Fast Consensus and Metastability in a Highly Polarized Social Network

TL;DR

This paper models a highly polarized social network using interacting point processes, demonstrating that the network rapidly reaches a metastable consensus as polarization increases.

Contribution

It introduces a novel stochastic model for polarized social networks and proves the emergence of metastable consensus under high polarization conditions.

Findings

Network reaches instantaneous metastable consensus at high polarization

Model captures influence of social pressure on opinion dynamics

Proves convergence properties of the proposed process

Abstract

A polarized social network is modeled as a system of interacting marked point processes with memory of variable length. Each point process indicates the successive times in which a social actor expresses a "favorable" or "contrary" opinion. After expressing an opinion, the social pressure on the actor is reset to 0, waiting for the group's reaction. The orientation and the rate at which an actor expresses an opinion is influenced by the social pressure exerted on it, modulated by a polarization coefficient. We prove that the network reaches an instantaneous but metastable consensus, when the polarization coefficient diverges.