With most examples of the types of judgment mistakes people tend to make, I had to admit that I’m susceptible to the same mistakes. I’d like to think that understanding these tendencies will help me avoid some future mistakes, but I’m not sure it will help as much as I hope. The current “powerful trend to mistrust human intuition and defer to algorithms” is sensible.
Most of the time, the automatic judgments our brains make work very well for us. But when they don’t work well, it’s hard to stop making mistakes. It’s easier to understand this problem when it’s our eyes that make the mistake. Consider the Müller-Lyer optical illusion.
The two horizontal lines are the same length, but the bottom one looks shorter. After we measure and convince ourselves that they’re the same length, our eyes continue to tell us the bottom line is shorter. We can’t turn off the automatic wrong answer our eyes give. So it is with certain types of judgments our brains make. A key insight by Kahnemen and Tversky was that people “weren’t just making random mistakes ... they were doing something systematically wrong.”
An example of a common mistake comes from the “Linda story.”
“Linda is 31 years old, single, outspoken and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.”The question Kahneman and Tversky asked was “which of the following alternatives is more probable?”
“Linda is a bank teller.85% of subjects chose the second option, even though logic dictates that it can’t be more likely than the first option because the second is a subset of the first. Even when they changed the question to make it clear that the first option doesn’t mean that Linda is “NOT active in the feminist movement,” subjects still favoured the second option. Graduate students with training in logic and statistics made this mistake. Even doctors made this type of mistake.
Linda is a bank teller and is active in the feminist movement.”
The book also contains plenty of the humour we’ve come to expect from Michael Lewis. Tversky was once told by an English statistician that “I don’t usually like Jews but I like you.” His reply was “I usually like Englishmen, but I don’t like you.”
Another example came from the way Tversky didn’t worry about the many demands for his time: “The nice thing about things that are urgent is that if you wait long enough they aren’t urgent anymore.”
In an odd definition of “winning,” Tversky’s wife once had to take her eldest son to an emergency room because “He had won a contest with his brother to see who could stick a cucumber farther up his own nose.”
Asked “if their work fit into the new and growing field of artificial intelligence,” Tversky replied “We study natural stupidity instead of artificial intelligence.”
Kahneman and Tversky faced a lot of opposition from economists. One economist’s paper opened with “The agent of economic theory is rational, selfish, and his tastes do not change.” That may make it easier to build theory, but it doesn’t sound much like actual people. But economists clung to this view that people were rational even in the face of mounting evidence from Kahneman and Tversky.
In this battle over whether people are rational, Kahneman and Tversky even reached back to a paper written in 1738 by Daniel Bernoulli (English translation here). Bernoulli offered a model of financial decision-making that we now call utility theory. He believed that people should treat a doubling of one’s net worth as just as good as halving it is bad, even though the dollar amounts won and lost are different.
Tversky wrote that “Bernoulli sought to account a bit better than simple calculations of expected value for how people actually behaved.” I’ve read the translation and I see no evidence this is true. My reading of Bernoulli’s paper was that he sought to model rational choices, not the irrational choices people actually make. I went into more detail on this subject a few years ago.
In explaining Bernoulli’s theory, Lewis writes “The marginal value of the dollars you give up to buy fire insurance on your house is less than the marginal value of the dollars you lose if your house burns down—which is why even though insurance is, strictly speaking, a stupid bet, you buy it.”
Just because fire insurance on a house has a negative expectation, it isn’t a “stupid bet.” I don’t know if this misunderstanding is just Lewis’ or if it is Kahneman and Tversky’s as well. It is irrational to accept a 50/50 bet to either double your net worth plus a dollar or lose everything, even though it has a positive expectation of 50 cents. Bernoulli’s theory isn’t a perfect model of rational decision-making, but it is far better than simple expectation.
All that said, Kahneman and Tversky’s prospect theory is an excellent model of how people actually make decisions, which is clearly different from a purely rational method.
Overall, I found this book up to Lewis’ usual high standard. It gives very accessible insight into Kahneman and Tversky’s work and gives interesting accounts of their lives and collaboration. Anyone who enjoys this book should consider reading Kahneman’s Thinking Fast and Slow.