Humans are imperfect and inconsistent decision makers
Noise. By Daniel Kahneman, Olivier Sibony and Cass Sunstein. Small, brown; 464 pages; $ 32. William Collins; £ 25
NOTOISE IS UNWANTED variation in judgments that should be the same, leading to inaccurate and unfair decisions. He’s around people all the time, although people don’t notice him. To get a feel for how this happens, do a “noise audit” right now: Open your phone’s stopwatch app and practice counting ten seconds. Now, with your eyes closed, count several times, pressing the turn button whenever you think ten seconds have passed.
Your answers were not perfect but loud: slightly above or below the ten-second mark. And if they have consistently gone wrong in one direction, there is also a bias, which is another form of error (you counted too quickly or too slowly).
The problem of decision bias is well known and there are strategies people can adopt to minimize it. For example, customers may be “anchored” to the first price presented to them in a transaction, so they learn to consciously reject it before negotiating. But the noise is different precisely because it is less noticeable. “It only becomes visible when we statistically think of a set of similar judgments. Indeed, it then becomes difficult to miss, ”write Daniel Kahneman, Olivier Sibony and Cass Sunstein in their new book.
The differences are glaring. In a Miami courthouse, one judge would grant asylum to refugees 88% of the time while another would do so 5% of the time. A large study of radiologists found that the false positive rate ranged from 1% to 64%, meaning that two-thirds of the time a radiologist said a mammogram showed cancer when it was not. not. Doctors are more likely to prescribe opioids at the end of a long day. The judges made tougher decisions before their breaks and on the hottest days. Underwriters at an insurance company were estimating premiums that ranged from 55%, a difference five times greater than its management had imagined.
Not only do individuals differ from their peers, they often fail to agree with themselves. Oenologists tasting the same samples a second time obtained less than one in five identically. Four in five fingerprint reviewers changed their initial identification decision when presented with contextual information that should not have been a factor in the fingerprint match. In a medical study evaluating angiograms, doctors disagreed with their previous judgments more than half the time.
The noise is sometimes good. When different investors rate a business or book reviews reach different ratings, diversity of opinion is beneficial. But more often than not, it creates problems. In law, noise is synonymous with injustice. In business, it can get expensive.
Still, it can be reduced. The authors’ remedies include a “noise audit” to measure the degree of disagreement over the same cases, to quantify the variation that is usually invisible. They also call for better “decision hygiene”, such as the designation of an observer for group decisions, in order to avoid common prejudices and loud judgments. For example, they can ensure that team participants get independent assessments before they meet as a group to consolidate their decisions.
Another solution is to do without people altogether. Statistical models, predetermined rules and algorithms are in many cases more precise than human judgment. The authors welcome artificial intelligence to make many decisions in society, but recognize that people are predisposed to resist their responses for lack of personal and emotional quality in decision-making, even if it leads to inferior decisions. , or at least variable.
The trio speak with credibility. Mr. Kahneman is a Nobel Laureate whose ideas about bias in human reasoning have reshaped the economy and society; Mr. Sunstein is a math scholar at Harvard and an occasional government official putting his ideas into politics; Mr. Sibony is a former McKinsey partner who teaches decision science at a French business school. Yet despite the book’s title, the authors struggled to extract signal from noise, so to speak, requiring some 400 pages to make their point. A tighter argument would have improved the ideas they present. ■
This article appeared in the Books & Arts section of the print edition under the title “Sound pollution”