Social Polarization Caused by the Estimation of Others' Opinions

Abstract

With the development of social media in recent years, individuals are now able to broadcast their opinions to the world, and the opportunities to come into contact with diverse opinions have increased. Whereas, polarization tends to increase on social media due to the echo chamber, and the social polarization has become a problem. To design a social system that suppresses polarization, it is essential to understand the mechanisms that cause it. The purpose of this study is to clarify the effect of attitude estimation on polarization and its process under the assumption that opinion is expressed based on the estimation of others' attitudes. We construct a model in which agents update their beliefs based on rewards and punishments for expressing opinions, and analyzed the time series of opinion expressing by multi-agent simulation. The results suggest that misestimation of the surrounding attitudes causes bias in the opinions expressed, resulting in polarization.