Modeling misinformation spread for policy evaluation: a parsimonious framework

We develop a parsimonious framework for evaluating the efficacy of different approaches for limiting the spread of misinformation. We use this framework and simulation studies to determine the evolution of truthful and fake messages on social media platforms and then investigate the following policy interventions: (1) our suggested approach of having the platform require senders of messages to also state their perceived (possibly incorrect) veracity of the message, (2) provide some accuracy nudge to increase the number of potential readers who can accurately identify fake messages, (3) have the platform flag fake messages, and (4) have the platform demote or down-rank fake messages. We find that when a significant number of senders are able to correctly identify the veracity of the message, the market can self-regulate under our suggested approach. If this is not the case, we find that augmenting our approach with any of the other approaches is effective in reducing the spread of misinformation.