An Active Opinion Dynamics Model: the Gap Between the Voting Result and Group Opinion

Yiru Jiao,Yongli Li

DOI: https://doi.org/10.1016/j.inffus.2020.08.009

IF: 18.6

2021-01-01

Information Fusion

Abstract:Originally developed to simulate the evolution of public opinion, opinion dynamics models have also been successfully applied to market pricing and advertising. However, passive interactions initiated by locational or social relationships in these models are insufficient to characterize purposeful behaviours such as canvassing or trading, where people are driven by their specific intrinsic motivations. Here, we propose an active model in which people tend to communicate with someone who is more likely to be an ally and game theoretically decide whether to interact. Model simulations highlight the macroscopic development of opinion evolution, showing the ubiquitous gap between people's voting result and their collective opinion, and how it narrows with the stabilization of opinion evolution. Our results help explain why group opinion rarely reverses its initial stance and the significance of a level of inclusiveness that is neither too high nor too low. Additionally, we find and attest to the probability distribution of group opinion change, which contributes to predicting how much the collective opinion of a group will change after full discussion.