The Role of Information in the Evolution and Stability of Cooperation in Stochastic Environments

The aim of this thesis is to study the emergence and stability of cooperation in fluctuating environments, in presence of varying levels of available information about the current state of the environment. Cooperation's intrinsic susceptibility to individual deviant behaviours makes its stability a central aspect of this collective behaviour. We consider a model composed of two agents whose resources undergo stochastic multiplicative growth. These growth processes are influenced by the state of a shared fluctuating environment while being guided by the agents' individual strategies of investment in the environment's states. To improve their growth, the agents are allowed to interact through two mechanisms: partitioning the total available information about the fluctuating environment among them, and sharing fractions of their resources as a common good. We analyse the interplay between these mechanisms at varying levels of total information about the environment, thereby exploring the link between uncertainty and cooperation. We show that full cooperation, achieved through complete resource-sharing and equal information partition, is stable and evolvable, provided the initial state exhibits a sufficient level of cooperation. We also find that the size of the basin of attraction of the full cooperation state is determined by the predictability of the environment.