Here’s a nominee for strangest psychology experiment ever, or at least spookiest. Yesterday I talked about the theory that religion, or at least supernatural agent concepts, serve to activate representational concerns, and thus increase prosocial behavior, or decrease selfish behavior. The experiment I’m about to describe was designed to test part of that theory. But given how odd the experiment is, I don’t really care what the theoretical motivations for it were. It’s post-worthy simply by virtue of its bizzarness.
The experiment, conducted by Bering et al.1, involved a boring enough task. They selected the 25 most difficult mental rotation problems (designed for “experienced users”) from a common spatial reasoning test, and told participants that they were going to be completing a new and as of yet untested spatial reasoning task. Participants were told that because the task was new, there were some glitches in the program. One of these glitches would occasionally cause the answer to a mental rotation problem before the problem was presented. If this happened, participants were instructed to hit the space bar as quickly as possible so as to avoid being able to use the answer to solve the problem. Here’s the relevant passage from the instructions:
IMPORTANT NOTE: Because this is a new test, the computer program periodically malfunctions. In some instances, the correct answer may appear on the screen BEFORE the actual problem. If you see the word “ANSWER” at any time, this is a mistake (this is the correct answer to the following problem). If this happens, please press the space bar immediately so that you can solve the problem honestly. ONLY BY PRESSING THE SPACE BAR WILL THE SCREEN BE CLEARED. Thank you for your patience while we attempt to fix this problem.
Before starting, participants were also told that the top scorer on the test would win $50. So, the possibility of a financial reward, combined with the difficulty of the problems, provided participants with a lot of incentive to cheat by waiting to press the space bar when the answers were “accidentally” displayed before the problems. Bering et al. then used the time it took participants to press the space bar when an answer appeared as a measure of cheating or willingness to cheat.
Not very interesting so far, right? Well, here’s where it gets strange. The participants were divided into three conditions. Those in the control group just took the test as I’ve described it. For those in the “In Memoriam” condition, the following note was attached to the instructions:
In Memoriam: This test is dedicated to the memory of Paul J. Kellogg, who died unexpectedly in May 2004. Paul was a graduate student in the department, and his contributions to the development of this spatial intelligence test were invaluable.
Still not all that strange. But then there’s the third condition. These participants read the “In Memoriam” notice, and were told “as a casual but serious aside” that the experimenter had recently seen the ghost of the dead graduate student in the room. The participants in this “Ghost Story” condition had therefore been primed to think about the presence of a supernatural agent — a dead graduate student. The “representational concern” theory of the purpose of supernatural agents predicts that participants primed to think about a dead graduate student in the room will act as though someone is watching them, and therefore be less likely to cheat than participants in the control or in memoriam conditions.
Perhaps because they were scared of a dead graduate student’s ghost hanging around, participants in the ghost story condition performed worse than those in the other two conditions on the mental rotation problems. More importantly, participants in the ghost story condition pressed the space bar faster than those in the other two conditions (though the difference between their times and those of the participants in the in memoriam condition only approached significance) when the answers were “mistakenly” shown prior to the problems. It appears, then, that thinking about the presence of a ghost in the room made the participants less likely to cheat by looking at the answers before getting the problems.
Aside from providing evidence for the “reputational concern” function of supernatural agents, this study also provides a method that educators can use to prevent their students from cheating. I, for one, plan to include in all my syllabi a memorial notice for a fictitious faculty member, and to casually mention at the beginning of the first class that the ghost of that faculty member has been seen in the classroom on several occasions. While this may cause their performance in the course to drop, it should also curb cheating, and I’m willing to make that trade off.
1Bering, J. M., McLeod, K., & Shackelford, T. K. (2005). Reasoning about dead agents reveals possible adaptive trends. Human Nature, 16, 360-381.