On 2 October 2025, all eyes turned to the shocking events unfolding at a synagogue in Manchester, United Kingdom. The attack at Heaton Park Hebrew Congregation claimed the lives of two Jewish congregants after the assailant drove a car into worshippers and proceeded to stab them. Footage, images and commentary of the Heaton Park Synagogue terrorist attack began filtering down from news sources to social media. Within a few hours, vast quantities of mis/disinformation, conspiracy theories and violent reactionary content exploded across the online ecosystem.
The ‘Attention Window’, in this context, refers to the initial 24-48 hours following an event. A spike in engagement on social media sites characterises the point at which public attention is largely hooked onto one focal point. This Insight will analyse the cross-platform, online responses in the hours that followed the Manchester Heaton Park Synagogue Attack. Using snowball sampling and media monitoring to map online reactions to the attack, it compares discourse within and across four platforms. This analysis is solely focused on harmful reactionary content and stipulates that it should not be taken as a general characterisation of content found on the respective social media platforms. It will discuss content seen on two fringe and two mainstream platforms: Quora, Gab, TikTok, and X. Each platform was monitored using consistent search terms such as ‘Heaton’, ‘Manchester’, and ‘Synagogue’, and operated under new, blank accounts to reduce any algorithmic bias of content found.
All platforms saw a surge of content posted on the attack, alongside the identifiable commonalities in rhetoric and reaction patterns – primarily characterised by framing the attack narrative through pre-existing social fault-lines, giving weight to extremist worldviews. This Insight will explore individual platform dynamics, highlight broader cross-platform patterns during this volatile period immediately following the attack, and discuss the importance of positive interventions within this crucial ‘Attention Window’ towards the goal of preventing further extremist radicalisation.
The Attention Window: Platform-Specific Responses
Within the crucial hours following the Heaton Park Attack, this primary analysis explores (1) the nature of issue-mobilisation, (2) the presence of conspiratorial narratives or mis/disinformation, (3) the level of racist content or hate-speech and (4) the prevalence of violent sentiment or retaliatory threats. All platforms see the presence of these themes to varying degrees. However, before characterising the commonality of the attention window, platform ecosystems must be individually understood by discursively analysing the different narrative types of reactive content.
TikTok
It is important to note that, in the case of TikTok, this depth of research was not characteristic of a typical user experience, and moderation removed the bulk of such content in the following 24 hours. Nonetheless, initial harmful responses to the Heaton Park attack on TikTok can be characterised by considerable amounts of racism and anti-migrant discourse, but lower levels of violent content than on other platforms. Many videos that gained attention in this initial period had a few key markers: emotional appeals, alongside far-right dog whistles and Islamophobic hate speech, as shown in Figs. 1 and 2. Before any information was released about the attacker’s identity or motive, there were many false reports circulating on TikTok. As the escalating comments in Fig. 2 demonstrate, this stoked fear: “we are being invaded by immigrants”, and “nobody is safe anymore” as “we are always under attack”. Fear prompts outrage, and discussions of how to fight back, which map onto and confirm pre-existing anti-migrant narratives. This same type of issue-mobilisation can be seen across all four platforms to varying degrees.
Figure 1: Screenshot of a TikTok video posted on 02/10/25 at 13:25.
Figure 2: Screenshot of selected few comments. Taken 02/10/25 at 13:58.
Quora
Immediate reactions on Quora reflected pervasive conspiratorial discourse across the political spectrum, with discussions shifting away from immigration in the United Kingdom toward debates concerning Israel and Islamophobic rhetoric in the United States.
As seen in Fig. 3, a small minority pushed the conspiratorial narrative that the terror incident was a ‘false flag’ attack carried out by Israel itself. However, a larger Islamophobic conspiratorial narrative also emerged, suggesting this attack was evidence of a wider ‘Islamic’ plot to “attack and kill… Jews, Christians, Israel and their allies” [sic], with violent depictions of what should be done in Gaza in retaliation, as displayed in Fig. 4.
Fig. 3: Screenshot of Quora comments. Taken at 02/10/25 at 13:51.
Fig. 4: Screenshot of Quora, Question and Answer. Taken 02/10/25 at 13:54.
Quora’s high proportion of users in the United States meant reactions to the Manchester attack were mapped onto pre-existing American social cleavages. Violent depictions drew comparisons to the assassination of Charlie Kirk and the conspiratorial threat of the ‘radical-left’ in the United States, as shown in Fig. 5. In Figs. 5-9, the effects of the narratives produced in the attention window are stark, with the resulting dialogue from the following days suggesting continued belief that violence is necessary to combat the ‘radical left’. Such posts continued irrespective of the now-released facts indicating that the attack had no such comparative motivations.
Fig. 5: Screenshot of excerpt from longer Quora post from 07/10/25 at 04:25.
Fig. 6-9: Screenshot of selected few comments. Taken 08/10/25 at 09:37.
Gab
The ecosystem on Gab similarly mapped the attack onto pre-existing conspiratorial narratives, with higher proportions of antisemitism compared to the other platforms.
Several of Gab’s users reported the attack as the latest in a long line of ‘hoax’ terror incidents in the United Kingdom and abroad. Antisemitic accounts promoted content suggesting that the attack was fake, “low-rent theatre”, with one user stating, “nothing sells a staged attack like a ‘blood-stained’ rabbinic prop” in Fig. 10. The platform, which hosts conspiracy theory minded communities, centred around hoax-identification, flooded the platform with graphic content of this and previous attacks, such as the “2017 Westminster Bridge Attack Hoax”, as seen in Figs. 10 and 11.
Gab also saw high prevalence of Islamophobic hate speech and similar calls for violence against immigrants in the United Kingdom as a key area of issue-mobilisation. However, the anti-Jewish rhetoric was more significant than any other platform investigated. Posts on other platforms saw conspiratorial narratives regarding a ‘false flag’ operation and attributed this only to Israel, making the distinction between the country and the religion, whilst Gab users targeted Jewish communities themselves, as Figs. 10-11 show.
Fig. 10 – 11: Screenshot excerpts of a longer post on Gab. Taken at 03/10/25 at 11:50.
X
Comparably to TikTok, immediate reactions on X saw a high frequency of anti-immigrant and Islamophobic hate speech dominating mainstream discourse. The rhetoric was more political, with much of the discussion centred on anger at Keir Starmer and the Labour government, alongside support for Nigel Farage and Reform – as evident in Figs. 12 – 15. Reactions to the attack were mapped onto politicised discourse surrounding the respective policies of these parties and the recent far-right anti-migrant push. Users suggested the attack was the fault of Starmer’s “lax approach to immigration and real crime whilst stupidly attacking legitimate free speech” in Fig. 14. The attention window on X can be characterised by a polarising use of the attack to reinforce a pre-existing, misleading conflation of ‘increasing violence’ in the United Kingdom with immigration (see Figs. 14-15).
Antisemitic content on X was significantly lower than observed on Quora and Gab. As displayed in Figs. 13 – 15, anti-immigrant and Islamophobic hate speech was, however, highly prevalent, as well as overtly racist content that included references to Muslims as “violent beasts” and Islam as a “death cult”.
Figs. 12: Screenshot of X post from 02/10/25 at 18:42
Figs. 13: Screenshot of X post from 03/10/25 at 16:16
Figs. 14: Screenshot of X post from 03/10/25 at 14:54
Figs. 15: Screenshot of X post from 02/10/25 at 17:25
The Attention Window: Cross-Platform Patterns
The analysis of the four platforms explored demonstrates platform-specific nuances in rhetoric, as well as cross-platform commonalities in the nature of responses to the Manchester attack within the crucial attention window. Chief among these are how responses to the attack are mapped onto pre-existing beliefs and narratives, exploiting and striking at the heart of social fault lines, used to evidence a range of extremist – and sometimes violent – worldviews. Such views as the belief in terror hoaxes on Gab, anti-migrant narratives on TikTok and X, the right-wing conspiracy of the ’radical left’ on Quora, the ‘false flag’ conspiracy on Gab and Quora, anti-Labour, pro-Reform rhetoric on X, and either Islamophobic or antisemitic hate across all four platforms.
The prevalence of mis/disinformation, conspiracy theories, and widespread Islamophobic and antisemitic content follows similar patterns across TikTok, Quora, Gab and X. In an era of polarisation, emotionally instinctive responses tend to take the form of either grieving, fearful or pre-emptively defensive attitudes, which spread rapidly. All of which can be whipped up by certain platforms’ affordances and algorithmic amplification, often resulting in outrage, blame, and violent retaliatory rhetoric.
The focus of such rhetoric depends on the issue-alignments of the individuals on each respective platform. Across all, however, grief and speculation were fomented through emotive online mobilisation in the short and immediate online ‘attention window’. A combination of rage and mis/disinformation crystallises into persistent false narratives that were not deconstructed or tempered by ensuing facts or evidence that emerged following police investigations and mainstream journalism. Fig. 16 demonstrates the continued prevalence of disinformation a week after the attack.
Fig. 16: Screenshot of a Quora question and answer. Taken 08/10/25 at 13:45.
The Attention Window: A Critical Moment for Mis/Disinformation and Radicalisation
This case of immediate cross-platform reactions to the Heaton Park Synagogue attack is not unique. The cases of the 2024 Magdeburg and Southport attacks mirror the same cross-platform dynamics to a significant extent. Following the Southport attack in 2024, anti-migrant and Islamophobic users weaponised false information to mobilise conspiracies, culminating in the Southport riots. Similarly, following the 2024 Magdeburg attack, the European Commission labelled online platforms as “echo chambers of manipulation and disinformation” where “far-right groups exploited the incident to influence public opinion”.
This critical moment manifests particularly when there is no immediate identification of the attacker or motive-attribution, as was also the case in Magdeburg and Southport, leaving a knowledge gap to be filled by rampant speculation.
As such, the post-attack attention window presents a critical challenge. The lack of information, combined with high-intensity content production and fearmongering, provides a ripe opportunity for the rapid spread of resentment, fear, outrage and conspiracy that fuels extremism.
Preventing Radicalisation Online: Practical Solutions
Effective prevention of radicalisation, following a public health model, should operate at three levels: primary (untargeted), secondary (targeted towards vulnerable populations), and tertiary (targeted towards known radicalised individuals).
From a primary prevention standpoint, since the attention window represents a high-risk environment for radicalisation, this should drive responses mitigating risk. So far, most tech platforms remove content that violates their terms of service; however, it is impossible to do so quickly and comprehensively. Moreover, they are unlikely to remove borderline content in the realm of hate, misinformation, or conspiracy theories, instead focusing on the most egregious and violent content. And as we have seen, such borderline content helps foment and cement extremist narratives; so, this alone is insufficient. Public authorities engage in strategic communications during this attention window, with the principal aim of reassuring the public, reducing panic, and de-escalating violence (whilst avoiding prejudicing legal proceedings) to prevent reprisal attacks.
Given the changing nature of the information environment post-attack, small tweaks to the current approach could have a hugely positive effect; public authorities should:
- Revisit their strategic communications playbooks and consider whether they need to be more authoritative.
- Issue safety messages warning of the prevalence of misinformation and conspiracy theories during this window.
- Conduct preventative measures ahead of time to build general media/digital literacy and ‘prebunk’ such narratives.
Regarding secondary prevention, the attention window in the hours after a violent incident is crucial. Because extreme narratives are mapped onto pre-existing belief patterns and reinforce existing prejudices, this window presents a risk for acceleration or reinforcement of existing radicalisation pathways. Similarly, this window presents an opportunity to identify and engage individuals who are at heightened risk of exhibiting radicalisation indicators.
Online positive interventions are key; they can be prepared ahead of time and activated during the attention window. If we can predict extremist narratives and understand the target audience, we can design compelling positive interventions and counter-narratives. For example, tech companies should:
- Link towards educational content or support resources.
- Direct individuals off-platform and out of echo chambers.
- Do more to explore the temporal aspect and surge them during the attention window.
Effective tertiary prevention efforts require more individual-oriented, holistic approaches in the long term, recognising that individuals can change their behaviour when supported. Whilst seemingly at odds with a timebound attention-window, we acknowledge that sometimes extreme acts of violence are a bridge too far for vulnerable individuals, and incidents can present a cognitive opening to begin a prevention intervention. This is a crucial time to raise awareness of the support services available to provide pathways to de-escalation for those most at-risk. Immediate, content-neutral, behaviour-focused communications and tech interventions responding to specific indicators, such as increasing content violations, can prompt individuals to engage off-platform with experienced practitioners before these narratives solidify. Through partnerships with tech platforms, Violence Prevention Network aims to target these to the most at-risk individuals: identified based on their on-platform behaviours that are indicative of vulnerability to radicalisation – such as multiple, severe violations of a platform’s terms of service or strong connections to users removed for terrorism.
With this targeting already in place, platforms could legitimately read more into their engagement with hateful or conspiratorial discourse during this attention window. Attacks such as Heaton Park would hold less power in encouraging the cycle of extremism.
—
Ella Goldschmied works at Violence Prevention Network International based in London. She works primarily with the European Observatory of Online Hate and supports VPN’s projects worldwide. She studied Political Science at the University of Amsterdam, and her personal research focuses on the violent rhetoric of far-right groups on Quora and the impact of conspiratorial narratives in the US.
—
Are you a tech company interested in strengthening your capacity to counter terrorist and violent extremist activity online? Apply for GIFCT membership to join over 30 other tech platforms working together to prevent terrorists and violent extremists from exploiting online platforms by leveraging technology, expertise, and cross-sector partnerships.