 2010. Ben says he has an IQ of 159 putting him in the genius range but doesn’t think too much about it and only joined Mensa to talk about it with Friends.

An IQ of 160 is so uncommon that most people in their lifetime will never encounter someone so intelligent, and is considered upper threshold for the reliability of IQ testing. I suspect the same also applies for speed reading, which I hypothesize is also highly correlated with IQ, with smart people reading faster and retaining more [I'll probably do a post on speed reading soon].

I think there is a trend or problem of super-geniuses, such as Cal Newport (a Ph.D. from MIT, which is as smart as it gets) who try to promote this mythos that average people can do that they do, which may be inspiring, but possibly unrealistic and not backed by science. There is also a misconception that skeptics want to rain on people’s parades, but we just want all the data. We want to unvarnished truth. To quote Ronald Reagan, “Trust but verify.”