Posts Tagged ‘experiment’
I’ve been reading Philip Zimbardo’s book The Lucifer Effect: Understanding How Good People Turn Evil, a book which contains a details description of the Stanford Prison Experiment, and an explanation of the dangers which exist in a system where there are roles which create an expectation of certain behaviours there is a likelihood that these behaviours will be expressed. It leads me to believe that there are Nash Equilibria here, looking at the SPE it is clear that it coordination game in which the Guard and the Prisoner should adopt the same or corresponding strategies to achieve the highest pay-off.
I was amused to not see a reference to Game Theory in Zimbardo’s book, as the Prisoners Dilemma is the most “fundamental problem in game theory that demonstrates why two people might not cooperate even if it is in both their best interests to do so.” The Prisoners Dilemma is a game with cooperation being a strictly dominated strategy – dominated by abuse in the SPE game – this doesn’t entirely translate to the SPE. The Guard does better than the Prisoner in most strategies both, although cooperation strictly dominates all other strategies.
The influence of role on the game is clear from the experiment, and many other examples, this means that role-based actions will influence strategy, this could be called a role-based strategy. The role-based strategy does not necessarily correspond with an optimal strategy, the fable of the the scorpion and the frog is an instance of choosing a strictly dominated role-based strategy.
Good read, food for thought.