Abstract
In a long sequence of iterated Prisoner's Dilemma played by a pair of subjects, "lock-in " effects are frequently observed both on the CC and on the DD outcome. That is, in the course of the sequence either CC or DD becomes predominant sometimes persisting to the end. A lock-in on CC can be explained by a deterrence effect: The subjects learn that defection to D, although immediately advantageous if the co-player continues with C, is likely to be followed by retaliation. A lock-in on DD can be explained by the inhibition of unilateral C (the "peace initiative"), which is associated with the sucker's payoff In the present experiment, subjects played repeated round robin sequences of Prisoner's Dilemma with other subjects in their experimental group. Prominent lock-in effects were observed when the subjects were informed on each play how their current co-players chose when paired with other subjects but not when no such information was given.
Get full access to this article
View all access options for this article.