Combining interventions to reduce the spread of viral misinformation

Abstract

Misinformation online poses a range of threats, from subverting democratic processes to undermining public health measures. Proposed solutions range from encouraging more selective sharing by individuals to removing false content and accounts that create or promote it. Here we provide a framework to evaluate interventions aimed at reducing viral misinformation online both in isolation and when used in combination. We begin by deriving a generative model of viral misinformation spread, inspired by research on infectious disease. By applying this model to a large corpus (10.5 million tweets) of misinformation events that occurred during the 2020 US election, we reveal that commonly proposed interventions are unlikely to be effective in isolation. However, our framework demonstrates that a combined approach can achieve a substantial reduction in the prevalence of misinformation. Our results highlight a practical path forward as misinformation online continues to threaten vaccination efforts, equity and democratic processes around the globe.

Identification and characterization of misinformation superspreaders on social media

“…However, the potential for suspensions to reduce harm may conflict with freedom of speech values [23]. The effectiveness of other approaches to moderation should be evaluated by researchers and industry practitioners [2]. For instance, platforms could be redesigned to incentivize the sharing of trustworthy content [3].…”

“…Tweets are gathered from a historical collection based on Twitter's Decahose Application Programming Interface (API). 2 The Decahose provides a 10% sample of all public tweets. We collect tweets over a ten-month period (Jan. 2020 -Oct. 2020).…”

Wild animals suppress the spread of socially-transmitted misinformation

“…We designed an individual-based model to simulate the dynamics of populations comprising individuals who follow a response rescaling rule interacting on a featureless 2-dimensional plane. We compared the results of these simulations to expectations from two widely-used models of social contagion: simple and fractional contagion [7, 8, 25, 38, 41]. Model parameters, including those that control agent movement, sensing, and decision-making were set to their corresponding empirical estimates (Table S2).…”

“…E. Observed responses (colored points, colors as in (A) center; points vertically jittered), empirical fraction responding (black points) and predicted (black line) response probability from response rescaling model. Note on (A): we were unable to fit standard phenomenological formulations of simple [7, 8] and fractional contagion [8, 38] models to this data set because, under both models, probability to respond when no neighbor has yet responded is zero; thus these models cannot predict onset of escape events. Nevertheless, we analyze predicted spreading properties of these models following cascade onset in Fig 4.…”

“…Understanding the mechanisms by which information and misinformation spread through groups of individual actors is essential to the prediction of phenomena ranging from coordinated group behaviours [1][2][3] to global misinformation epidemics [4][5][6][7]. Transmission of information through groups depends on the decision-making strategies individuals use to transform the perceived actions of others into their own behavioural actions [8][9][10].…”