The geographical propagation of attitudes on social media

In real life interactions, emotional states have been shown to transfer to others through a process called emotional contagion. An increasing body of research suggests that this phenomenon also occurs on social media. Moreover, computerised interactions have been found to increase and facilitate group polarisation and radicalisation of attitudes. The present study seeks to identify the extent to which contagion effects are influenced by a circulation of extreme views on social media. Specifically, it is hypothesised that attitudes around Brexit would disperse not only as a function of posts' density but also depending on how extreme are the posts that are being circulated on social media. Moreover, extremity is hypothesised to moderate the relationship between posts density and social activism (e.g. protests). These hypotheses will be addressed using geo-tagged Twitter posts and sentiment analysis to identify the extremity of the views that have been circulated after the Brexit vote. Activism will be operationalised as the number of protests identified using news media open databases such as the Global Database of Events, Language and Tone. These results would identify the extent to which attitudes as presented in the social media are linked to behaviour and would help identify how social media incentivises to action compared to real life interactions.