Dynamic analysis of false information spread over social media : 5G-covid 19 conspiracy theory

Abstract

The spread of false information via online social networks is a critical societal issue with various potential harms. Although there are huge efforts both in research and application to mitigate this problem, it persists with an increasing magnitude of results ranging from political manipulation to violent attacks. In our research, we built a causal simulation model to combine the existing accumulated knowledge in the liter ature and provide a formal model to evaluate the governing dynamics for the specific case of the viral spread of the 5G-COVID-19 conspiracy theory. The model makes use of both qualitative and quantitative data and successfully generates the observed dynamics for the 5g narrative. Results from the base run suggest that the dominance of believers in the active discussion on social media is overrepresented relative to the total population. Moreover, common mitigation strategies proposed in the literature such as limiting the interaction with believers of the misinformation often seem to pro duce worse outcomes for specific cases which indicates policy resistance. In addition, scenario analysis suggests that the involvement of neutral people in sharing misinforma tion or superspreader actors might be enough to induce the system to pass the tipping point and generate an infodemic. The current analysis presents several trade-offs while discussing the underlying reasons through posterior analysis. In further research, we plan to expand our analysis by the inclusion of other user profiles, experiment with other mitigation strategies, and discuss the potential similarities and differences of our case with other types of false information dynamics.