Here’s an idea that has been bugging me for a while: what if the Tit for Tat game theory strategy is not successful in the real world, ever?
Here’s why – Tit for Tat works because it creates a (technically unstable) equilibrium. In models of reward/retribution between two players Tit for Tat is always the best option.
But we know that in the real world there are rarely two systems that are isolated, and grievances are normally propagated down generations when retribution is against a different individual that has been strereo-typed as having similar characteristics. This creates a potentially infinite cycle of violence, even when Tit for Tat is used as a strategy by every actor.
I believe this could be modeled very simply by creating a 3 party risk/reward game where a certain percentage of plays from a -> b would randomly affect party c (but the players would not know about this random variation). This is a better simulation of real-life and I believe would show that Tit for Tat does not work without a more altruistic dampening affect.