This is not a Prisoner’s Dilemma, or is it?

A little bit ago, Cory Doctorow posted a story from Inside Higher Ed about students organizing to beat the curve in a Johns Hopkins computer science class. The professor, Peter Fröhlich, scales grades based on the highest grade1. The students all refused to take the test, making the highest grade a 0. Thus, a 0 was an A, meaning they all got As.

…students in Fröhlich’s…classes decided to test the limits of the policy, and collectively planned to boycott the final. Because they all did, a zero was the highest score in each of the three classes, which, by the rules of Fröhlich’s curve, meant every student received an A…The students waited outside the rooms to make sure that others honored the boycott, and were poised to go in if someone had. – Zack Burdyk, “Dangerous Curves” from Inside Higher Ed

Doctorow labeled this as a solution to the Prisoner’s Dilemma. A brief perusal of the 113 comments on his post will convince you that the internet thinks that this is NOT an example of the Prisoner’s Dilemma. It certainly was not an example of the classic version of a Prisoner’s Dilemma as the participants were able to directly communicate with each other. In the classic version, the two prisoners involved are not able to exchange any information. There are, however, many variations on the basic structure that manipulate the variables (eg, number of participants, incentives, number of times played, etc) over wide ranges, making it much harder to say that something is not a Prisoner’s Dilemma than it is to say that something is a Prisoner’s Dilemma.

It was a cooperative effort on the part of the students, which may have brought the Prisoner’s Dilemma to mind. In The Selfish Gene, Richard Dawkins popularized the work of Robert Axelrod, which showed that their were dominant strategies in an iterated Prisoner’s Dilemma (ie, played multiple times by the same players) that favored cooperation. Variations on this result have been important for research on the evolution of altruism beyond genetic relatedness.

I am not going to say that this is not a solution to a Prisoner’s Dilemma. It is a solution to a Tragedy of the Commons problem, in which individual, short term self-interest leads to over exploitation  Some define the Tragedy of the Commons as a multiplayer version of the Prisoner’s Dilemma. Some do not. Importantly, research has now shown that, both theoretically and historically (eg, the work of Elinor Ostrom), small, local groups can “solve” the Tragedy of the Commons via communication and regulation. This appears to be what these students did.

They communicated and organized using social media. They enforced that communication with the social pressure of sitting immediately outside the classroom, forcing any defectors to face those they were betraying and censure by the group. Unlike, the classic Prisoner’s Dilemma, defectors could not defect in secret. They stood ready to punish defectors by all entering and taking the test, thus reducing a defectors potential benefit2.

So, BoingBoing commenters. This is not really an issue of whether Cory Doctorow is right or wrong, but whether you think the Tragedy of Commons is a multiplayer Prisoner’s Dilemma or not3.

1. I am an active opponent of “grading on a curve” for a variety of reasons. Perhaps most important among these reasons is that it treats grading students as a method of ranking them, not as a method for assessing their competency in the subject. Ranking students alone suggests that the goal of education is to sort the “wheat” from the “chaff”, not to transfer subject matter knowledge and improve critical thinking.
2. This represents another deviation from the classical form of the Prisoner’s Dilemma, which is a simultaneous game (ie, players play at the same time – or, at least, with no knowledge of the other player’s choice). This situation was turn based. The students sitting outside the classroom door had the ability to change their choice of action if they observed another student defecting.
3. I think I come down on the side of the Tragedy of the Commons not being a Prisoner’s Dilemma variant, but I freely admit that I do not know what I’m talking about.

Author: Josh Witten

One thought on “This is not a Prisoner’s Dilemma, or is it?”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: