Opinion Dynamics and Social Power Evolution: A Single-Timescale Model

TL;DR

This paper introduces a single-timescale modification of the DeGroot-Friedkin model to analyze opinion dynamics and social power evolution, revealing similar behavior in irreducible networks but distinct outcomes in reducible networks.

Contribution

It proposes and analyzes a new single-timescale model for opinion and social power evolution, extending the original DF model to reducible influence networks with novel equilibrium behaviors.

Findings

In irreducible networks, the model predicts power distribution aligns with centrality.

In reducible networks, individuals can retain all social power if initially set so.

Multiple sinks in the network lead to diverse equilibrium power partitions.

Abstract

This paper studies the evolution of self-appraisal and social power, for a group of individuals who discuss and form opinions. We consider a modification of the recently proposed DeGroot-Friedkin (DF) model, in which the opinion formation process takes place on the same timescale as the reflected appraisal process; we call this new model the single-timescale DF model. We provide a comprehensive analysis of the equilibria and convergence properties of the model for the settings of irreducible and reducible influence networks. For the setting of irreducible influence networks, the single-timescale DF model has the same behavior as the original DF model, that is, it predicts among other things that the social power ranking among individuals is asymptotically equal to their centrality ranking, that social power tends to accumulate at the top of the centrality ranking hierarchy, and that an autocratic (resp., democratic) power structure arises when the centrality scores are maximally nonuniform (resp., uniform). For the setting of reducible influence networks, the single-timescale DF model behaves differently from the original DF model in two ways. First, an individual, who corresponds to a reducible node in a reducible influence network, can keep all social power in the single-timescale DF model if the initial condition does so, whereas its social power asymptotically vanishes in the original DF model. Second, when the associated network has multiple sinks, the two models behave very differently: the original DF model has a single globally-attractive equilibrium, whereas any partition of social power among the sinks is allowable at equilibrium in the single-timescale DF model.