A Game-Theoretic Model of Misinformation Spread on Social Networks

Abstract

We develop a game-theoretic model of  sharing decisions among users of  a social network. Agents are endowed with a subjective prior on an unobservable state, representing their beliefs on a topic that is subject to vote. Agents cast their binary vote, matching the sign of the state. Before the voting stage, a small fraction of agents receives a piece of news which impacts their belief. Those who receive the news update their belief and  make a decision as to whether to share the news with their followers in order to influence their beliefs, and in turn their votes, given a fixed cost for sharing. We characterize  the underlying news spread as an endogenous Susceptible-Infected (SI) epidemic process and derive agents'  sharing decisions as well as the size of the sharing  cascade at the equilibrium of the game showing that  lower credibility news can result in a larger cascade than truth.

Keywords: Persuasion, strategic news sharing, spread of information, social networks