The Invisible Gorilla: And Other Ways Our Intuitions Deceive Us

by Christopher Chabris

  Taylor relied on his instincts when he decided to bet his company’s future on the new logistics software for mom-and-pop Internet businesses. He’d learned from his reading that he was not using as much of his brainpower as he could be. His left brain was so busy analyzing every option in cost-benefit detail that his more emotional right brain never had a chance to take in the big picture. “I had two warring groups within Chimera on this launch question,” he says later in the day after coming out of a meeting with the project team. One group was gung ho for the new product, but the other had a laundry list of objections. Taylor had to referee and make the final call. “This time I told myself from the outset that I wouldn’t get bogged down in the specifics of the market, the pricing, the project timelines, and so on. Our marketing folks had prepared a profile of the target customer”—a thirty-five-year-old single mother who runs an eBay business out of a spare bedroom in her house—“and I just thought about that woman, and how important her business was to her family and her future, and I visualized her making more money from that business thanks to our software, and I knew that jumping into this market was the right thing to do.”

  The product launch is set for the end of the year. On the drive home, Larry Taylor is a bit more relaxed than he was at the office, but he’s not completely at rest. He is on the phone again—this time talking to his kids.

  In case it wasn’t obvious, the story you just read was entirely made up—100 percent fictitious. Taylor and Flynt don’t exist, and Chimera Information Systems is a chimera. We constructed this fake profile to mimic many similar articles we have seen in the business press.1 It’s full of commonsense notions, assumptions, and beliefs that portray Taylor as a somewhat unconventional, but no doubt successful, business leader. Realizing that the profile must be fake wasn’t the real test, though.

  We intentionally constructed the Larry Taylor story to spotlight the six everyday illusions we have discussed in this book. Did you catch all of them at work? Let’s look back and see where Taylor—and the “writer” of the profile—were led astray by everyday illusions:

  Taylor starts his day by talking nonstop on his cell phone while he drives to work. We saw in Chapter 1 that the illusion of attention insidiously makes us think we can do both of these things at once just as well as we can do either one alone.

  During his “interview,” Taylor gives an extremely precise recollection of how he hired his chief financial officer, emphasizing his own cleverness in announcing a surprise challenge. He may think he remembers the episode “like it was yesterday,” but as we learned in Chapter 2, our memories of even the most salient events are subject to distortion—even as we remain confident that we are recalling them accurately.

  Confidence was an important signal to Taylor when he decided to hire his CFO: Jane Flynt stood out over more experienced, better-educated candidates precisely because of the confidence she exuded. But as we told you in Chapter 3, that sort of confidence is exactly what Jennifer Thompson exuded on the witness stand when Ronald Cotton was sentenced to life in prison for a crime he didn’t commit.

  What makes Taylor such a good manager? According to Taylor himself, it is his broad and deep knowledge of Chimera; others praise his ability to grasp complex information quickly. But as Chapter 4 illustrated, we habitually overestimate our own knowledge (especially of how things work), and we quickly make important decisions that we might profitably stop to reflect on if we realized how little we really do know.

  What’s behind Chimera’s recent success? The experts think it’s Taylor—before he became CEO, the company was an also-ran, but now it’s a leader. From Chapter 5, we can recognize the illusion of cause that can result from a chronological sequence of events: By itself, the fact that Chimera did better after Taylor than it did before him doesn’t prove that his arrival caused the improvement. Other changes to the company around the same time, or changes outside the company, like a general upswing in its industry, might have been responsible.

  The profile also reports that Taylor plays classical music and subliminal messages to his employees and has been trying to access the unused capacity of his own brain. He seems to be under the sway of the illusion of potential that we covered in Chapter 6.

  Earlier we mentioned that everyday illusions have a common characteristic: They all make us think that our mental abilities and capacities are greater than they actually are. There’s another common thread that connects all of the illusions. In each case, we confuse how easily our minds can do something with how well they are doing it. In psychological lingo, we take the fluency with which we process information as a signal that we are processing a lot of information, that we are processing it deeply, and that we are processing it with great accuracy and skill. But effortless processing is not necessarily illusion-free. For example, retrieving memories almost never feels difficult to us. We experience the ease of retrieval, but we don’t experience all the distortions that happened to our memories after they were first stored. These distortions happen beneath the surface of our mental lives, without our awareness. We then mistakenly attribute the perceived fluency of our recall to the accuracy, completeness, and permanence of our memories. Fluency plays a similar role in our understanding of perception, attention, confidence, knowledge, and many other mental processes, and in all of these cases, we have seen that significant illusions result.2

  We aren’t arguing that everyday illusions are inherently bad or that they are simply bugs in the mind’s software that could have been avoided with better programming. Although the illusions result from our mental limitations, those limitations usually have a countervailing benefit. As we pointed out in Chapter 1, the inattentional blindness that causes us to miss the gorilla is an inevitable consequence of our generally salutary ability to focus attention on a primary goal—in that case, counting basketball passes. As in many other situations, the ability to focus is useful precisely because it greatly increases our ability to carry out otherwise difficult tasks.

  In recent years, psychologists have proposed that most of our thought processes can be divided into two types: those that are fast and automatic and those that are slow and reflective. Both contribute to everyday illusions. The rapid, automatic processes involved in perception, memory, and causal inference have serious limitations, but these limitations become much more consequential when our higher-level, reflective, more abstract reasoning abilities fail to see that we are going astray and make appropriate corrections. In other words, we get into more accidents when we talk on a phone while driving both because our attention is limited and because we don’t notice this limitation while it’s happening.3

  It’s not only Larry Taylor and the misguided “author” of his profile who labor under these everyday illusions. We all do. When we uncritically consume stories like the one about Taylor, or when we do the things that Taylor does, we too fall prey to these illusions. Everyday illusions are so woven into our habits of mind that we don’t even realize that they undergird all of the “common sense” that leads us to accept stories like Larry Taylor’s.

  This type of common sense has another name: intuition. What we intuitively accept and believe is derived from what we collectively assume and understand, and intuition influences our decisions automatically and without reflection. Intuition tells us that we pay attention to more than we do, that our memories are more detailed and robust than they are, that confident people are competent people, that we know more than we really do, that coincidences and correlations demonstrate causation, and that our brains have vast reserves of power that are easy to unlock. But in all these cases, our intuitions are wrong, and they can cost us our fortunes, our health, and even our lives if we follow them blindly.

  That’s not a message that has been popular lately. Among the general public and among some psychologists, it has become fashionable to argue that intuitive methods of thinking and making decisions are superior to analytical methods. Intuitive thinking is faster and easier, to be sure. And
the idea that it might also be more accurate is seductive, because it flies in the face of society’s long-standing celebration of rationality and logic as the purest and most objective forms of thought. Toward the end of the profile, we see that Larry Taylor has absorbed this contrarian message. Citing an adage from his days as a poker pro—“think long, think wrong”—and his reading of Malcolm Gladwell’s Blink, he ignores all the analysis his staff has done and goes with his gut, which tells him that customers will benefit from the new product. He bets the company on this instinct, but he is at peace—and back on the phone while he drives home.

  Taylor’s decision might seem like an appalling way to gamble with the money of his investors and the careers of his employees. But sadly, it’s not at all far-fetched to portray a CEO who makes a billion-dollar decision on instinct. Business magazines routinely celebrate this kind of “decisive” leadership. For example, in its profile of Percy Barnevik, the celebrated CEO of the Swedish-Swiss company ABB, the magazine Long Range Planning gushed, “To meet him … is to become immediately aware of an incisive, original approach to management in which the ability to make swift, confident decisions is paramount.”4

  To cite just one concrete example of the instinctual risk-taking that businesspeople engage in all the time, the decision by top executives of Motorola to launch the Iridium satellite telephone business was driven largely by an intuitive “vision” of customers being able to use a single portable phone to place calls from anywhere in the world, despite the extensive data that Motorola itself generated showing that this would be an economically unsound business. The phone would have cost $3,000, the service would have cost $3 per minute, and communication would have been impossible indoors or in cities with skyscrapers. The product was ideal for the desert nomad with a few thousand dollars burning a hole in his pocket, but impractical for everyone else. According to one outside analyst, even if Iridium captured the entire worldwide market for international business calls from developing countries, it still could not pay for the equipment its system required, let alone its operating expenses. Iridium failed within a year of launch and ultimately lost almost $5 billion.5

  When First Impressions Are Wrong Impressions

  Thomas J. Wise was a celebrated British collector of rare books and manuscripts in the late nineteenth and early twentieth centuries. The catalog of his private collection, which he named the Ashley Library, filled eleven printed volumes. Around 1885, an author named W. C. Bennett showed Wise several copies of a privately printed edition of Sonnets from the Portuguese, a famous series of poems written by Elizabeth Barrett Browning during her courtship with Robert Browning. (“How do I love thee? Let me count the ways …”) The sonnets were thought to have been first published in a two-volume collected edition of her poems that appeared in 1850. Bennett’s forty-seven-page pamphlet, labeled “not for publication,” was dated 1847, making it a previously unknown, earlier printing of the sonnets. Wise realized its value as a rarity and purchased a copy for £10. He also alerted several collector friends, who did the same, exhausting Bennett’s stock.

  Wise’s story of how he came upon the Browning volume was corroborated by detailed accounts from one of his friends, Harry Buxton Forman, and from a writer named Edmund Gosse. Over the ensuing years, Wise found and distributed previously unknown volumes of minor works by other writers, including Alfred Tennyson, Charles Dickens, and Robert Louis Stevenson. Numerous private collectors and libraries snapped them up; Wise’s fame and wealth grew commensurately. He eventually became known as the leading book collector and bibliographer in all of England.

  By the turn of the century, however, some American book dealers were becoming discomfited by the steady stream of newly discovered, author-printed pamphlets. In 1898, George D. Smith’s Price Current of Books wrote, “Grave suspicions are entertained that some of these are being manufactured—but that these suspicions are well-grounded, cannot be said … Maybe ‘The Last Tournament’ by Tennyson is worth $300, but it is curious that every Tennyson collector of note has been supplied with one lately!” Despite this and other isolated challenges to their provenance, the pamphlets were broadly respected as genuine for decades.

  In the 1930s, two young British dealers, John Carter and Graham Pollard, formed their own suspicions about the authenticity of some of Wise’s finds. They began a meticulous program of research in which they gathered and analyzed all of the evidence about the provenance of the Browning Sonnets pamphlet. They identified eight separate ways in which the existence of the volume was inconsistent with other facts known about Browning and her work, or with typical experience in rare books. For example, no copies inscribed by the author had ever been found, no copies existed that had been trimmed and bound in the way that was customary at the time of their printing, and the special private printing was not mentioned in any letters, memoirs, or other documents left by the Brownings.

  Carter and Pollard next turned to direct scientific analysis. Although the forensic science of the 1930s was not what it is today, it was possible to examine the paper used to print the Sonnets under a microscope. All paper manufactured in the United Kingdom before 1861 was made from rags, straw, or a strawlike material called esparto. Wood pulp was not used to make paper until 1874. Carter and Pollard put the Browning pamphlet under their microscope and saw a substantial amount of chemically treated wood pulp in its fibers. From this, and much other carefully gathered evidence, they concluded that the putative 1847 printing of the Sonnets had to be a forgery produced after 1874. They performed similar analyses on fifty other pamphlets and found decisive evidence that twenty-one of them were similarly forged.

  The two dealers published the results of their research in 1934, in a 412-page book titled An Enquiry into the Nature of Certain XIXth Century Pamphlets. They stopped short of explicitly accusing Wise of forgery, but their case left no doubt that he was guilty.6 He denied the charges until his death three years later. Subsequent investigations revealed that he had also stolen pages out of many rare books in the British Library. He is still celebrated today, but no longer as a great collector or bibliographer; instead, he is universally regarded as one of the greatest literary forgers of all time.

  How did Wise pull off this fraud on such a massive scale? In evaluating his individual items for their collections, private buyers and institutional librarians didn’t have the opportunity to analyze the full scope of Wise’s offerings, or the skill to perform chemical analysis. Individually, the items looked authentic, and each nicely filled a gap in an author’s known body of work. Intuition was of no help in discovering the fraud. The deception was only uncovered through the use of deductive logic, based on the overall pattern of newly discovered pamphlets, a careful comparison with other historical sources and facts, and scientific study of the items themselves. The story of Thomas Wise and of the detective work performed by John Carter and Graham Pollard illustrates a triumph of deliberation and analysis over instinct. Gut feelings led professional, expert collectors to spend small fortunes accumulating Wise’s pamphlets; rigorous analysis revealed their mistake.7

  Ironically, one of the best-known cases used to demonstrate the power of intuition also involved the detection of a forgery. In his bestselling book Blink, which is subtitled The Power of Thinking Without Thinking, Malcolm Gladwell opens his case for “rapid cognition,” which is another name for intuition, with the story of art experts who could tell immediately that a purported ancient Greek statue known as a kouros was a fake, while scientific experts incorrectly judged it to be authentic.8 Gladwell’s compelling narrative vividly portrays a case in which intuition outdid analysis. And as we’ve seen repeatedly, a single vivid example that illustrates a causal argument may be taken as proof unless we think carefully about the information we haven’t been given—and thinking about what is missing from a story does not come naturally. The case of the kouros might be an anomaly. After all, how often do art experts intuit that a piece is forged when scientific analysis says it’
s genuine? Cases like Wise’s—in which intuitions are refuted by analysis—might well be more common. Moreover, neither story informs us whether intuition or analysis is more accurate when the artwork actually is genuine.

  The story of Thomas J. Wise is just one example of deliberate, scientific analysis overcoming flawed, intuitive judgments; but just as Gladwell’s kouros story does not prove that intuition trumps analysis, our Wise story does not prove that analysis always trumps intuition. Intuition has its uses, but we don’t think it should be exalted above analysis without good evidence that it is truly superior. The key to successful decision-making, we believe, is knowing when to trust your intuition and when to be wary of it and do the hard work of thinking things through.9

  Picking Preserves and Recognizing Robbers

  Are there times when deliberation consistently yields worse judgments than snap decisions and gut intuition? Yes, and here’s an example from a classic experiment. Suppose you were asked to participate in a blind taste-test of five different brands of strawberry jam. After tasting all of the jams, but before being asked to rate their quality, you spend a couple of minutes writing down your reasons for liking and disliking each jam. Then you rate each one on a scale from 1 to 9. How accurate would your ratings be, assuming we judged accuracy by comparing your ratings with those given by a panel of experts assembled by Consumer Reports magazine?