Invasion by a Tit-for-Tat Minority
Main Menu Previous Topic Next Topic
Four Tit-for-Tats amongst All-D's
Note: in order to run the simulation referred to in this slide, go here, to the Java Applet version. You will be directed to download the latest version of the Java plug-in.
In the simulation to the left the four agents in the middle of the square start off with Tit-for-Tat strategies, while the rest start off as All-D's. Note that the Tit-for-Tats are quite close to each other. Consequently, we can increase α to increase the likelihood of a Tit-for-Tat interacting with another Tit-for-Tat.
Indeed, every time you restart this graphic, and press "Go" once, you may notice that the four agents in the middle are more likely to connect to each other for high values of α. Furthermore, when they do connect to each other, they get a non-zero cooperation payoff. On the other hand, if one of them connects to an All-D partner instead, her payoff stays at zero. Subsequent clicks of the "Go" button will run the simulation faster, and you may gauge for yourself how territorial structure affects the success of the strategy invasion.
Note that for higher values of α, the simulation takes longer to reach a stasis: agents don't interact with as many different partners during each generation, and thus learning occurs at a slower rate.
Previous Slide Next Slide