Homor Among Thieves: College Students Twice More Likely To Rat Than Inmates

PRISONERS THWART PRISONER’S DILEMMA

JULY 24, 2013 AVERY APPELMAN LEAVE A COMMENT

A study by two economists at the University of Hamburg revealed that when presented with the “prisoner’s dilemma”, inmates actually achieved a more favorable outcome than college students.

The prisoner’s dilemma goes as follows: Two criminals are arrested on suspicion of burglary. The cops believe the prisoners worked together during the heist, but they don’t have enough evidence to convict them on 1st degree burglary Prionunless the criminals testify against one another. The suspects are brought into separate rooms for questioning with no means of communication.

They are then told that if they testify and their accomplice remains silent, they will walk free and their partner will get three years in jail. On the other hand, if they remain silent and their partner agrees to testify, they will get three years in prison and their accomplice will go free. If both agree to testify, they will each receive two years in prison. If neither agrees to testify, the cops will only be able to book them on trespassing charges, and they’ll each receive one year in prison.

According to game theory, also known as strategic decision-making, choosing to testify is always the more popular answer. Let’s assume, for example, you knew what your accomplice was going to do. In each case, for the lesser sentence, you would choose to testify. If you knew your accomplice was staying silent, you could walk free if you testified. If you knew your accomplice was talking, your testimony would reduce your sentence from three years to two. In this rational sense, it leads one to believe that testifying will be your best option, when in fact, mutual cooperation leads to the greatest result for both parties.

Testing the Theory on Prisoners

Although termed the prisoners dilemma, the study had never been conducted on a prison population. Menusch Khadjavi and Andreas Lange decided to put the theory to test, and they used a group of college students as a control group.

Because humans are not purely rational individuals, researchers did not expect all 100% of either group to betray their partner. They did, however, believe that the inmate population might be more jaded and distrustful, meaning they may be more likely to betray their partner. The results showed quite the opposite:

• When students were presented with the dilemma, only 37% opted not to betray their partner.
• When prisoners were presented with the dilemma, 56% of inmates decided not to betray their partner.
• When examined in a pair setting, only 13% of students managed to get the best mutual outcome for both players. Prisoners obtained the best mutual outcome for both players 30% of the time.

Although they weren’t actually presented with years in prison (money for the students and coffee and cigarettes for the prisoners were used), the study highlights what some behavioral economists have been arguing for years; that convicts aren’t as distrusting as one might naturally believe.