The Undoing Project

Michael Lewis, “The Undoing Project: A Friendship That Changed Our Minds,”  W.W.Norton, 2017, 352 pgs.

Forty years ago, Israeli psychologists Daniel Kahneman and Amos Tversky wrote a series of original studies undoing conventional assumptions about the decision-making process. Their papers showed the ways in which the human mind erred, systematically, when forced to make judgments in uncertain situations. Their work eventually led to the creation of Behavioral Economics, revolutionized Big Data studies, advanced evidence-based medicine, and led to a new approach to government regulation. Kahneman and Tversky are more responsible than anybody for the powerful trend to mistrust human intuition and defer to algorithms.

The Undoing Project unfolds and tells the story of a remarkable collaboration between two men. They became heroes in the university and on the battlefield—both had important careers in the Israeli military—and their research was deeply linked to their life experiences. Amos Tversky was a brilliant, self-confident warrior and extrovert, the center of rapt attention in any room; Kahneman, a fugitive from the Nazis in his childhood, was an introvert whose questing self-doubt was the seedbed of his ideas. They became one of the greatest partnerships in the history of science, working together so closely that they couldn’t remember whose brain originated which ideas, or who should claim credit. They flipped a coin to decide the lead authorship on the first paper they wrote, and simply alternated thereafter.

When Tversky died young, in 1996, he was on the secret shortlist for a Nobel memorial prize in economics. Kahneman won the Nobel economics prize in 2002 and published his own bestselling book, Thinking, Fast and Slow, in 2011 (another book I highly recommend).

An example of their work is the “representativeness heuristic” — our tendency to make judgments by comparing an example to some mental model. When we meet a nervous, geeky person we note that he matches our stereotype of a programmer and, therefore, probably is a programmer. We forget that most of the people we meet are not, in fact, programmers, no matter how much they might resemble them, or at least our idea of them.  So when judging probabilities we often skip over the real question, “Is this likely?” in favor of a representativeness question: “Does this match my preconceptions?” “Is the lump likely to be a malignant tumor?” becomes “Does the lump match my idea of what a malignant tumor looks like?” It’s a reasonable rule of thumb that can lead us seriously astray.

Behavioral economics itself is not a major part of this book. Richard Thaler, the most important intellectual conduit between Kahneman and Tversky and economics, does not appear in the story until the closing chapters.  Anyone interested in Behavioral Economics should read Thaler’s books that I have reported on earlier:

Richard H. Thaler and Cass R. Sunstein, “Nudge (Improving Decisions About Health, Wealth, and Happiness)”, Penguin Books, C2009,  Pages 271

Richard H. Thaler, “Misbehaving: The Making of Behavioral Economics” W.W.Norton, 2015, 358 pp.

Addenda:

With Amos Tversky and others, Kahneman established a cognitive basis for common human errors that arise from heuristics and biases.  Together they developed prospect theory.

*Michael Lewis is an excellent writer and I recommend any and all books he has written.  For instance, his best work is exemplified in Liar’s Poker which described bond trader Lewis Ranieri and the way securitization revolutionized Wall Street in the 1980s. Moneyball covered baseball manager Billy Beane and anticipated the “quants” taking over the world. And The Big Short depicted Steve Eisman and Michael Burry, the men who spotted the financial crisis coming and bet vast sums on it.  I have read all of these book and highly recommend each one.

My Notes:

Pg. 79:  As far back in 1915, a psychologist named Thorndike had discovered what came to be known as the halo effect.  He had found that when people were asked to rate others according to some physical trait (‘physique’) and then assess some less tangible quality (‘intelligence, leadership’, and so forth) the feeling created by making the first ranking bled into the second: So if a person was first judged physically impressive, they would likely be found impressive in other ways.  Switch the order of assessment, and the same problem occurred: If a person was first judged to be generally great, he was then judged to be stronger than he actually was.  Thorndike concluded that he was convinced that even a very capable foreman, employer, teacher, or department head is unable to view an individual as a compound of separate qualities and to assign a magnitude to each of these in independence of the others.  Thus was born what is still called ‘the halo effect.’

Pg. 102:  Amos Tversky compared clinical psychology to medicine.  If you went to a doctor in the seventeenth century, you were worse off for having gone.  By the end of the nineteenth century, going to the doctor was a break-even proposition: You were as likely to come away from the visit better off as you were to be worse off.  Amos argued that clinical psychology was like medicine in the seventeenth century, and he had lost of evidence to support his case.

Pg. 160:  The Law of Small Numbers.  Psychologist often used sample sizes of around 40 which gave them only a 50 percent chance of accurately reflecting the population.  To have a 90 percent chance of capturing the traits of the larger population, the sample size needed to be at least 130.  Given the inherent variability of the small sample, the sample may be unrepresentative.  This belief in small numbers was a widespread intellectual error.  At the time, psychologists had so much faith in small samples that they assumed that whatever had been learned for a group must be generally true, even if one group seemed to contradict the other group. 

Pg. 193: Anchoring.  People could be anchored with information that was totally irrelevant to the problem they were being asked to solve.  For instance, Danny and Amos asked their subjects to spin a wheel of fortune with slots on it that were numbered 0 through 100.  Then they asked the subjects to estimate the percentage of African countries in the UN.  The people who spun a higher number on the wheel tended to guess that a higher percentage of the United Nations consisted of African countries than did those for whom the needle landed on a lower number. 

Pg. 198:  Danny and Amos wrote that,”In making predictions and judgments under uncertainty, people do not appear to follow the calculus of chance or the statistical theory of prediction.  Instead, they rely on a limited number of heuristics which sometimes yield reasonable judgments and sometimes lead to severe and systematic error.”

Pg. 216: The Representativeness Heuristic.   It appears that doctors do not think statistically.  Eighty percent of doctors don’t think probabilities apply to their patients, just like 95 percent of married couple don’t believe the 50 percent divorce rate applies to them.  Doctors and others must be careful when there is one simple diagnosis that instantly pops into your mind that explains everything all at once, even if the probability metric is low.  It isn’t that the first thought that comes to mind is always wrong, it is that its existence in your mind will lead you to feel more certain than you should be that it is correct.  You have to consider base rates unless you are completely certain you are correct.

Example: Miss De Jong is orderly, neat, fairly quiet, and shy. She enjoys reading in her spare time and belongs to a social club that includes three librarians, nine real estate agents, and eight social workers. A tendency to conclude that Jan must be one of the three librarians would illustrate the powerful influence of the representativeness heuristic.  The danger of using the representativeness heuristic is that it may lead us to disregard probability information that it is improbable that any one person is a librarian which in this case is relevant to our judgments.

Pg. 283:  The Endowment Effect (Thaler): People attach some strange extra value to whatever they happen to own, simply because they own it, and prove surprisingly reluctant to part with their possessions, or endowments, even when trading them makes economic sense. For instance: people are reluctant to sell stocks that have fallen in value even when admitting they would never buy them at their current price.

Pg. 219:  It is not surprising that people make mistakes.  What is so compelling is that the mistakes are predictable and systematic.  They seem ingrained in human nature. 

Pg. 304:  The Undoing Project:  With the passage of time, the consequences of any event accumulate, and leave more to undo.  And the more there is to undo, the less likely the mind is to even try.  This was perhaps one way time heals wounds, by making them less avoidable. 

Pg. 325: Representativeness (the similarity between whatever people are judging and some model they have in their mind of that thing):   People were blind to logic when it was embedded in a story.  For instance: when Amos asked people, Which is more likely to happen in the next year, that a thousand Americans will die in a flood, or that an earthquake in California will trigger a massive flood that will drown a thousand Americans? People went with the earthquake.