The Reality of Prisoner’s Dilemma
Most people who have taken basic Econ or Networks probably know the concept “The Prisoner’s Dilemma”. The underlying concept of Prisoner’s Dilemma goes as follows: Two prisoners who are arrested are put into separate rooms without means of communication. The authorities do not have enough evidence to convict either criminals of the primary charge. However, if one testifies against the other, he/she walks free while the other gets a 3 year sentence; this works for either criminal. Though, if both prisoners testify, they each get 2 years. Lastly, if both remain silent, they each get one year.
In this game, the dominant strategy, which is the best strategy for each player, is to testify regardless of the decision of the other. It is the dominant strategy because if he/she doesn’t testify, he will either get 1 year if the other doesn’t testify or 3 years (if the other does testify) of prison. However, if he/she does testify, then the prisoner can either walk free 0 years (which better than 1) or 2 years (which better than 3). Thus, the best decision, AKA Nash equilibrium, of the criminals would be to testify against the other even though the total payoff would be higher if they both remained silent (combined 2 year sentence).
Despite the expected outcome of the Nash Equilibrium, the reality did not align with expectations. Two economists from the University of Hamburg ran the experiment on a group of female inmates and students, but instead of jail time, the participants were offered coffee, cigarettes, and cash instead. To many people’s surprise, the inmates actually cooperated 56% of the times while the students only cooperated 37% of the time (i.e both remained silent). Despite the prediction that the majority of prisoners and students will testify against one another, prisoners were far more cooperative than expected. Not only does this experiment challenge game theory’s actual implications on reality, but the experiment also showed that prisoners are not as “calculating, self-interested, and un-trusting” as people may expect. From an economics perspective, prisoners, according to their dominant strategy, would benefit more from betraying one another, but evidently from the experiment, the supposed violent, shunned criminals showed more trust, bonding and faith than our students.
Resources:
http://www.businessinsider.com/prisoners-dilemma-in-real-life-2013-7