Thinking, Fast and Slow

by Daniel Kahneman

“The emotional tail wags the rational dog.”

The affect heuristic simplifies our lives by creating a world that is much tidier than reality. Good technologies have few costs in the imaginary world we inhabit, bad technologies have no benefits, and all decisions are easy. In the real world, of course, we often face painful tradeoffs between benefits and costs.

He points out that experts often measure risks by the number of lives (or life-years) lost, while the public draws finer distinctions, for example between “good deaths” and “bad deaths,” or between random accidental fatalities and deaths that occur in the course of voluntary activities such as skiing. These

“Risk” does not exist “out there,” independent of our minds and culture, waiting to be measured. Human beings have invented the concept of “risk” to help them understand and cope with the dangers and uncertainties of life. Although these dangers are real, there is no such thing as “real risk” or “objective risk.”

He goes on to conclude that “defining risk is thus an exercise in power.”

the availability cascade. They comment that in the social context, “all heuristics are equal, but availability is more equal than the others.” They have in mind an expanded notion of the heuristic, in which availability provides a heuristic for judgments other than frequency. In

An availability cascade is a self-sustaining chain of events, which may start from media reports of a relatively minor event and lead up to public panic and large-scale government action.

coming to mind. As Slovic has argued, the amount of concern is not adequately sensitive to the probability of harm; you are imagining the numerator—the tragic story you saw on the news—and not thinking about the denominator. Sunstein has coined the phrase “probability neglect” to describe the pattern.

The combination of probability neglect with the social mechanisms of availability cascades inevitably leads to gross exaggeration of minor threats, sometimes with important consequences.

Terrorism speaks directly to System 1.

“She’s raving about an innovation that has large benefits and no costs. I suspect the affect heuristic.”

“This is an availability cascade: a nonevent that is inflated by the media and the public until it fills our TV screens and becomes all anyone is talking about.”

To decide whether a marble is more likely to be red or green, you need to know how many marbles of each color there are in the urn. The proportion of marbles of a particular kind is called a base rate.

Using base-rate information is the obvious move when no other information is provided.

we expected them to focus exclusively on the similarity of the description to the stereotypes—we called it representativeness—ignoring both the base rates and the doubts about the veracity of the description.

A question about probability or likelihood activates a mental shotgun, evoking answers to easier questions. One of the easy answers is an automatic assessment of representativeness—routine in understanding language.

Although it is common, prediction by representativeness is not statistically optimal. Michael

In all these cases and in many others, there is some truth to the stereotypes that govern judgments of representativeness, and predictions that follow this heuristic may be accurate. In other situations, the stereotypes are false and the representativeness heuristic will mislead, especially if it causes people to neglect base-rate information that points in another direction. Even when the heuristic has some validity, exclusive reliance on it is associated with grave sins against statistical logic.

One sin of representativeness is an excessive willingness to predict the occurrence of unlikely (low base-rate) events.

The students who puffed out their cheeks (an emotionally neutral expression) replicated the original results: they relied exclusively on representativeness and ignored the base rates. As the authors had predicted, however, the frowners did show some sensitivity to the base rates. This is an instructive finding.

When an incorrect intuitive judgment is made, System 1 and System 2 should both be indicted. System 1 suggested the incorrect intuition, and System 2 endorsed it and expressed it in a judgment. However, there are two possible reasons for the failure of System 2—ignorance or laziness.

Some people ignore base rates because they believe them to be irrelevant in the presence ...

The second sin of representativeness is insensitivity to the quality of evidence.

You surely understand in principle that worthless information should not be treated differently from a complete lack of information, but WYSIATI makes it very difficult to apply that principle.

There is one thing you can do when you have doubts about the quality of the evidence: let your judgments of probability stay close to the base rate. Don’t expect this exercise of discipline to be easy—it requires a significant effort of self-monitoring and self-control.

The correct answer to the Tom W puzzle is that you should stay very close to your prior beliefs, slightly reducing the initially high probabilities of well-populated fields (humanities and education; social science and social work) and slightly raising the low probabilities of rare specialties (library science, computer science). You are not exactly where you would be if you had known nothing at all about Tom W, but the little evidence you have is not trustworthy, so the base rates should dominate your estimates.

the logic of how people should change their mind in the light of evidence.

There are two ideas to keep in mind about Bayesian reasoning and how we tend to mess it up. The first is that base rates matter, even in the presence of evidence about the case at hand. This is often not intuitively obvious. The second is that intuitive impressions of the diagnosticity of evidence are often exaggerated. The combination of WYSIATI and associative coherence tends to make us believe in the stories we spin for ourselves. The essential keys to disciplined Bayesian reasoning can be simply summarized:

Anchor your judgment of the probability of an outcome on a plausible base rate. Question the diagnosticity of your evidence.

“They keep making the same mistake: predicting rare events from weak evidence. When the evidence is weak, one should stick with the base rates.”

introduced the idea of a conjunction fallacy, which people commit when they judge a conjunction of two events (here, bank teller and feminist) to be more probable than one of the events (bank teller) in a direct comparison.

The most coherent stories are not necessarily the most probable, but they are plausible, and the notions of coherence, plausibility, and probability are easily confused by the unwary.

The uncritical substitution of plausibility for probability has pernicious effects on judgments when scenarios are used as tools of forecasting.

What have we learned from these studies about the workings of System 2? One conclusion, which is not new, is that System 2 is not impressively alert. The undergraduates and graduate students who participated in our studies of the conjunction fallacy certainly “knew” the logic of Venn diagrams, but they did not apply it reliably even when all the relevant information was laid out in front of them.

The laziness of System 2 is an important fact of life, and the observation that representativeness can block the application of an obvious logical rule is also of some interest.

Intuition governs judgments in the between-subjects condition; logic rules in joint evaluation.

“They constructed a very complicated scenario and insisted on calling it highly probable. It is not—it is only a plausible story.”

However, you can probably guess what people do when faced with this problem: they ignore the base rate and go with the witness. The most common answer is 80%.

Statistical base rates are facts about a population to which a case belongs, but they are not relevant to the individual case. Causal base rates change your view of how the individual case came to be. The two types of base-rate information are treated differently: Statistical base rates are generally underweighted, and sometimes neglected altogether, when specific information about the case at hand is available. Causal base rates are treated as information about the individual case and are easily combined with other case-specific information.

The costs are worth paying to achieve a better society, but denying that the costs exist, while satisfying to the soul and politically correct, is not scientifically defensible.

causal base rates are used; merely statistical facts are (more or less) neglected. The next study, one of my all-time favorites, shows that the situation is rather more complex.

Statistical results with a causal interpretation have a stronger effect on our thinking than noncausal information. But even compelling causal statistics will not change long-held beliefs or beliefs rooted in personal experience.

the feedback to which life exposes us is perverse. Because we tend to be nice to other people when they please us and nasty when they do not, we are statistically punished for being nice and rewarded for being nasty.

It took Francis Galton several years to figure out that correlation and regression are not two concepts—they are different perspectives on the same concept.

The general rule is straightforward but has surprising consequences: whenever the correlation between two scores is imperfect, there will be regression to the mean.

Others involve intuition and System 1, in two main varieties. Some intuitions draw primarily on skill and expertise acquired by repeated experience. The rapid and automatic judgments and choices of chess masters, fireground commanders, and physicians that Gary Klein has described in Sources of Power and elsewhere illustrate these skilled intuitions, in which a solution to the current problem comes to mind quickly because familiar cues are recognized. Other intuitions, which are sometimes subjectively indistinguishable from the first, arise from the operation of heuristics that often substitute an easy question for the harder one that was asked.

Intuitive judgments can be made with high confidence even when they are based on nonregressive assessments of weak evidence.

Intensity matching yields predictions that are as extreme as the evidence on which they are based, leading people to give the same answer to two quite different questions:

The objective of this study was to compare the percentile judgments that the participants made when evaluating the evidence in one case, and when predicting the ultimate outcome in another. The results are easy to summarize: the judgments were identical. Although the two sets of questions differ (one is about the description, the other about the student’s future academic performance), the participants treated them as if they were the same.

People are asked for a prediction but they substitute an evaluation of the evidence, without noticing that the question they answer is not the one they were asked. This process is guaranteed to generate predictions that are systematically biased; they completely ignore regression to the mean.