Thinking, Fast and Slow

Kahneman demonstrates that two statements about the results of the 2006 World Cup final: “Italy won” and “France lost” are logically equivalent but evoke different associations and meanings. This feature of System 1 makes it difficult for people to act consistently when presented with different frames.

Kahneman and Tversky applied frames to gambles in these two scenarios. They asked some participants, “Would you accept a gamble that offers a 10% chance to win $95 and a 90% chance to lose $5?” They asked others, “Would you pay $5 to participate in a lottery that offers a 10% chance to win $100 and a 90% chance to win nothing?” These two problems are identical, but the second usually attracts many more positive answers. Losses evoke stronger negative feelings than costs.

In an essay, Richard Thaler describes another example: the credit card lobby pushed against gas stations that charged more if people paid with a credit card. But their fallback position was to request that the vendors call it a “cash discount” rather than a “credit surcharge,” because people will more readily forgo a discount than pay a surcharge.

In another experiment conducted by British psychologists, participants are given £50 and then asked to choose between a sure outcome of £20 and a game of chance, in which the person has a 2/5 chance to keep the entire amount that they have been given. The £20 can either be framed as KEEP £20 or LOSE £30. When the frame is designated as KEEP £20, subjects more likely choose the sure thing. When it is LOSE £30, subjects are more likely to gamble.

People’s brain activity is monitored during this experiment. When the subject chose the more frequent option (in either frame), a region associated with emotional arousal was active. When subjects did not do what comes naturally, a brain region associated with conflict and self-control was active.

An experiment that Tversky carried out is another example of emotional framing, in which two outcomes of surgery are described to physicians: “The one-month survival rate is 90% and “There is a 10% mortality in the first month.” Recommending surgery was more popular in the former frame than in the latter.

Kahneman and Tversky also explored framing with this example: the U.S. is preparing for the outbreak of an unusual Asian disease, which is expected to kill 600 people. If program A is adopted to combat it, 200 people will be saved. If program B is adopted, there is a 1/3 probability that 600 people will be saved, and a 2/3 probability that no people will be saved. A majority of people choose program A.

Now consider different framings: If program A’ is adopted, 400 people will die. If program B’ is adopted, there is a 1/3 probability that nobody will die and a 2/3 probability that 600 people will die. The consequences of A and A’ are the same, as are B and B’. In the second frame, however, most people choose the gamble.

When people are confronted with this inconsistency, they often don’t know how to decide. They know intuitively that saving lives with certainty is good, and certain death is bad. But System 2 does not always have a way of answering the question on its own. Our moral intuitions too often rely on descriptions, not substance.

Some frames can be more useful than others. Kahneman asks readers to consider a pair of problems. In the first, a woman loses two $80 tickets. In the second problem, a woman loses $160 in cash. Kahneman asks whether the first woman would buy two more tickets, and whether the second would buy tickets anyway. People who see the problems believe that the first woman not buy two more tickets, and the second woman will buy tickets. Kahneman advises that in each case, sunk costs should be ignored. He would ask the first woman, “Would you have bought the tickets if you had lost the equivalent amount of cash?” If yes, she should buy two more tickets.

Another example of bad framing centers on two drivers: Adam and Beth. Adam switches from a gas-guzzling car with 12 mpg to one with 14 mpg. Beth switches from a 30 mpg car to a 40 mpg car. Intuitively, people think that Beth is saving more gas, but if each person drives 10,000 miles, Adam will reduce his consumption by 119 gallons, and Beth will only reduce hers by 83 gallons. The mpg frame is wrong, and Cass Sunstein, a psychologist who worked with Richard Thaler, helped to change policy that now requires fuel economy information to be printed in gallons-per-mile instead of mpg.

Framing is important in yet another example regarding organ donation. In 2003, a study was done that discovered that nearly 100% of people in Austria choose to be organ donors, but only 12% in Germany; 86% in Sweden, but 4% in Denmark. The difference is that the high-donation countries must check a box to opt out of organ donation; low-contribution countries must check a box to become a donor. An important choice is controlled by an inconsequential feature of the situation.