Skip to main content

Blog 1, Networks 2040

Using Prisoner’s Dilemma in a College Course

             I recently read an article by Professor Selterman from the University of Maryland that implemented a prisoner’s dilemma exercise on his class. He asked his students to select whether they wanted either 2 or 6 points added to their final term paper grade; however, the catch was that if more than 10% of the class selected 6 points as their choice, then no student in the class would get any extra credit. This exercise is definitely one that revolves around selfishness, greed, and social trust.  Naturally, the best choice for the class as a whole is to pick 2 points so that you guarantee yourself, and everyone else, extra credit. Yet, the professor noted that with the exception of one class, every class has exceeded the 10% rule and, consequently, got zero extra credit.

This article relates directly to what we have learned in class about game theory. In our textbook, we have studied examples such as deciding to prepare for either a joint presentation with another person or study for an individual exam.  Naturally, the two people in this dilemma have to do what is in their best interests, and so their final choices are based on these interests. Yet, their final choices lead them to an outcome that is not the optimal one.  Studying for the exam is a strictly dominant strategy, just how picking 6 points is a dominant strategy. Yet, just how the partners know that preparing for the presentation will make them both better off, they each still have an incentive to prepare for the exam so as to achieve an even higher payoff of 92, as opposed to 90; likewise, the same idea goes for the extra credit. But, with this considered, and referring to the graph on page 141 in our textbook, the outcome of 90,90 for each student cannot be achieved by “rational play,” as each student will ultimately go the route of studying for the exam and get the result of 88,88 (Networks p. 141-143).  Similarly, more than 10% of the class in Professor Selterman’s exercise will usually (every case except for 1) choose 6 points out of an incentive for a higher payoff, and end up with zero extra credit. This extra credit exercise is yet another good example of reasoning about behavior in a game.


Link to Article:

Networks, Crowds and Markets,” page 141-143


Leave a Reply

Blogging Calendar

September 2015
« Aug   Oct »