Association in Multi-agent Simulations of Dynamic Random Social Networks

Social groups form where individuals who are attracted to each other - usually by a common interest – interact and form clusters. These groups exist within structural networks that rely on the patterns of links between members through which communication and resource transfer occurs. Individual influence impacts on emergent characteristics of a group, for example, global opinion and collective behaviour. However, individuals join and leave groups, thus changing the system’s dynamics. What impact do these structural changes have on the emergence of sub-groups? Here our interest is in the association of members around a particular ideology and real social network systems provide our bio-inspired simulation models. We address the effects of dynamic structural changes to randomly connected networks on global behaviour and the emergence of subgroups that associate with specific states. Results from multi-agent simulations demonstrate that social cohesion and collection of nodes around particular states are dependent on group dynamics and can have an impact on social management that effects social order and stability.