A study done at Stanford Univerisity has parents and teachers re-thinking our use of positive reinforcement with children. Dweck (2006) assigned middle-schoolers randomly to two different groups. They had to answer ten questions of medium difficulty, similar to those on the non-verbal section of an intelligence test. All were successful at the task. The difference was how they were praised for success. Students in the first group were told, “Great! You got (say) eight right. Clearly you are very smart at this.” Those in the other group were told, “Great! You got eight right. You obviously worked very hard at this.” Since students were assigned to the two groups at random, initial performance of both groups was similar.
However, after the different positive reinforcement that each group received, differences between the two groups became evident. The next task they were assigned had an element of choice. They had to elect between another round of questions similar in difficulty to the first ten, or choose a set of questions of greater difficulty from which they could “learn”. None of the students who were praised for their intelligence wanted to try the more difficult questions, while those praised for effort overwhelmingly (90%) chose the harder set of questions. All were given harder questions anyway. Then all were asked if they had enjoyed the exercise. Those praised for effort indicated they did enjoy it, while none of those praised for intelligence did. It seems that the harder questions put their “intelligent” self-concept at risk.
In each subsequent exercise, the performance of those praised for intelligence went down, even when they were given easier problems, similar to the first ones. Students praised for effort did better and better on each successive round. Questions were the same type used on intelligence tests, so it could be said that praising children for their intelligence made their IQ go down, while praising them for effort made them more intelligent.
In one last exercise students were told, “We are going to other schools, and we think that the students there would like to know more about the problems we did together.” They were given a piece of paper to write down their observations about the questions, with a space provided to record their scores on each round. Surprisingly, almost 40% of those praised for their intelligence lied about their scores by inflating their results, while none of those in the other group did. Researchers summarized, “The alarming thing is that we took a group of normal kids and turned them into liars just by telling them they were smart.”