Large-Scale Conflicts in Massively Multiplayer Online Games

Complex systems are of interest to the scientific community due to their ubiquity and diversity in daily life. Popularity notwithstanding, the analysis of complex systems remains a difficult task, due to the problems in capturing high-volume data. Massively Multiplayer Online Games (MMOGs) have recently emerged as a tractable way to analyze complex system interactions, because these virtual environments are able to capture a great amount of data and at high-fidelity, often tracking the actions of many individuals at a time resolution of seconds. MMOGs have been used to study phenomena such as social networks and financial systems; our focus is to identify behaviors related to Large-Scale Conflict (LSC). In this paper, we review how one particular MMOG allows large-scale complex behavior to emerge and we draw parallels between virtual-world LSCs and real-world LSCs. The LSC-related behavior that we are interested in identifying deals with the conditions that lead a participant in the virtual environment (a game player) to engage and actively participate in a LSC, with the goal of informing an agent-based model that predicts when any one player is likely to engage in conflict. We identify virtual world behavioral analogues to real-world behavior of interest (i.e. insurgent behavior), and link the virtual behavior to a (previously derived) theoretical framework that analyzes the determinants of participation in the civil war of Sierra Leone. This framework identifies three general theories that collectively predict participation in civil war (a type of LSC); we operationalize one of the theories (Theory of Social Sanctions), and look at how insurgent behavior can occur as a function of community networks, which are assumed to impose social sanctions for non-participation in an LSC.