Negative image of people produces selfish actions

People's opinion of others determines how cooperative they are

April 12, 2011

The expectations people have about how others will behave play a large role in determining whether people cooperate with each other or not. And moreover that very first expectation, or impression, is hard to change. "This is particularly true when the impression is a negative one," says Michael Kurschilgen from the Max Planck Institute for Research on Collective Goods in Bonn, summarising the key findings of a study in which he and his colleagues Christoph Engel and Sebastian Kube examined the results of so-called public good games. One’s own expectation thereby becomes a self-fulfilling prophecy: those who expect people to act selfishly, actually experience uncooperative behaviour from others more often.

Surprisingly, there are significant differences between Bonn and London in the willingness to invest in the common good. Londoners invested a mere 43 per cent, on average, in the common good. In Bonn, on the other hand, the figure was 82 per cent. "This is probably down to differing expectations of what constitutes normal behaviour," postulates Kurschilgen. Individuals who assume that the others will act selfishly too are hardly likely to commit altruistic deeds themselves. "From that point of view, Londoners have a more pessimistic view of man than do the participants in Bonn," he concludes in respect of the Brits' reticence. Consequently, whether a person decides to behave cooperatively or not depends strongly on how that person thinks the other players will behave.

In their series of experiments, Engel, Kube and Kurschilgen told their newly recruited players from Bonn the results of the London study. The players in the new round of games evidently reacted very negatively to the information that few of the players in the previous experiments in London had exhibited cooperative behaviour. Unlike the virtuous people of Bonn from the previous rounds, they showed far fewer pretensions of being good citizens: instead of investing more than 80 per cent in the common good, the participants in these experiments contributed just 51 per cent, on average. Therefore, the negative information was enough to revise the previously positive image held by the Bonn residents. This model did not really work the other way around – good examples did not make bad teammates into goody two-shoes.

"Our findings demonstrate that the core of the 'broken windows' theory does actually hold true. Faced with a social dilemma, people are guided to a very great extent by their original expectations of what other people will do, but they are also particularly sensitive to negative impressions," says Kurschilgen, summing up the observations. Given this conclusion, it is clear to him that every cent spent on maintaining residential districts does more than just make the neighbourhood look prettier – it also represents a sound investment against crime.


Go to Editor View