A physicist's approach to social dilemmas on networks: we've been wrong all along

One proposal to explain the origin of cooperative behavior that was put forward 20 years ago by Nowak and May posits that cooperation can prevail over free riders if interactions take place in a network instead of being universal. This idea has inspired very many researchers to explore this possibility and literally hundreds of papers have been published about it. At the beginning, most results tend to confirm the hypothesis that networks helped cooperation, but lately it has become clear that this alleged support is very much dependent on the details of the model studied. In order to sort out the ensuing confusion in the field, we realized that we have to supersede Aristotle and proceed to the traditional approach in physics since Galileo, namely to do experiments and to compare them with theoretical descriptions. As I will show, this has allowed to make a breakthrough in the field, pointing to the preliminary but likely conclusion that networks are irrelevant to cooperation, at least with humans, and at least in the context of a specific model, the Prisoner's Dilemma. In the way to this conclusion we have also gathered new knowledge on the way human people behave when interacting with each other in this context, that can be summarized by saying that humans are equal in reciprocity, and diverse in generosity. These and related aspects will also be discussed in the talk.