"Perhaps the most revealing experimental “anomaly” came from a study at Stanford University, where one group of subjects was instructed to play a public goods game called “the community game,” and another asked to play one called “the Wall Street game.” The two problems were in fact identical, but the rates of cooperation in “the community game” were twice as high as in “the Wall Street game.” (Cooperation in the former was around the “normal” rate of two-thirds; in the latter it was abnormally low, at about one-third.) Another version of this study, carried out among Israeli air force trainers, showed approximately the same effect. In that study, however, instructors were asked to predict how their trainees (whom they knew quite well) would behave. Interestingly, not only were the instructors wrong (they did no better than chance at predicting the behavior of their trainees), but they also made the mistake of ignoring the effect that the name of the game might have upon rates of cooperation and defection. In this respect, the trainers fell victim to the same fallacy that has plagued generations of economists. It is, in fact, a general flaw in our everyday social reasoning, which social psychologists refer to as an extrinsic incentive bias. Simply put, people have a tendency to overestimate the importance of external incentives in motivating human conduct. In particular, we have a natural tendency to overestimate the influence of power, money, and status in motivating other people’s decisions."