Rumour Propagation on Social Networks as a Function of Diversity

This paper investigates how the diversity of the population affects the propagation of a rumour. Mathematical models of rumour have only recently begun to consider the social network by which the interlocutors are connected. If the outgroup minority is very poorly integrated do they have a greater probability of resisting a derogatory rumor or is it better for the minority to be well mixed in the social network as a whole, allowing them to better rebut the rumour? In populations comprised of two conflicting groups the propagation of an outgroup negative rumour is dependent on the proportion and distribution of the two groups in the social network. The social network over which the rumour propagates is comprised of individuals from two social groups; the ingroup majority and the outgroup minority, and the relationships between those individuals. Because the rumour content reflects negatively on the members of the outgroup, the rumour’s flow across the minority outgroup population is markedly different from the flow across the majority ingroup. Thus rumor propagation over such a social network is clearly dependent on the size of the minority outgroup population; but the integration of the outgroup minority is also a factor in rumour propagation. Minority outgroups of similar size can consist of a tight knit clique or have their members evenly distributed across the social network. Consider the two social networks below. In the two 10 people social networks shown there are 11 edges are 4 outgroup nodes (shaded). The social network on the left (Network A) appears to be more integrated than that on the right (Network B). That is, the minority group of the 4 shaded nodes is more mixed into the network; relationships are less based on group status. It is not enough to simply state that 40% of the people in both social networks are minorities; we need metrics that quantify the integration of the minority into the social network as a whole. The diversity of a social network depends on both the size of the minority population and its distribution. The diversity is quantified by the Minority Integration Metric (MIM). The affect of that diversity on the propagation of an outgroup negative rumour is investigated using Monte Carlo simulations over various network topologies. The GBN-Dialogue model is used to simulate the rumour propagation on social networks of various topologies. Based on rumour theory, the GBN-Dialogue model focuses on characteristics of person to person interaction, and is dynamic. The GBN-Dialogue model calculates the probability of rumour transmission between two interlocutors based on their strength of belief and level of perceived novelty and their group memberships. Psychological and sociological research informs this transmission probability function. Belief and perceived novelty levels of each participant adjust as a result of rumour transmission and time; research literature on attitude change guide these updating functions.