Opinion trajectories of three representative simulation runs with 100 agents generated by the Durkheimian model.

<p>In all three runs, the opinions are restricted to values between −250 and 250, and all agents hold the same opinion initially ( for all ). In all runs, we assume the same social influence range , but vary the strength of the disintegrating force. (A) Monoculture, resulting in the case of a weak disintegrating force (). Agents do not hold perfectly identical opinions, but the variance is low. We studied dynamics over 10.000 iterations (Each iteration consists of simulation events .). (B) Anomie (i.e. extreme individualism), generated by a very strong disintegrating force (). Agents spread over the complete opinion scale. The black line represents the time-dependent opinion of a single, randomly picked agent, showing significant opinion changes over time, which is in contrast to the collective opinion formation dynamics found in the monocultural and pluralistic cases (A) and (B). (C) For a moderate disintegrating force (), the population quickly disintegrates into clusters. As long as these clusters are small, they are metastable. However, clusters perform random walks and can merge (e.g. around iteration 5500). As the disintegrating force grows with the size of a cluster, big clusters eventually split up into subclusters (e.g. around iteration 7000). The additional graph, in which each agent's opinion trajectory is represented by a solid black line, is an alternative visualization of the simulation run with . It shows that the composition of clusters persists over long time periods.