By Alastair Reed, Jack Springett-Gilling and Ellie Rogers
This post is a lightly edited excerpt from “Online P/CVE Interventions,” a chapter in the just-published Routledge Handbook of Online Violent Extremism.
Over the past decade and more, as terrorist and violent extremist actors have become increasingly adept at exploiting the internet for radicalisation and recruitment, policymakers and practitioners have grappled with how best to prevent and counter this exploitation.
Initially, online preventing and countering violent extremism (P/CVE) focused largely on two approaches: preventing the circulation of extremist propaganda and developing campaigns to challenge extremist messaging. Content moderation on mainstream platforms became increasingly effective at identifying and removing terrorist content, but new problems emerged. Extremist material was often simply displaced to less regulated online spaces. Moreover, beyond content produced by proscribed terrorist organisations, moderation becomes more complicated when decisions depend on what content says rather than who produced it. This creates a substantial grey area around so-called “lawful but awful” content: material that may contribute to pathways towards radicalisation but does not meet the threshold for removal.
Alongside content moderation, P/CVE practitioners sought to respond to terrorist and violent extremist propaganda through counter- and alternative-narrative campaigns. These initiatives have proliferated, generating a wide variety of approaches. Yet they have also been hampered by a lack of robust evaluation. Where evaluations do exist, evidence of effectiveness has often been limited.
But online PCVE has continued to develop. In a recent chapter “Online P/CVE interventions” for the Routledge Handbook of Online Violent Extremism, we examined two evolving approaches: digital media literacy and redirection. Both seek to address some of the gaps left by these earlier interventions.
Digital media literacy
Rather than trying to prevent individuals from encountering terrorist and violent extremist (TVE) content, or attempting to challenge extremist communications directly, digital media literacy (DML) focuses on strengthening the resilience of those encountering that content. DML encompasses a range of approaches designed to strengthen individuals’ ability to recognise and resist the manipulative techniques used in violent extremist propaganda. These can include direct taught instruction, digital games, and multiformat communications campaigns. The rationale is straightforward: by giving people the skills to critically evaluate the content they encounter online, they become better equipped to recognise manipulation and make their own judgements about what they see.
Evaluations of DML interventions piloted for P/CVE have already demonstrated promising effects on factors associated with resilience. But DML is a relatively new addition to the P/CVE toolkit, and there is a much longer track record of its application in other fields. A forthcoming meta-analysis by Springett-Gilling, drawing on 113 independent studies involving 113,865 participants across a broad range of contexts, found clear evidence that DML interventions can strengthen individuals’ ability to recognise and reject manipulative and misleading content.
The lesson for P/CVE is that we do not need to start from scratch. Evidence from specifically targeted P/CVE interventions remains relatively limited, but the much broader evidence base demonstrates the potential for DML to build public resilience against the complex mix of extremist propaganda, disinformation and other forms of manipulative content encountered online.
Redirection
Redirection takes a different approach. It uses targeted advertising on search engines and social media platforms to reach individuals searching for violent extremist content and offer them links to counter-speech or support services, which they can choose to click on or not.
In some respects, redirection represents an evolution of earlier counter- and alternative-narrative campaigns. However, it addresses one of their central challenges: how to reach individuals who may be at risk of radicalisation at the point at which they are actively seeking extremist material.
Later iterations have moved beyond simply redirecting users towards counter-speech. Instead, some programmes connect individuals with intervention partners offering services such as counselling, employment assistance, mental-health support and other forms of specialist help. This creates a potential pathway from an online intervention to sustained, individualised support that may help people disengage from violent extremist movements or address some of the underlying motivations that led them to seek out extremist content.
The impact of redirection is, however, much harder to establish. Many early campaigns were evaluated primarily using the metrics of digital advertising, such as impressions, click-through rates and engagement. These measures can demonstrate whether an intervention reached its intended audience and whether users engaged with the material, but they tell us much less about whether that engagement produced meaningful or lasting changes in attitudes or behaviour. Consequently, evidence that redirection can reach people searching for extremist material is considerably stronger than evidence that redirecting them towards counter-speech can reduce radicalisation.
Campaigns that connect individuals directly with support services offer a more promising pathway to measuring impact. Evaluations have documented cases in which individuals engaged with intervention providers and, in some instances, continued into longer-term support. Yet the numbers reaching this stage remain small, with substantial drop-off between being exposed to a redirection advertisement, clicking on it, contacting a service and ultimately engaging with support.
While redirection has demonstrated that carefully targeted online interventions can connect some individuals with assistance at a potentially important moment, the challenge now is to determine whether these individual successes can be translated into effective interventions at greater scale.
Where next for online PCVE?
While both approaches show considerable potential, important questions remain.
The first is one of scale: how can promising interventions be further developed and expanded without losing the qualities that make them effective?
The second question is one of continuity: how can promising but often short-lived and fluctuating interventions be sustained into longer-term programmes to enable comprehensive evaluation and iterative efforts to improve their effects?
The third is one of integration. There is no silver bullet for countering violent extremism. The challenge is therefore not simply to identify which individual interventions work, but to understand how different approaches can be combined so that they complement and reinforce one another. Digital media literacy, redirection, content moderation, counter-messaging and offline interventions should not necessarily be viewed as competing alternatives, but as potential components of a broader P/CVE framework.
The final question is about what comes next. The online environment is changing rapidly, as are the ways in which violent extremists exploit it. This inevitably creates new gaps in our response. Identifying those gaps, and developing interventions capable of addressing them, should be a central priority for the next generation of online PCVE.
The future of online PCVE may therefore depend less on finding the next single solution than on building a more sophisticated ecosystem of interventions: one that limits the circulation of the most harmful content, strengthens people’s resilience for when they encounter misleading and manipulative content, and targets interventions at those on a pathway to radicalisation or seeking a way out. Such an ecosystem could hit efforts to radicalise and recruit online from all sides. The task going forward will be to realise and sustain it.
Professor Alastair Reed is Professor of Security and Strategic Communication at the Centre for Peace and Security, Coventry University.
Jack Springett-Gilling is a PhD candidate at Swansea University within the Cyber Threats Research Centre (CYTREC).
Ellie Rogers is a PhD candidate at Swansea University within the Cyber Threats Research Centre (CYTREC). Her research focuses on the algorithmic amplification of counter-speech as a response to online extremism.