Beyond consensus and polarisation: complex social phenomena in social networks

Speaker: Emeritus Prof. Brian D. O. Anderson

Social network analysis is a rich and exciting area of interdisciplinary research that has been tackled by many different scientific communities. Much research draws on ideas of control and systems theory to infer how the opinions of a group of interacting individuals evolve, and to explain the resulting behaviour, often observed in experiments of sociologists, psychologists and the like, in terms of system dynamics concepts.

This lecture will survey several distinct recent developments of this character. We shall present an opinion dynamics model which describes how an individual’s private and expressed opinions (which are not the same in general) evolve under pressure to conform to the majority opinion. In another direction we shall present new results on the recently proposed DeGroot-Friedkin model, which describes how an individual’s self-confidence (termed social power) in his/her own opinion evolves over discussion of a sequence of topics. One key finding is that every individual forgets his/her perceived (i.e. initial) social power exponentially fast, even when the network topology is dynamic. Lastly, we shall describe the opinion dynamics of interacting individuals holding multiple and logically dependent opinions on a number of issues.