Cooperation poses a bit of a problem; when selfishness represents the optimal strategy why would entities op to cooperate, especially if it means losing out in the short term and opening up oneself to exploitation? We can show acting for one's own self interest, regardless of anyone else's, represents the optimal strategy through game theory, especially prisoner's dilemma which shows defecting as the optimal strategy. This forms the foundation in classical economics of the "economic man"; a rational entity that seeks to optimise personal gain. Yet such a model doesn't fit with reality. Experiments and real world applications of the prisoner's dilemma has demonstrated that it doesn't model real world human interaction in either a social or economic sense and it fails (actually, it works well for psychopaths and economists but not for the rest of us). Psychology has show that the foundation of economics, the "economic man", also fails to model reality well. In reality we tend to favour cooperation over purely selfish actions. Why?
Why should we cooperate? Axelrod and Hamilton looked at the evolution of cooperation through the application of multiple prisoner's dilemma. They found that if rational entities could meet again and play prisoner's dilemma over and over that tit-for-tat represented the optimal strategy. They ran hundreds of trials with many different strategies and although tit-for-tat didn't come top it came second more time than other strategies meaning that overall it represented the best strategy. In tit-for-tat, cooperation gets rewarded with cooperation and defecting gets punished with defection, thus each entity has a vested interest in cooperating. So long as entities have a chance of meeting again and so long as defecting or freeloading gets punished cooperation represents to optimal strategy for self interested entities, which matches closer what actually happens in the real world.
The real world example of reciprocal altruism comes close to theoretical model of tit-for-tat from game theory. Examples include symbiosis between different animals such as cleaner fish or human actions such as voluntary charity work. This strategy can give greater benefits to those involved if those involved meet multiples times and can spot free loaders and punish them.
In the end, cooperation can work as the optimal strategy for selfish entities giving them greater benefits, especially in the long run, than purely selfish actions provided the entities interact multiple times over time and freeloaders or those that would explode the cooperation get pushed and thus, do not gain from the exportation.
Cooperation forms an important part of the socioeconomic design of EOS. We propose a network multiple autonomous entities working together and cooperating on common tasks. These entities remain separate with their own ideas doing their own things but as each works towards a common overall goal they can come together to form common project as needed. This results in entities forming higher entities to work on a common project in a holonic system of cooperation.