Tuesday, July 01, 2008

Changing belief in free will can cause students to cheat

This time, instead of just reading a passage, two groups of students read a booklet fifteen phrases such as "A belief in free will contradicts the known fact that the universe is governed by lawful principles of science," and were required to ponder each phrase for a full minute before turning to the next page. Another group read similar statements advocating free will, and another group read neutral statements. A final baseline group didn't read any statements at all. Everyone was given the same test, which was composed of 15 problems from the GRE exam, a rigorous test designed to assess applicants for graduate school.

One determinism group, the free will group, and the neutral group took the test in sets of 2 to 5 individuals. For these groups, a tempting possibility of cheating was introduced. The experimenter told them that she had to leave the room for a meeting, and that they were to take no more than 15 minutes to take the test, which they should grade themselves. She would not see their test sheets: they were to be destroyed in a document shredder in the room. Then they could award themselves $1 for every correct response from an envelope of coins she left on a table in the room.

The other two groups (the second determinism group and the baseline group) were tested individually, the experimenter graded their tests, and paid them based on their actual scores...

The beauty of this experiment is that the experimenter honestly didn't know how much money each person in the "cheating possible" groups took. She could only count the money after they left and find the average take per person. She couldn't even assess their answers on the test -- those were shredded. Yet those who had read the determinism statements took significantly more money when they were given the opportunity to cheat. The fact that reading about determinism didn't lead to higher scores when cheating was not possible suggests that indeed, cheating was going on, not just super-performance due to reading about determinism.

No comments: