Abstract
Opinion Dynamics is an interdisciplinary area of research. Disciplines of Psychology and Sociology have proposed models of how individuals form opinions and how social interactions influence this process. Sociophysicists have interpreted the observed patterns in opinion formation in individuals as arising out of nonlinearity in the underlying process and helped shape the models. Agent‐based modelling has offered an excellent platform to study the Opinion Dynamics of large groups of interacting individuals. In this paper, we take recent models in opinion formation in individuals. We recast them to create a proper dynamical system and inject the idea of clock time into evolving individuals’ opinions. Thus, the time interval between two successive receipts of new information (i.e., the frequency of information receipts) by an individual becomes a factor that can be studied. In recent decades, social media has continuously shrunk time intervals between receipt of new information (i.e., increased frequency of information receipts). The recast models are used to show that as the time interval between successive receipts of new information gets shorter and the number of individuals in one’s network becomes larger, the propensity for polarization of an individual increases. This explains how social media could have caused polarisation. We use the word “polarisation” to mean an individual’s inability to hold a neutral opinion. A polarisation number based on sociological parameters is proposed. Critical values of the polarisation number beyond which an individual is prone to polarization are identified. These critical values depend on psychological parameters. The reduced time intervals between the receipt of new information and an increase in the size of groups that interact can push the polarisation number to approach and cross the critical value and could have played a crucial role in polarising individuals and social groups. We also define the extent of polarisation as the width of the region around neutral within which an individual is unable to have an opinion. Reported results are for values of model parameters found in the literature. Our findings offer an opportunity to adjust model parameters to align with empirical evidence. The models of opinion formation in individuals and the understanding arrived at in this study will help study Opinion Dynamics with all its nuances and details on large social networks using agent‐based modelling.