Near the University of Washington campus, there’s a “love lab” where married couples go for testing by researcher John Gottman. The couples are hooked up to electrodes and sensors and videotaped as they interact about contentious issues in their relationship. Gottman codes every second of video for each spouse’s displayed emotions, such as disgust, contempt, anger, whining, neutrality, sadness, and so forth. The data is plugged into equations, and the marriages’ likely outcomes can be predicted.

According to Gottman, the ratio of positive to negative feelings should be at least five to one or the marriage may be in trouble. Gottman asked couples to tell the story of how they met; their initial interaction predicted their behavior later. Gottman and his colleagues discovered that it takes only three minutes of watching the tapes of interactions to accurately estimate the likelihood that a couple will remain married.

Gladwell discusses a similar phenomenon that occurred during World War II. Throughout the war, the British used thousands of women as interceptors of German Morse Code transmissions. Before long, the women learned to recognize individual operators by their “fist,” or style and cadence of code clicks. They could tell a sender from only a few characters. Combined with other information, the women distinguished which German operators moved to new locations to inform the British about troop movements. At Gottman’s lab, couples also betrayed a signature style of interaction. Sometimes they fell into “negative sentiment override,” in which whatever one spouse said was taken badly by the other. Once this pattern began, it uncontrollably snowballed.

Most people, from marriage counselors to divorcées, are unable to do better than 50/50 at predicting which marriages will succeed. Gottman simplified the process by focusing on “the Four Horsemen: defensiveness, stonewalling, criticism, and contempt” (54); contempt expressed by either or both spouses is the most dangerous signal. Gladwell explains that winnowing down all the complexities to a simple signal—such as whether couples express contempt toward each other—is an example of “thin-slicing,” or finding patterns from limited bits of experience. Thin-slicing often provides better answers than careful, deliberate thinking. Indeed, when a new group watched Gottman’s videos in 30-second cuts, viewed each snippet twice—once for each spouse—and followed simpler instructions about which emotions to note, observers were more than 80% accurate at predicting successful marriages.

In a similar case, psychologist Samuel Gosling worked up personality profiles on 80 students, using the “Big Five Inventory” of extraversion, agreeableness, conscientiousness, emotional stability, and openness to new experiences. Gosling had friends of the students rate them and found the friends made accurate assessments. Gosling then asked strangers to simply visit the students’ dorm rooms for 15 minutes and then rate them. The strangers were less accurate on extraversion and agreeableness, but much more accurate on the latter three traits.

In a similar vein, Gladwell notes that people often misunderstand themselves, which is why asking them what type of person they are can be misleading. It’s often better to ask indirect questions whose answers can reveal personality. Likewise, to determine which doctors are most likely to be sued for malpractice, a careful analysis of medical records and diagnostic decisions will yield less information than simply listening in as doctors talk to their patients. In malpractice cases, patients often say they feel hurried, ignored, or mistreated. One study shows that tone of voice alone—dominant versus concerned—predicts which doctors may see a lawsuit.

Tennis coach Vic Braden had the knack of predicting when a tennis player was about to default on a serve, predicting as much as 16 out of 17 times in a row. The problem is that “he simply cannot figure out how he knows” (84). Moreover, Braden notes that professional tennis players often have inaccurate understandings of how they play. Baseball player Ted Williams, one of history’s greats, described how he could “look the ball onto the bat” (116), but this is physiologically impossible during the last few milliseconds before contact. Williams professed, “Well, I guess it just seemed like I could do that” (117). 

Gladwell argues that this is not unusual: People are often unable to explain why they suddenly know something is going to happen. Further examples include art historian Bernard Berenson, who gets a queasy stomach or feels “off balance” in the presence of a fake, and investor George Soros, who has back spasms that warn him it’s time to make a trade.

Gladwell introduces the idea of priming via an experiment designed by researcher John Bargh, who gave subjects word tests sprinkled with words like “aggressively,” “bold,” and “rude,” or, for other participants, “respect,” “considerate,” “patiently,” and so on. The design of the tests did not allow subjects to know of the deliberately added words. After the test, each test taker had to go down the hall to an office for further instructions; however, at the office, someone was arguing with the professor, blocking the door. Subjects whose tests included the aggressive words tended to interrupt, while those who took tests sprinkled with diplomatic words simply waited.

In another example of priming, Black students took a 20-question sample of the Graduate Record Examination. Some had to indicate their race first; these subjects correctly answered only half as many questions. Apparently, “that simple act was sufficient to prime them with all the negative stereotypes associated with African Americans and academic achievement” (96).

Gladwell disputes the idea that priming means people are simply pawns of their environment. More likely, he says, unconscious processes keep people’s actions as appropriate as possible while they concentrate on the real problems at hand. The benefits of this kind of automation are evident when one considers the experiences of people who have sustained damage to the ventromedial prefrontal cortex, a part of the brain just behind the nose. These people are unable to have intuitions. They may be very intelligent but show poor judgment and must reason slowly through every situation, often without ever reaching a decision. Moreover, those who attempt to override their intuition may end up in trouble. Speed dating works because people can decide in moments whether they’re interested in someone. By contrast, people have trouble describing the type of person they’re attracted to: They say they want one thing but choose another. Their conscious ideals don’t match their intuitive choices.