The Invisible Gorilla: And Other Ways Our Intuitions Deceive Us

by Christopher Chabris

  15. The first quote is from Robert Burns, the second is from Helmuth Graf von Moltke, and the third is from Douglas Hofstadter.

  16. This quip is usually attributed to Yogi Berra, whose sayings often had this sort of twisted logic, but a version of it was apparently said earlier by the physicist Neils Bohr.

  17. This study is described on p. 142 of P. B. Carroll and C. Mui, Billion Dollar Lessons: What You Can Learn from the Most Inexcusable Business Failures of the Last 25 Years (New York: Portfolio, 2008).

  18. The classic volume on the positive nature of most self-deception is S. E. Taylor, Positive Illusions: Creative Self-Deception and the Healthy Mind (New York: Basic Books, 1989). The idea that depressed people are less subject to everyday illusions is speculative; there is a controversial line of research suggesting that depressed people have a more realistic understanding of how much they can control events (e.g., L. B. Alloy and L. Y. Abramson, “Judgment of Contingency in Depressed and Nondepressed Students: Sadder but Wiser?” Journal of Experimental Psychology: General 108 [1979]: 441–485).

  19. The idea of the “outside view” is described in detail in D. Lovallo and D. Kahneman, “Delusions of Success: How Optimism Undermines Executive Decisions,” Harvard Business Review (July 2003): 56–63. The tendency to underestimate the time to complete a task is often called the “planning fallacy,” and the formal name for the technique of comparing a project to similar ones to estimate completion time is called “reference class forecasting.” This method has been endorsed by the American Planning Association. See B. Flyvbjerg, “From Nobel Prize to Project Management: Getting Risks Right,” Project Management Journal (August 2006): 5–15. Another way to use the disinterested knowledge of other people to help in forecasting project durations (and other future events) is to set up a prediction market, a sort of artificial financial futures market in which individuals invest or gamble money on making the most accurate prediction. The aggregation of multiple, independent predictions, each from someone motivated by financial gain and not personally involved in carrying out the plan, can yield much more accurate forecasts than those made by even expert individuals. For discussion, see C. R. Sunstein, Infotopia: How Many Minds Produce Knowledge (Oxford: Oxford University Press, 2006); and R. W. Hahn and P. C. Tetlock, Information Markets: A New Way of Making Decisions (Washington, DC: AEI Press, 2006).

  20. Techniques like these were studied experimentally in R. Buehler, D. Griffin, and M. Ross, “Exploring the ‘Planning Fallacy’: Why People Underestimate Their Task Completion Times,” Journal of Personality and Social Psychology 67 (1994): 366–381.

  21. Information on Brian Hunter and Amaranth Advisors comes from: A. Davis, “Blue Flameout: How Giant Bets on Natural Gas Sank Brash Hedge-Fund Trader,” The Wall Street Journal, September 19, 2006, p. A1 (online.wsj.com/article/SB115861715980366723.html); and H. Till, “The Amaranth Collapse: What Happened and What Have We Learned Thus Far?” EDHEC Business School, Lille, France, 2007. The comparison between Amaranth and other debacles is based on “List of Trading Losses” in Wikipedia, en.wikipedia.org/wiki/List_of_trading_losses (accessed March 27, 2009).

  22. Information on various investment strategies comes from the following sources: “Dow Theory” in Wikipedia, en.wikipedia.org/wiki/Dow_theory (accessed March 25, 2009); discussion of the Nifty Fifty in Chapter 8, “The Amazing Two-Tier Market,” in D. N. Dreman, Psychology and the Stock Market: Investment Strategy Beyond Random Walk (New York: Amacom, 1977). “Dogs of the Dow” is a nickname for a strategy proposed by Michael O’Higgins in his book Beating the Dow: A High-Return, Low-Risk Method for Investing in the Dow Jones Industrial Stocks with as Little as $5000 (New York: HarperCollins, 1991). The “Foolish Four” strategy, a derivative of one of O’Higgins’s ideas, is described by Robert Sheard in The Unemotional Investor: Simple Systems for Beating the Market (New York: Simon & Schuster, 1998). Both of the latter two books were bestsellers.

  23. It is arguably wrong to view a house as an investment. A typical asset bought for investment purposes is not usable while you own it; there’s nothing you can physically do with your Google stock or your municipal bonds or your money-market funds. (You can’t even frame your pretty stock certificates anymore, unless you make a special request for them from your broker.) The right way to think of a house is as a hybrid of a consumable product that must be repaired and upgraded over time, like a car or a computer, and an underlying investment (which is based partly on the value of the land where it stands).

  People make mistakes when thinking about housing prices for a variety of reasons, one of which is failing to make this distinction. For example, many homeowners mistakenly believe that improving their homes will increase the home’s value by a greater amount than the cost of the improvement; in fact, every one of twenty-nine common home improvements yields an average increase in resale value less than 100 percent of its cost (see “Remodeling 2007 Cost Versus Value Report” [www.remodeling.hw.net/costvsvalue/index.html]; and D. Crook, The Wall Street Journal Complete Homeowner’s Guidebook [New York: Three Rivers Press, 2008]). Remodeling a home office costs $27,193 on average, but increases the home’s value by only $15,498, or 57 percent of the original expenditure, not counting any interest paid if the remodeling was financed. Even remodeling a kitchen, one of the classic value centers of a house, returns only 74 percent of the money spent. Look at it this way: If your house would sell for $500,000 today, but you decide to “invest” $40,000 in a new kitchen before you put the house on the market, you should expect to get about $530,000 for it. Putting the same money in the bank would be a much better investment: You wouldn’t earn much in interest, but at least you wouldn’t lose the $10,000!

  When told these facts, people often become incredulous and even angry—precisely because they contradict a foundational piece of “knowledge” homeowners have about their “investments.” We will return to this subject later in this chapter when we discuss the necessary conditions for financial bubbles and panics. There are, of course, other reasons to remodel a house besides any expected “investment” gain: A recent study showed that additional full or half bathrooms in a house were more strongly associated with owner satisfaction than any other feature measured, including additional bedrooms, air conditioning, and a garage. See R. N. James III, “Investing in Housing Characteristics That Count: A Cross-Sectional and Longitudinal Analysis of Bathrooms, Bathroom Additions, and Residential Satisfaction,” Housing and Society 35 (2008): 67–82.

  24. M. Piazzesi and M. Schneider, “Momentum Traders in the Housing Market: Survey Evidence and a Search Model,” Stanford University manuscript, 2009, www.stanford.edu/~piazzesi/momentum%20in%20housing%20search.pdf (accessed August 17, 2009).

  25. Alberto Ramirez’s mortgage story is from C. Lloyd, “Minorities Are the Emerging Face of the Subprime Crisis,” SF Gate, April 13, 2007 (www.sfgate.com/cgi-bin/article.cgi?f=/g/a/2007/04/13/carollloyd.DTL). Ninja loans, and other bad home-finance ideas, are mentioned in S. Pearlstein, “‘No Money Down’ Falls Flat,” The Washington Post, March 14, 2007, p. D1 (www.washingtonpost.com/wp-dyn/content/article/2007/03/13/AR2007031301733_pf.html). Ed Glaeser’s quote comes from E. Glaeser, “In Housing, Even Hindsight Isn’t 20–20,” The New York Times Economix blog, July 7, 2009 (economix.blogs.nytimes.com/2009/07/07/in-housing-even-hindsight-isnt-20–20/?hp).

  26. R. Lowenstein, “Triple-A Failure,” The New York Times Magazine, April 27, 2008 (www.nytimes.com/2008/04/27/magazine/27Credit-t.html). Similar problems beset so-called “quant” funds, which were hedge funds that made trading decisions entirely or mostly based on the predictions of computer models that were calibrated with historical data that didn’t include market conditions like the increasingly risky environment of 2007. See H. Sender and K. Kelly, “Blind to Trend, ‘Quant’ Funds Pay Heavy Price,” The Wall Street Journal, August 9, 2007.

  27. R. H. Thaler, A. Tversky, D. Kahneman, and A. Schwartz, “The Effect of Myopia and Loss Aversion on Risk Taking: An Experimental Test
,” Quarterly Journal of Economics 112 (1997): 647–661.

  28. Interestingly, the most active traders also tended to have smaller portfolios at the beginning of the study than did the least active ones; obviously this difference would tend to magnify over time since their net returns would be lower as well. See B. Barber and T. Odean, “Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors,” Journal of Finance 55 (2000): 773–806. Men, especially single men, also trade much more frequently than women, and earn correspondingly lower returns on their investments. See also B. Barber and T. Odean, “Boys Will Be Boys: Gender, Overconfidence, and Common Stock Investment,” Quarterly Journal of Economics 116 (2001): 261–292.

  29. Unless you are a coin collector, you don’t know enough to distinguish a counterfeit penny from a real one. Even coin collectors might fail to recognize subtle changes unless they’re actively looking for them. As a child, Dan collected coins, and he did spot one obvious fake. He was at a coin show, and a vendor was selling a really old coin that he claimed was from ancient Greece. The coin was well worn, with few details still visible. It certainly looked like it could be more than two thousand years old, and the figure on the front looked like a Greek hero. Dan didn’t buy it, though—it had a date of “300BC” partially visible below the figure! (Some counterfeiters apparently are not terribly bright.)

  30. The idea that the mind works like a Web browser comes from R. A. Rensink, “The Dynamic Representation of Scenes,” Visual Cognition 7 (2000): 17–42. In philosophy and psychology, metaphors for the workings of the mind often draw on the latest and greatest in technology. Early models of the mind appealed to the notions of hydraulics, with the flows of fluids causing different thoughts and actions. Such models were gradually replaced by the notion of the mind as a mechanical device, with metaphorical gears. In the 1960s, the dominant model of the mind was as an information-processing device. Essentially, the mind was treated as a powerful computer. The computer metaphor continues to hold sway in psychology, with some adjustments corresponding to further changes in technology: an emphasis on the parallel nature of processing, off-loading of some types of processing to specialized modules (just as computer graphics are often handled by a special chip set), and so on. For an interesting discussion of the effects of technological developments on the nature of scientific theories, see G. Gigerenzer, “From Tools to Theories: A Heuristic of Discovery in Cognitive Psychology,” Psychological Review 98 (1991): 254–267.

  31. B. Popken, “Do Coat Hangers Sound as Good as Monster Cables?” The Consumerist blog, March 3, 2008, consumerist.com/362926/do-coat-hangers-sound-as-good-monster-cables (accessed June 29, 2009).

  32. If you want some snarky entertainment, read the user reviews of the Denon cable at Amazon.com. Just search the site for “Denon Ethernet cable.” As of August 2009, one Amazon user was even offering one of these cables “used” for sale at $2,500!

  33. D. S. Weisberg, F. C. Keil, J. Goodstein, E. Rawson, and J. R. Gray, “The Seductive Allure of Neuroscience Explanations,” Journal of Cognitive Neuroscience 20 (2008): 470–477. The “curse of knowledge” described in the example we gave from this experiment has implications for the illusion of knowledge. If we assume that other people know what we know, and we think we know more than we do, then we must think other people know more than they do as well!

  34. These results are from Experiment 1 of D. P. McCabe and A. D. Castel, “Seeing Is Believing: The Effect of Brain Images on Judgments of Scientific Reasoning,” Cognition 107 (2008): 343–352.

  35. The Allstate ad is on the company’s website, www.allstate.com/content/refresh-attachments/Brain-Ad.pdf (accessed November 15, 2009).

  36. Agricultural facts taken from Wikipedia, en.wikipedia.org/wiki/Illinois (accessed February 27, 2009).

  37. Details about Illinois weather forecasting and WILL are from an interview with Ed Kieser conducted by Dan on February 27, 2009.

  38. P. Hughes, “The Great Leap Forward: On the 125th Anniversary of the Weather Service, A Look at the Invention That Got It Started,” Weatherwise 47, no. 5 (1994): 22–27.

  39. J. P. Charba and W. H. Klein, “Skill in Precipitation Forecasting in the National Weather Service,” Bulletin of the American Meteorological Society 61 (1980): 1546–1555. There has been much discussion of “chaos” in physical systems like the earth’s climate, and the now-clichéd idea that a butterfly can flap its wings on one side of the world and influence the weather weeks later on the opposite side of the world. None of this makes it impossible to predict whether it will rain tomorrow.

  40. This demonstration was suggested by one of Dan’s teaching assistants, Richard Yao, who experienced it in a class as an undergraduate at Northwestern University.

  41. R. A. Price and S. G. Vandenberg, “Matching for Physical Attractiveness in Married Couples,” Personality and Social Psychology Bulletin 5 (1979): 398–400.

  42. The meteorologist preference question was asked of the 72 chess players in Philadelphia who participated in the study of overconfidence in chess ability that we discussed in Chapter 3. The question was first used in G. Keren, “On the Calibration of Probability Judgments: Some Critical Comments and Alternative Perspectives,” Journal of Behavioral Decision Making 10 (1997): 269–278. See also G. Keren and K. H. Teigen, “Why Is p = .90 Better Than p = .70? Preference for Definitive Predictions by Lay Consumers of Probability Judgments,” Psychonomic Bulletin and Review 8 (2001): 191–202. The popular preference for certainty in weather reports was noted anecdotally over a century ago. When William Ernest Cooke introduced estimates of uncertainty to weather forecasting in 1906, he predicted that the public would prefer his new method, but immediately below his first article, a note by one Professor E. B. Garriott appeared, giving no fewer than five specific arguments why Cooke’s “scheme” was impractical, concluding with “because our public insist upon having our forecasts expressed concisely and in unequivocal terms.” W. E. Cooke, “Forecasts and Verifications in Western Australia,” Monthly Weather Review 34 (1906): 23–24.

  43. P. E. Tetlock, Expert Political Judgment: How Good Is It? How Can We Know? (Princeton, NJ: Princeton University Press, 2005). In weather forecasting, meteorologists understand the need to show that over time their methods outperform a simple model that assumes that tomorrow’s weather will be the same as today’s weather. And they are easily able to make enough verifiable predictions to show that they can beat such models. People in many other disciplines lack that ready source of feedback and they often do not check whether their models can outperform such simple heuristics. Even when they do have access to such data (e.g., public financial data can be used to determine whether a money manager’s method of actively picking stocks outperforms the returns of a passive index fund), they often do not bother to check. If they did, perhaps they would not express quite as much confidence as they do.

  44. We thank our editor, Rick Horgan, for suggesting these two examples.

  45. Citation for Herbert Simon from Nobel Prize website (nobelprize.org/nobel_prizes/economics/laureates/1978/index.html).

  46. In August 2009, Amaranth agreed to a settlement with the U.S. government over the charges, but Brian Hunter did not. As of earlier that year, he was an adviser to Peak Ridge Capital Group, where his “Commodity Volatility Fund” was up 138 percent in its first six months. “To have lost that amount of money and get back into the market with a similar-type trade takes a lot of confidence, if not arrogance,” said one industry analyst. See S. Kishan, “Ex-Amaranth Trader Hunter Helps Deliver 17% Gain for Peak Ridge,” Bloomberg.com, May 19, 2009 (www.bloomberg.com/apps/news?pid=20601087&sid=aUlBVaEHAk04&refer=home); “Ex-Amaranth Trader Makes Good, Possibly,” the New York Times DealBook blog, April 11, 2008 (dealbook.blogs.nytimes.com/2008/04/11/ex-amaranth-trader-makes-good-possibly/); A. Davis, “Amaranth Case Shows Trading’s Dark Side,” The Wall Street Journal, July 26, 2007, p. C3; C. Kahn, “Federal Judge Orders Amaranth Advisors to Pay $7.5M for Price M
anipulation,” Associated Press, August 12, 2009 (ca.news.finance.yahoo.com/s/12082009/2/biz-finance-federal-judge-orders-amaranth-advisors-pay-7–5m.html); J. Strasburg, “A De cade Later, Meriwether Must Scramble Again,” The Wall Street Journal, March 27, 2008, p. C1 (online. wsj.com/article/SB120658664128767911.html); and G. Zuckerman and C. Karmin, “Rebounds by Hedge-Fund Stars Prove ‘It’s a Mulligan Industry,’” The Wall Street Journal, May 12, 2008, p. C1 (online.wsj.com/article/SB121055428158584071.html).

  Chapter 5: Jumping to Conclusions

  1. Details from this case and the subsequent outbreak of measles in Indiana were taken from the CDC report “Import-Associated Measles Outbreak—Indiana, May–June 2005,” Morbidity and Mortality Weekly Report (MMWR) 54 (October 27, 2005): 10731075. Other details came from A. A. Parker, W. Staggs, G. H. Dayan, I. R. Ortega-Sánchez, P. A. Rota, L. Lowe, P. Boardman, R. Teclaw, C. Graves, and C. W. LeBaron, “Implications of a 2005 Measles Outbreak in Indiana for Sustained Elimination of Measles in the United States,” New England Journal of Medicine 355 (2006): 447–455. Other information about measles discussed in this section comes from the preceding sources as well as the following additional sources: World Health Organization Measles Fact Sheet, www.who.int/mediacentre/factsheets/fs286/en/ (accessed March 24, 2009); CDC report “Outbreak of Measles—San Diego, California, January–February 2008,” Morbidity and Mortality Weekly Report (MMWR) 57 (February 22, 2008): 203–206; “Confirmed Measles Cases in England and Wales: An Update to End–May 2008,” 2008, Health Protection Report 2, no. 25 (2008); S. B. Omar, W. K. Y. Pan, N. A. Halsey, L. H. Moulton, A. M. Navar, M. Pierce, and D. A. Salmon, “Nonmedical Exemptions to School Immunization Requirements: Secular Trends and Association of State Policies with Pertussis Incidence,” Journal of the American Medical Association 296 (2006): 1757–1763; CDC report “Measles—United States, January 1–April 25, 2008,” Morbidity and Mortality Weekly Report (MMWR) 57 (May 1, 2008): 494–498; CDC report “Update: Measles—United States, January–July 2008,” Morbidity and Mortality Weekly Report (MMWR) 57 (May 1, 2008): 893–896. Information about the measles outbreak in Romania from: Associated Press, “Measles Outbreak Sickens 4000 in Romania,” December 5, 2005. After we wrote this chapter, an excellent article reporting on this case and its implications was published in Wired: A. Wallace, “An Epidemic of Fear: How Panicked Parents Skipping Shots Endangers Us All,” Wired, November 2009, www.wired.com/magazine/2009/10/ff_waronscience/.