The Influence of Media Systems on Polarisation Patterns

The next speaker in this ICA 2024 conference session is Harry Yan, whose begins by noting the increase of animosity and affective polarisation against opposing parties in the United States. What role do mass media play in this context? We already know that greater Internet use in itself is not to blame here: this has been shown by a range of studies already. More complex explanations need to be found.

The present study is interested in understanding active and interactive audiences through agent-based modelling – in doing so, this also moves beyond assumptions of passive mass media audiences in past media effects theory. Two hypotheses can be tested here: more media options might lead to a more fragmented audience and public opinion space; while a more centralised media system might lead to more homogenised audiences and public opinions.

In agent-based modelling, this might be measured through entropy (showing opinion diversity, on a continuous scale), and the number of opinion communities (showing fragmented audiences, as a discrete distribution). The number of media outlets and the total reach of each media outlet can also be varied in such modelling. Active audience members are modelled as seeking out opinion-compatible messages and selecting media on this basis.

The more media options are introduced into such a system, the more opinion diversity emerges. However, the dominant pattern remains homogenisation: bipolar polarisation into two distinct communities emerges only with a system that contains very limited distinct media options.