City Research Online

Abstract

The self-organisation of social behaviour is observed across populations of varying complexity. Evolutionary game theory models such systems by describing interactions between individuals as evolutionary games. This thesis leverages a wide range of tools from evolutionary game theory to develop models of population dynamics and advance our understanding of the evolution of social behaviour, spanning from simple to complex social interactions and their applications.

We start by analysing fixation processes on finite, well-mixed populations. We show that new strategies have higher probability of fixation in larger populations
for half of all pairwise games, including the widely studied Prisoner’s Dilemma, Hawk-Dove and Stag Hunt games.

Next, we consider multiplayer social dilemmas on networks to examine the evolution of cooperation under various assumptions about population structure and
individual mobility. We find that limited movement leading to community organisation strongly promotes the evolution of cooperation in public goods dilemmas. In particular, large networks of small communities prove highly effective. Comparisons with completely mixed populations show that increased community mixing often hampers cooperation. We also observe the robust co-evolution of cooperation and high-mobility strategies under conditional movement. In regular networks, co-operators are able to find each other while evading defectors for extended periods.

These two mechanisms for the evolution of cooperation differ fundamentally: community structure relies on the viscosity of the evolutionary process on the network, whereas conditional movement depends on the evolution of mobile assortative behaviour. This distinction is supported by a detailed analysis of six different evolutionary dynamics.

Further, we develop two new dynamic models for infinite populations, adapting evolutionary game theory concepts to study other systems. The first examines the Win-Stay, Lose-Shift strategy in common goods usage. We propose its implementation by mobile users accessing Internet services, supported by its good performance in realistic stochastic simulations. This theory extends to the distribution over grazing and foraging land and may be used to propose solutions to operators of public transport or alternative technological common goods. The second model explores productive interaction and the transmission-selection of skills. Advances in economic complexity and evolutionary economic geography show us the impact of a skilled workforce on industrial development and its geographic organisation. We use this as motivation to study the evolution and co-existence skills and their synergistic production within social systems.

Keywords: evolutionary game theory, social dilemmas, cooperation, population structure, movement

Publication Type:	Thesis (Doctoral)
Subjects:	Q Science > QA Mathematics
Departments:	School of Science & Technology > Mathematics School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads