Thinking about Predictors of Radicalization to Violence
Exposure to Online Extremism - National Findings
Given the proliferation of social media platforms and myriad other online avenues, the internet is increasingly considered as an easily accessible platform for radicalization. The volume of extremist content online has grown exponentially in recent years. Based on an analysis of the Profiles of Individual Radicalization in the US (PIRUS) dataset, the internet played either a secondary or primary role in the radicalization of over 90% of US extremists over the last 10 years. Most extremists were active on social media platforms, using them for a vast array of activities including disseminating content, participating in extremist dialogue and facilitating domestic terror plots. Similarly, a 2023 report from the National Threat Assessment Center of the United States Secret Service revealed that most attackers in the US from 2016-2020 had an identified presence on the internet.
We fielded nationally representative surveys via YouGov in May 2023 (n=2000) where respondents were asked a range of questions related to targeted violence and domestic terrorism. Nearly 50% of respondents thought that radicalization was more likely to occur online, compared to face-to-face settings. Out of those identifying the internet as the most likely source of radicalization, over 65% thought that such radicalization was most likely to occur on social media platforms.
Survey responses indicated that a significant percentage of individuals were frequently exposed to a variety of extremist content online. These included exposure to conspiracy theories, far-right ideologies, far-left ideologies and hate speech. On average, nearly 20% of respondents reported having been exposed to such content regularly or very often. In addition, another 30% or so were sometimes exposed to such extremism. Respondents also reported being exposed to violent extremist material including content explicitly praising previous mass attacks, calls for threats to specific individuals, calls for violence against elected officials and calls for violence against certain groups of people. On average, slightly over 10% of respondents regularly or very often saw such content with another 20% sometimes exposed. Troublingly, just over half of those surveyed have seen content praising previous mass attacks. Overall, the data reveals that a non-trivial percentage of respondents are exposed to vitriolic material online on a regular basis. While some seek this content out, others have it recommended to them via algorithm or shared with them by someone in their social network. Given the prolificity of such content and the sheer number of people having easy access to extremist material, it is imperative that policymakers, technological companies, academics, activists and all concerned come together and redouble their efforts to staunch the flow of violent extremist content and work to mitigate its impact.
Exploring the link between exposure to online extremism and support for political violence
Scholars have found that increased exposure to extremist content online is a powerful predictor of support for political violence, even when controlling for alternative explanations that account for such phenomena (Farag 2017; Hasan et al 2018; Reeve 2021). Findings from our surveys are strongly indicative of this assertion. For example, those frequently exposed to hate speech are more than twice as likely to support threats against elected officials and support violence to achieve ideological and partisan objectives in comparison to those with no exposure to hate speech. Similar findings emerge for exposure to other kinds of extremist material, as seen in the graphs below.
Exploring the link between exposure to violent online extremism and support for political violence
Frequent exposure to unambiguously violent extremist content online also increases respondents’ support for threats and violence to achieve political objectives. Often, the effect of this relationship is stronger than observed with measures such as hate speech or conspiracy theories. As a case in point, respondents frequently viewing content praising previous mass attacks are four to five times more likely to agree that the use of threats and violence is justifiable to achieve various political objectives. Likewise, individuals who frequently see online content threatening elected officials are several times more likely to offer support for political violence.
Given unprecedented levels of political polarization and the pervasive presence of social media echo chambers, it is also highly plausible that individuals seeing violent content online seldom get exposed to alternative viewpoints or perspectives. In such a context, repeatedly being served a staple diet of divisive and virulent rhetoric can reinforce respondents’ opinions, which in turn can boost their support to achieve a variety of ideological and political objectives through violence.
Survey data identify a clear need for improved media literacy skills
Survey data reveals that only one third of respondents found it easy or extremely easy to determine whether information they encounter online is trustworthy. Just over 40% find it easy/extremely easy to distinguish facts from opinions and recognize when information is used out of context. Somewhat higher percentages are able to easily ascertain if information is slanted politically or recognize when someone is trying to manipulate them. However, across the five measures, between 10 to 40% of respondents have a difficult or extremely difficult time with media literacy.
Survey respondents favored addressing radicalization in various places including online and in schools. While the redirect method, which exposes those seeking out extremist content to counter messaging, and prebunking, which teaches people to spot disinformation and develop resilience against disinformation before it is encountered, offer some promise at disrupting or preventing radicalization to violence (Helmus and Klein 2018; Roozenbeek, Van der Linden, Goldberg, Rathke, and Lewanddowsky 2022), programs focused on building individuals’ digital media literacy skills can also boost resilience to violent extremist content online. To develop resilience to online extremism and radicalization to violence, it is crucial to strengthen digital media literacy skills.
The research presented here is part of a larger project funded by the Department of Homeland Security’s Center for Prevention Programs and Partnerships (opportunity number DHS-22-TTP-132-00-01). Analysis and opinions are those of the authors.
