Coevolutionary spatial game theory: The impact of abstention and dynamic networks on the evolution of cooperation
MetadataShow full item record
This item's downloads: 253 (view details)
Since the dawn of evolutionary game theory, the standard models in evolutionary dynamics have been developed under the assumption of a well-mixed population where agents (individuals) can interact with all the other agents in the population (i.e., unstructured population). However, in real-world scenarios, populations are usually not well-mixed nor unstructured. In this manner, concepts of graph theory to consider structured populations in a network form have been extensively applied to evolutionary game theory, giving birth to the so-called spatial evolutionary game theory domain. Although there has been a lot of research undertaken into spatial evolutionary game theory over the past number of years, the majority of this research involves the use of static networks in compulsory games such as the prisoner’s dilemma game. Recent studies have considered coevolutionary spatial models adopting dynamic networks where both the game strategies and the network itself are subject to evolution, which, in fact, constitute a natural upgrade in the field for being more accurate in describing many social scenarios. For instance, in scenarios involving real biological networks and social networks, the number of individuals, their connections and the environment are often dynamic. Those studies have been successful in exploring the effects of coevolution in a compulsory game, where coevolution has been recognized as a key mechanism to support cooperative behaviour. However, there has been little success in accounting for the impact of abstention (voluntary/optional participation) in such coevolutionary models. Thus, within the bounds of multi-agent systems and network science, this thesis aimed to expand the knowledge of evolutionary game theory by bridging the gap between abstention and dynamic networks in social dilemmas. Besides investigating the impact of dynamic networks on the evolution of cooperation on both compulsory and voluntary games, this work also provided a novel perspective for understanding the foundations of cyclic dominance behaviour in the context of the voluntary games.