“Vulnerable and easily influenced”: examining the vulnerabilities and susceptibilities associated with the engagement of young Australians in extremism

ACE

= Adverse childhood experiences;

ADHD

= Attention deficit hyperactivity disorder;

ASD

= Autism spectrum disorders;

ASIO

= Australian security intelligence organisation;

CCE

= Child criminal exploitation;

CSE

= Child sexual exploitation;

DCJ

= Department of communities and justice (NSW);

ESP

= Engagement and support program;

FACS

= Family and community services;

ODD

= Oppositional defiance disorder;

OOHC

= Out of home care;

VCCYP

= Victorian commission for children and young people; and

XRW

= Extreme right wing.

Young people are increasingly being targeted for engagement by violent extremist organisations and individuals. This occurs across the extremist landscape but has gained prominence recently in association with extreme right-wing (XRW) milieus. Western security agencies have warned that the age of radicalisation and engagement into XRW organisations and online subcultures is decreasing. One recent example comes from the UK, where Counter Terrorism Policing established that while referrals to PREVENT’s anti-radicalisation program had declined overall, the number of young people being referred to the program had increased to 38%, constituting 48% of PREVENT’s total caseload (Home Office, 2021; Counter Terrorism Policing, 2021). In Australia, there have been several recent high-profile incidents of extremism-fuelled violence perpetrated by young people, including the fatal stabbing of a schoolboy in Perth and six teens arrested in New South Wales (NSW) relating to a stabbing in a Sydney church. The June 2024 arrest of a 19-year-old male in Newcastle, who was allegedly planning a terrorist attack on a NSW Labour politician, Tim Crakanthorp, is another indication of this growing problem (Beazley and Canales, 2024).

The Director-General of the Australian Security Intelligence Organisation (ASIO) stated in his 2021 annual threat assessment that children as young as 13 were becoming radicalised (Burgess, 2022). Where youth radicalisation once consumed only 3% of ASIO’s new counter-terrorism investigations, it soon consumed 15% of new investigations and 50% of priority investigations (Burgess, 2022). The intensity and reach of radicalised youth were also noted to have increased, with the Director-General suggesting that young people are assuming operational and leadership positions. This sentiment was echoed in the Inquiry into Extremism in Victoria, where the Legal and Social Issues Committee fielded several submissions highlighting the risk to Australians as young as 10 (Wong and Rollason, 2022). This was subsequently reiterated in the 2024 Annual Threat Assessment (Burgess, 2022). While youth radicalisation and engagement into violent extremism is not a new area of inquiry, with seminal studies like Horgan et al. (2017) and Asal et al. (2008) leading the field, the decreasing in the age of people involved in the XRW milieus suggests a problem worthy of further inquiry.

Using Australia as a case study, article sheds light on the nexus of vulnerabilities that may make a young person more susceptible to engaging in or supporting extremism, through examining demographic, personal and social environments. By using key criminal justice and extremism paradigms, it exposes the presence of childhood trauma and adverse experiences, mental health issues, grooming, educational and socialisation factors and indications of neurodiversity among young Australians associated with extremism. We suggest that the susceptibility of these young people is informed by multiple overlapping vulnerabilities, rather than a single, isolated cause. This confluence of factors leads us to suggest how therapeutic and trauma-informed systems might be best positioned to support the disengagement of young people from extremism. We achieve this through a case file analysis of young people engaged or formerly engaged with extremism in NSW and through interviews with expert practitioners managing young people. This study was undertaken in association with the Engagement and Support program (ESP) and associated team members within the Department of Communities and Justice (please see funding note). The ESP is a voluntary program, unlike PREVENT which has mandatory referrals. ESP works with clients and their families to identify, support and manage complex needs in the long term for sustainable disengagement. Engagement with ESP is not, nor has ever been, compulsory or made a condition of bail, and clients can withdraw at any time.

This article presents our findings, focusing specifically on the key susceptibilities affecting young Australians in the cohort considered. We start by reviewing the three intersecting sub fields of literature, being youth, vulnerability and trauma, before detailing our methodological approach and research questions, followed by an overview of the findings. We analyse these findings across the themes of childhood trauma, neurodiversity, mental health, grooming, education and socialisation. This suggests that an intersection or interplay of these factors may increase the susceptibility of vulnerable young people to supporting extremism. We suggest therapeutic and trauma-informed systems are required to support the disengagement of young people from violent extremism in Australia.

It is well established that vulnerable children are susceptible to exploitation. This includes those that go missing (“run away”) and those from out-of-home care (Colvin et al., 2018; Jay, 2014; Shaw and Greenhow, 2020; Victorian Commission for Children and Young People, 2021). Franchino-Olsen’s (2021) study revealed that children vulnerable to child sexual exploitation (CSE) have several relevant risk factors, including involvement with child protection and a history of trauma and abuse. Jay’s (2014) report into the exploitation of children in Rotherham, UK, revealed a staggering scale of abuse and exploitation with an estimate of 1,400 victims, many of whom were used to gain access to other vulnerable children. Locally, the Victorian CCYP’s (2021: 18) inquiry into children going missing in care found that over a third of the files they reviewed had references to sexual exploitation. Child victims’ involvement in the child protection system, coupled with their histories of trauma and abuse, had made them more vulnerable to CSE, including helping perpetrators access other children.

The Victoria CCYP (2021) report also found evidence of child criminal exploitation (CCE), where people or groups target vulnerable children to participate in criminal activity (see also evidence from England in (Home Office, 2023; All Party Parliamentary Group for Runaway and Missing Children and Adults (APPG), 2019) and the Howard League for Penal Reform Howard League, 2020). Yet the extent of this issue is unknown as data have typically not been collected, either in the UK (Howard League for Penal Reform Howard League, 2020) or Australia (Baidawi et al., 2020). Neurodiverse children are more susceptible to CCE (Baidawi et al., 2020). CCE can range from coercing children to commit criminal acts online, minor offending such as theft, to involvement in large, organised crime gangs. The “county lines” phenomenon is an example of the type and scale of CCE that vulnerable children may be coerced into committing. Vulnerable children were used to traffic drugs across county lines in England and Wales to avoid police detection (Stone, 2018).

Cognitive impairment and neurodiversity may also engender more vulnerability to engagement. Neuroatypical and cognitively impaired children are also over-represented in the criminal justice system (Greig et al., 2019). Furthermore, young people with learning disabilities may be more susceptible to online engagement and may be groomed into extremist activities through online mechanisms (Caton and Landman, 2021). Caton and Landman also found that grooming techniques used by terrorists are similar to those used by paedophiles. The UK Government (2019) noted its concerns over the rise in the use of the internet to recruit people to extremist behaviour and the specific targeting of vulnerable people. Research has shown that these concerns, particularly the targeting of young people who are cognitively impaired or on the autism spectrum by using social media, are warranted (Allely, 2017; Bujus et al., 2017). Recommendations have been made to address the online grooming of children into extremism. Some research notes the importance of increasing digital literacy and critical thinking for young people (Reeves and Crowther, 2019). However, these recommendations are still focused on young people themselves, not the perpetrators or systemic changes. At the heart of these challenges is the vulnerability of the child involved and the power exerted over them by their abusers, who see themselves as recruiters. As the exploited children age, they may become the recruiters of younger children into CCE and CSE (Jay, 2014).

Engagement into extremism refers to the process through which an individual becomes part of an organisation, spanning coercive engagement, child engagement and open, non-coercive forms (Alakoc et al, 2021). There is evidence to suggest that predatory adult recruiters use similar mechanisms to recruit vulnerable young people to extremism as paedophiles use to sexually exploit children (Almohammad, 2018). Various factors may contribute to susceptibility towards this type of engagement strategy. Studies of experiences growing up in war zones and conflict areas show that children exposed to violence and trauma are more desensitised towards violence and accepting of using violence to achieve personal value-driven forms of “justice” (Almohammad, 2018; McCue et al., 2017; Özerdem and Podder, 2011). Furthermore, a lack of strong family or community bonds could exacerbate this tolerance for violence. Children with fewer community ties were also less likely to develop coping mechanisms and resilience against engagement techniques (Özerdem and Podder, 2011).

Almohammad’s (2018) study of children recruited into Islamic State revealed that children were groomed using techniques that strategically targeted their families and wider community over long periods. In this study, Almohammad noted that children were selected by recruiters using similar criteria to paedophiles: children who had experienced violence, particularly domestic violence, children who had less adult supervision and children who had experienced neglect (see also Winters and Jeglic, 2017). A key finding was that children with no parents were specifically targeted through the ‘exploitation of the children’s psychological vulnerabilities’ (p. 7), similar to strategies used by paedophiles.

Özerdem and Podder (2011) found that children were often recruited through the promise of reward or loot, and where these offers could not be made, there was more likelihood of children being forcibly recruited into violence. Furthermore, children experiencing poverty and hardship may see engagement into an extremist group as a means to escape poverty and marginalisation (Özerdem and Podder, 2011). These studies demonstrate how recruiters use their power to coerce and force vulnerable children into joining their groups and activities. However, many of these studies focus on child soldiers in conflict zones and Salafi jihadist extremism, again highlighting the lack of research into how vulnerable children are recruited into other forms of extremism and how this happens domestically in Australia, as well as in other non-conflict countries.

There is a link between a young person’s involvement in illegal and problematic behaviour and their experience of trauma and deprivation. Developmental studies have drawn on Bowlby and Ainsworth’s theory of attachment to understand the lifelong impact of childhood experiences on behaviour and interpersonal relationships into adulthood (Ainsworth et al., 2015; Bowlby, 2005). Children with experience of trauma, such as being a victim of violence or losing a parent [known as adverse childhood experiences (ACEs)], may have disrupted attachments, which, while potentially having long-term consequences for mental health and behaviour, may also increase their susceptibility to coercion (Cook et al., 2005). Children in the criminal justice system have high levels of ACEs and trauma. A recent study showed that 87% of children in the youth justice systems across 13 countries presented at least one traumatic event in their histories, and children in youth justice are 12 times more likely to have trauma backgrounds stemming from ACEs (Malvaso et al., 2021). The link between trauma and criminal offending is evidenced by the significant crossover of children with experience in the out of home care (OOHC) and criminal justice systems (McGrath et al., 2020; Shaw and Greenhow, 2020). A recent study demonstrated that of the data sample, 50% of children with experience of OOHC appearing before the children’s court had trauma compared with 11% of the non-care-experienced sample (McGrath et al., 2020).

The link between trauma and extremism has also been documented in other research. Logan et al.’s (2022) research examined exposure to ACEs for those involved in extremism. They showed that people engaged in right-wing extremism in adulthood had a high prevalence of exposure to ACEs, with 70% of the sample reporting four or more ACEs. In recent Australian study by Oberg et al. (2023), the authors interviewed CVE practitioners and suggested that childhood trauma is a key vulnerability to explain the engagement of young people in extremism. This is supported by research in other contexts, which suggests that trauma interrelates with grievance and the potentiality for radicalisation to violence (Ellis et al., 2021), and which also highlights the significance of personal factors, setting events and social, political and organisational contexts (Ellis et al., 2015).

Research into the engagement of young people into extremism has drawn on several theories to understand the behaviour, including attachment theory, the prevalence of ACEs and autism spectrum disorder (ASD). A significant contribution to this discourse comes from an Australian study by Druitt et al. (2023), which examined suggestions around linkages between ASD and vulnerability for terrorism engagement. They suggest that such claims are lacking in evidence, or otherwise inconclusive as a singular explanation for vulnerability towards engagement in terrorism. In relation to vulnerable young people, a particular focus on attachment, social bonds and the impact of trauma is used to understand involvement in criminal behaviour (Anda et al., 2006; Bowlby, 2005). Social bonds, strong community ties and parental attachment are protective factors in criminal offending and involvement in extremism (Cherney, 2020; Loeber and Farrington, 2000). Vulnerability to offending behaviour and extremism is linked to weaker social and community ties. Life transitions, such as changing schools or moving home can become sites of vulnerability to engagement for young people (Cherney, 2020). When we examine the factors of vulnerability discussed above, there are similar risk factors present for vulnerable children, for example, the impact of trauma, low attachment and insecurity in housing placement. The literature demonstrates links between vulnerability, trauma, criminal offending and extremism but does not go far enough to unpack the similarities between the vulnerabilities of those who are criminally justice engaged and those who are extremism engaged – this is a gap we examine in this paper.

This study aims to better understand how young people may become involved in violent extremism. Australia, specifically the state of NSW, forms the case study for this. Ethical approval was provided by the investigators University Human Research Ethics Committee. The research had three stages: firstly, an exploratory literature review was undertaken to orientate the research in the field of knowledge. This also helped to scaffold the data collection points for the second and third stage of research. From this stage, data collection points were derived from terrorist radicalisation assessment protocol (TRAP)-18, in addition to the N Trilogy by Kruglanski et al. (2019). TRAP-18 by Meloy and Gill (2016) is a structured professional judgement tool to support professionals from across mental health, intelligence, law enforcement and other relevant areas who are attempting to assess whether an individual may seek to engage in lone actor terrorism. It lists 8 proximal warning behaviours and 10 distal characteristics. These aspects informed the data collection nodes, but were adjusted to be more appropriate for younger individuals (i.e. intimate relationships became relationship security). While TRAP-18 highlighted key aspects of personal context which was useful for data collection, Kruglanski et al’s N Trilogy highlighted other the factors of engagement with extremism. The N Trilogy is a theory developed to explain the key factors that lead to engagement in violent extremism specifically, being Needs, Narratives and Networks. This also informed data collection, where we understood needs as deficits in personal environment, narratives as informational content within the personal/social/environmental context and networks relating to social context of the young person, whether online or offline. Synthesizing key aspects of TRAP-18 and the N Trilogy allowed us to identify and collect meaningful data which would inform our understanding of the client’s personal environment (needs), and the client’s engagement with violent extremism (narratives/networks). Specifically, information was considered with respect to vulnerability and basic demographics, any noted diagnosis or suspected diagnoses, experiences of violence (as victim and/or perpetrator), life security (such as a stable home or experiences in OOHC or with child protection), personal health (including mental health or physical challenges), family dynamics, reports of trauma, secure attachments, ideology, identification (with in-group), attribution of grievance (out-group), setting incidents and education. Data collected on recruitment, radicalisation and referral details forms the basis of other papers (Campion and Colvin, 2024; Campion and Colvin, 2025).

From there, the second stage of research was to undertake semi-structured interviews (n = 6) with key staff in the ESP team. The ESP team had diverse backgrounds, including social work, policing, psychology, law, corrections and child protection. Some had worked in adjacent spaces for 25 years, whilst others had only moved into the role more recently. They were given the gender-neutral pseudonyms of Morgan, Rory, Blake, Alex, Quinn and Taylor. The interviews went for 45–70 min each. Once completed, they were transcribed, viewed and approved by the interviewees. They were anonymised and then analysed using an deductive thematic analysis, where themes are generated from the data (Braun and Clarke, 2021). The transcripts were coded and analysed by the research team and then organised into broad thematic and sub-thematic areas. This produced three broad themes: being demographic factors, personal environment and social environments, with additional subtheme discussed in the Findings and Discussion.

Synchronously, the third stage of research began, where de-identified case files were reviewed (n = 17). This also drew on, and was informed by, the data noted above. The files were viewed under supervision at the DCJ office. File notes were de-identified at the time of access with pseudonyms used, such a Midhat, Latif, Adam etc. All collected data was processed and stored in line with National Health and Medical Research Council procedures. The files were analysed using quantitative and qualitative methods. Quantitatively, descriptive statistics were derived from the raw data that provided an overview of clients’ key characteristics and demographic factors, such as age. This highlighted basic demographic factors. Qualitative analysis was also conducted on the data through an evidence-based thematic approach derived from the literature review. This provided insights into the raw trends and patterns within the data set. The outcomes of Phase two (interviews) were essentially matched with Phase three (case file analysis), to identify convergence and divergence. These outcomes were then examined in greater depth in the Findings and Discussion.

It must be acknowledged that there were many gaps in the data, due to procedural, reporting or information sharing practices. While this, plus the size of the data set, may be seen as a complication, it nevertheless has value by providing specific insight to the NSW cohort which may have meaning and generalisability elsewhere. Indeed, early indications showed alignment with research emerging from the USA (Logan et al., 2022; Windisch et al., 2022), and in complement to Australian research on vulnerability (Druitt et al., 2023; Oberg et al., 2023).

Given the ambit of the research, it was necessary to define numerous terms. We adopted the Australian Institute of Health and Welfare definition of youth, as those aged between 12 and 24 years; children as those aged between 10 and 17; and young people collectively referring to the 10–24 age bracket (Australian Institute of Health and Welfare, 2021). Vulnerability is being used to refer to factors that make people more susceptible to being drawn into problematic or unlawful activities, such as ACEs, mental illness, cognitive or developmental impairment, age, gender, sexuality, ethnicity, personal issues and so on. This, in association with social and environment factors such as insecure housing and home lives, poverty, experiences of violence or poor interpersonal relations combine to create overall susceptibility.

Extremism was defined as used here to refer to the use, support or threat of violence by others to promote ideological, religious or political agendas/goals – a definition contextually grounded in Australia (DFAT, 2022; Winter et al., 2020). An individual may be extreme by promoting the use of violence by others, praising or glorifying such violence or supporting it. An individual becomes a violent extremist when they commit acts in preparation for violence, engage in or threaten violence. Understandings of right-wing extremism came from Mudde (2000) and Carter (2018), which highlight core defining attributes as authoritarian, anti-democracy and exclusionary nationalism, of which racism and xenophobia tend to be consequences. Salafi jihadism, in this study, referred to a fundamentalist Islamist movement that seeks to revive practices from the first three generations of Islam, and in which jihadism is construed as physical, violent and obligatory struggle for all devout Muslims (Maher, 2016). This is relatively consistent with the broader body of literature.

This study produced three themes with additional sub-themes which speak to the young person’s demographic circumstances, personal environment and social environment. The demographic circumstances relate to gender, age, location (in terms of rural or metropolitan), time on the program, cultural or ethnic heritage and engagement with an extremist milieu. The personal environment details personal experiences, such as childhood trauma, neurodiversity or diagnosed mental health challenges. The social environment speaks to their experiences being groomed, their experiences at school or in social interactions. This is then assessed as a confluence of factors which overall impact on vulnerability and susceptibility towards extremism. We will discuss these in turn, drawing on both the case files and the interviews.

The reviewed files comprised 16 males and 1 female, meaning a gendered analysis was not possible. The age range (at the time of referral to the program) was 13–27 years, with a median age of 19 and a mean age of 20 (rounded up to the nearest year). Thirty-five per cent of the data set were from a regional area, and almost 60% were from the western suburbs of Sydney. The average time spent on the program since the referral was 2.7 years, ranging from a few months to 10 years. Seven of the files were individuals who identified as an ethnic minority. Some files did not contain information that explicitly stated cultural or ethnic heritage. Four people were identified as Aboriginal (24%). The clients were observably engaged with either Salafi jihadism (47%) or right-wing extremism (53%). No anti-government extremism or other forms were immediately observable.

We examined the files for any ACEs that clients may have noted. We found that 12 of the 17 files (70%) contained evidence of ACEs, and many of these clients experienced multiple ACEs. A couple of clients witnessed the effects of war in their country of origin. Several clients lived in deprivation and poverty. Family violence was a factor in the childhoods of seven clients either witnessing violence or as victims. The childhoods of some clients were described in case notes as “violent and chaotic”. Other traumatic events experienced by clients included witnessing deaths, finding dead bodies, the death of a parent or other family member, a family member imprisoned, sexual abuse or sexual abuse of a family member and the loss of the family home. Four of the clients’ files reviewed revealed some form of child protection involvement. Three clients lived in kinship care, two formally and one in an informal arrangement. Another client had two half siblings removed from the family home, although the client was not removed. Another client was referred to child protection services, but the department found the mother so hostile that a case note states, “community services are no longer assisting”.

Staff were clearly aware of the traumatic backgrounds of their clients. Morgan commented: “We do see a bit of it. Where there’s been some breakdowns in relationships. There’ve been victims of abuse to seeing domestic violence and all that sort of thing”. Quinn also notes: “All of the clients have some number of traumas in their background, unstable home life, drugs and alcohol, mental health issues, where they’ve had a traumatic event occur”.

Four clients had some form of contact with child protection. Taylor commented that given the backgrounds of other clients that “[…] probably some should have [been referred to child protection in childhood] to be honest with you”. Rory noted that “we do have kids that bounce to grandparents, and then bounce to care or bounce to mum, and then goes into the responsibility of the Minister. So FACS (Family and Community Services [FACS] – the NSW child protection authority) is often there in and out”. Furthermore, Rory saw an increasing interest from FACS staff in learning more about the program as they saw it relevant for some children in their care.

However, while observing trauma backgrounds in clients, there were mixed understandings of the link to involvement in extremism. Alex stated “I never really thought that trauma was that connected, but it certainly does seem to be with the current list of people. And maybe it’s that just instability in early years of life leads to greater instability in your adolescence as you’re trying to form your own identity. I’m not sure”. Taylor saw a connection between neurodiversity and childhood trauma as prevalent in their client list, “ASD is the common denominator with my clients who’ve had that background”. However, Blake saw the trauma backgrounds, particularly deprivation, as an instigator for their involvement in extremism:

While research has identified links between trauma and social cohesion (Ainsworth et al., 2015), trauma and vulnerability (Cook et al., 2005), trauma and criminal justice involvement (Malvaso et al., 2021) and trauma, conflict and engagement in extremism (Almohammad, 2018), this research demonstrates a clear link between vulnerability, trauma and involvement in extremism among young Australians.

While undertaking this study, the department highlighted neurodiversity as an area where more information was required. As such, we were keen to see what the data revealed and the thoughts of our interview participants. Several client files indicated that there was suspected ASD or that clients displayed symptoms of ASD, but either they had not yet been formally diagnosed or their diagnosis did not confirm the suspected presence of neurodiversity. Of our case file data set, three files had confirmed diagnoses of ASD, and an additional five files revealed suspected ASD or some form of neurodiversity, but it was either undiagnosed or the diagnosis did not confirm neurodiversity. Some files had contradictory information about neurodiversity. For example, notes on Murray’s file contain indicators of neurodiversity in the client. After referral for testing, the diagnosis was that the client did not have ASD but probably had some sensory processing disorder. Adam’s file contained a psychologist’s report that specifically confirmed an ASD diagnosis was not supported, but many notes on the file contradicted this.

Several clients had multiple diagnoses, not only ASD. In fact, when examining the presence of neurodiversity, it appears that the files revealed clients had behavioural and neurodivergent symptoms together. For example, Adam’s file, as noted above, did not contain an ASD diagnosis but noted symptoms perceived as autism by staff. However, confirmed diagnoses of oppositional defiance disorder and attention deficit hyperactivity disorder (ADHD) were present. Latif was diagnosed with both ASD and ADHD. Rodney was suspected of having Asperger’s syndrome and was formally diagnosed with ADHD. Kareem presented with both ASD and ADHD. In total, 10 out of the 17 files (almost 60%) contained some evidence of neurodiversity or behavioural disorders, often with comorbid presentations.

Participants raised neurodiversity as a key area of concern in their practice and were keen to explore this further. Our participants estimated that many, indeed most, clients were autistic. Blake estimated that the “majority have either ASD or autism possibly”. Taylor also believed that about half of the clients were neurodiverse:

Blake also noted that getting diagnoses was challenging: “It’s hard to get them assessed quite quickly. Takes a little time, and just having been on three years, some of them are still waiting to get those assessments done”. Other participants echoed this:

These comments reflect what was observed in the file reviews – that many clients present with certain characteristics or traits associated with neurodiversity but are not diagnosed. However, although we only examined a selection of files, there is a key difference between staff observations and what was in the files. There was a variance in staff observation of ASD and formalised diagnosis. In several files, staff had made comments that clients had neurodiverse traits, but formal assessment did not support these comments. Other factors, such as developmental concerns or behavioural disorders, may be creating the impression of ASD, further complicating case management.

Cognitive impairment and developmental disablement were other factors we considered in our analysis. Several clients presented with some form of impairment, whether from an acquired brain injury or cognitive or developmental issues. One-third of the clients in our sample (n = 6) had learning difficulties or delays. Taylor considered the functioning of clients in the program and noted that “without being a psychologist myself, I do think there’s a level of cognitive deficit, not necessarily neurodiversity, but maybe a parallel intellectual disability which I guess feeds into that vulnerability and susceptibility. Or potential ABIs [Acquired Brain Injury] as a result of parental substance abuse possibly”.

While studies such as Baidawi’s et al.’s (2020) and Greig et al.’s (2019) research explore criminal justice challenges for neurodiverse young people and Allely (2017) and Bujus et al. (2017) examine concerns over the role of online activities in their involvement in unlawful and extremist behaviours (see also Caton and Landman, 2021), our study indicates more specific expertise and knowledge is needed to better understand the role of neurodiversity and cognitive impairment. Furthermore, there is a need for better access to diagnostic services and better stakeholder training on neurodiversity.

Clients’ mental health was a significant factor in our theming and analysis. Twelve files (70%) revealed clients with mental health concerns. Seven clients had attempted suicide or had suicidal ideation. Other mental health issues reported in the files included mood disorders such as depression, anxiety, body image issues and post-traumatic stress disorder. Of note was a relatively low involvement with alcohol and other drug use than perhaps expected for a cohort of people with their ACE history (n = 5) (however, on a par with the proportion noted in Cherney et al.’s 2022 study, but at odds with Gill et al. (2021). While problematic alcohol and drug use by clients’ parents and wider family was reported, this did not mean children followed this pattern of behaviour. Indeed, commentary on file sometimes revealed an anti-drug stance either stemming from parental drug use or for ideological reasons related to their extremism.

Blake considered a mental health risk to be instrumental in the overall assessment of risk, noting that while their risk of ongoing or re-ignited enmeshment in extremism may be alleviated, the mental health risks that contributed to their initial involvement may still be present. Blake highlighted that this led to ongoing support for clients with mental health concerns to mitigate the risk posed. Blake further wondered how the risk might ever be mitigated for those with ongoing mental health concerns:

Noting the mental health issues of the clients, our interview participants talked about the challenges of securing appropriate support to address these issues, particularly in regional areas, as illustrated by Blake and Taylor.

The health services have been extremely challenging, especially in the rural areas. (Blake):

Although acknowledging the lack of services in regional areas is acute, this did not mean that urban areas fared much better:

Further early diagnosis and intervention were seen as opportunities to intervene that may have prevented clients from being caught up in extremism or criminal justice involvement. Taylor saw lots of “missed opportunities’ in clients” histories:

An important point made by Alex was the need for expertise in the particular issues for the clients caught up in extremist ideology. Alex noted that clients would benefit from being linked with professionals with expertise in extremist ideology:

Alex also noted that if an expert cannot be located, some training could be developed for allied health professionals to address the knowledge gap.

While our research focused heavily on trauma, particularly in childhood, and there is a clear connection between trauma and poor mental health (Cook et al., 2005; Malvaso et al., 2021), we did not anticipate the level of serious mental health concerns among the studied cohort. The prevalence and severity of mental health concerns are an important piece of evidence for criminal justice and law enforcement practitioners to consider when responding to young people involved in extremism. Given challenges around appropriate and adequately funded mental health supports, particularly those equipped to respond appropriately to vulnerable young people engaged in extremism, these concerns need to be at the forefront of ongoing responses and initiatives.

We also saw how these intersecting vulnerabilities made clients more susceptible to harmful and unlawful activity and how they could be groomed into such activities. Documents on Latif’s file revealed that he had experienced “[…] grooming and abuse of power by a former support worker”. Henry recognised that he had been groomed and, in turn, had started behaving similarly to those who groomed him. His file noted that he “reflected he was surprised he could groom others more vulnerable – he realised he became the bully he very much despised”.

Due to our interest in grooming behaviour derived from our literature review, in our interviews we specifically asked participants about their observations of clients’ experience of grooming behaviour. Interestingly, we often did not need to ask this question because grooming had already come up unprompted in the discussion. Blake observed: “Yeah, they’re all quite vulnerable, I guess so. Yeah […] I find it them quite easily groomed, that’s for sure”. This was echoed by Rory, “because they are vulnerable just as child sex offenders where people are groomed – it’s the lollies in the van”. Alex commented that grooming happened regardless of belief system:

Taylor demonstrated how prevalent grooming was, particularly online: “Absolutely 100% grooming. I don’t think any of my clients – and I can only speak to my clients here, have the capacity, knowledge, know how, education to be local factors. They’re all being led online by other people who are unnamed”. Rory noted that grooming was often done with groups of young people, and part of the grooming was to encourage others to peer pressure the young person into engaging in harmful or unlawful activity:

Blake provided an example of how groomers can coerce vulnerable clients into engaging in problematic behaviour through threats of blackmail and extortion:

Encouragingly, interviewees perceived an awareness of authorities of the susceptibility of some vulnerable young people to grooming. Rory talked about a level of understanding from some quarters: “[…] they’re trying to do that early disengagement, they’re seeing that they’re troubled kids, they’re seeing that they’re not outright crims. They’re obviously seeing in terms of the intel more complex workings of that grooming or the operationalisation of it”.

The prevalence of grooming vulnerable young people into extremism is a critical finding of this research. While the role of grooming vulnerable young people is recognised in research into CSE (Franchino-Olsen, 2021; Jay, 2014) and, to an extent, CCE (Shaw and Greenhow, 2020; Stone, 2018), this research provides solid evidence of these practices in extremism engagement. Of note is that the understanding of grooming behaviour and many official responses to grooming are limited, being viewed as sexual exploitation rather than, more broadly, adult predators recruiting young people into unlawful or extremist activities.

We noted that clients often had poorer educational experiences and outcomes as well as poor socialisation stemming from their mental health, neurodiversity and trauma backgrounds. Many of the clients in our file review sample had few, if any, friends. Many had problems forming relationships in adulthood. Clients were reported to be socially isolated or having very few friends; in eight files (almost 50%), this social isolation was specifically commented on in case notes. School for many clients in the sample was not a great experience. School refusal, suspension and expulsion were not uncommon. Bethany was asked to leave school in Year 10; Mohammad was asked to leave school after the incident that led to his referral to the program, as was Adam. Kareem had multiple suspensions and was expelled in Year 12. Latif had multiple suspensions and school refusals. Of most interest was the experience many of these clients had with bullying. Nine of the files (53%) contained evidence of bullying. Being bullied sometimes went together with being the bully and disruptive behaviour at school. Edward retaliated against his bullies with violence, egged on by his parents. Latif reacted strongly to perceived bullying and exhibited problematic behaviour as a result. Adam experienced bullying and, in response, often refused to go to school; however, when attending, he made threats and challenged teachers’ authority.

Quinn talked about the bullying experiences of clients, noting that this compounded existing trauma backgrounds for many: “There’s also trauma where things have happened through education. For some of the younger ones—as in the bullying that occurs with those cases”. Blake also noticed the experience of bullying and its incongruence to the “tough guy” image of extremism: “I think all our right-wing clients, what they seem, they’re not romper stompers, they’re not huge, they’re not massive, they’ve all had bullying issues at school”.

Alex also noted that bullying may have led the clients into exploring avenues that may have led to the behaviour that resulted in the program referral: “[they are] not doing well at school in terms of mostly being bullied or bullying others, not connecting with their teachers at school or finding their place in school in general and so looking outside”. Quinn observed how being bullied led to clients becoming the bully: “It is very easy for them to slide into being the antagonist and being that person that they once were very unhappy about—how they displayed behaviours towards them”. Rory talked about how bullying could lead to becoming a bully and engaging in extremism as a protective measure:

Blake explained how clients experiencing bullying felt empowered by their extremist involvement:

The files also revealed examples of bullied clients who had adopted “tough guy” personas and this had led to them engaging in unlawful and harmful behaviour, which ultimately contributed to their pathway to extremism. Case notes and reports from Kamal’s file illustrated this, noting that he “developed a persona as a neo-Nazi as a means to counter and ultimately avoid physical threats emanating from older peers”. Rodney experienced homelessness when he was quite young and connected with white supremacist gangs to protect himself. He also started carrying a knife on his person from the age of 10 and took anabolic steroids to appear physically intimidating.

Education and social bonds are critical protective factors for disengaging young people from extremism (Cherney, 2020; Loeber and Farrington, 2000). However, the cohort studied struggled to engage positively with education and were frequently victimised in educational settings, limiting their positive social bonds. Effective responses to these vulnerable young people must include opportunities for positive social engagement and supportive educational environments to strengthen these positive social bonds.

The presence of multiple, intersecting vulnerabilities for so many clients was seen by interview participants as contributing to their susceptibility to be caught up in illegal or harmful activities. Case notes and reports on file painted a picture of naïve and easily influenced clients. Alex saw the inability to form good social connections resulting from neurodiversity as a vulnerability factor in being susceptible to being drawn into problematic behaviours and unlawful activity: “There’s no good way to say it, it’s almost like there’s shortcuts if you’ve met that neurodiversity threshold because you do have issues with social skills and you do have issues with executive functioning and all those things that almost place you at a greater vulnerability”.

Case notes on file and comments in referral forms drew a picture of clients who were “young and impressionable” (Mohammad’s file), “vulnerable and easily influenced” (Malik’s file), “fearful and naïve” (Tony’s file), “highly impressionable” (Mustafa’s file) and “naïve and influenced” (Midhat’s file). Latif’s file illustrated a bleak picture of a troubling childhood with a report stating that he had “been taken advantage of and subjected to horrific abuse both physically and sexually in the past in home, school and community-based environment in addition to being exposed to significant domestic abuse perpetrated by former stepfather…grooming and abuse of power by a former support worker”.

A key understanding we drew from the data here is that vulnerability is linked to susceptibility. When professionals engage with vulnerable young people, they need to be aware of their susceptibility to engagement, both online and offline. This can aid professionals in responding with appropriate protective measures.

What became clear from our analysis of the clients’ histories and the perspectives of case managers, as outlined in the previous sections, was that clients presented with multiple intersecting vulnerabilities, from mental health concerns to neurodiversity to multiple ACEs. When asked about the commonalities of clients in the program, participants echoed what we found in the files – clients had challenging life experiences. Taylor said: “All of my clients have had pretty disruptive childhoods, and that’s a nice way to put it”. Quinn saw these experiences as linking in with the behaviour that led to program referral:

Rory also linked these complex histories of clients to their program referral:

This research demonstrates that the best practice responses to vulnerable young people engaged in extremism consider the multiple intersecting challenges these people face in their daily lives and provide wrap-around and tailored support services to assist with these underlying social factors. Any government or non-government organisations seeking to disengage and prevent young involvement in extremism should take this approach rather than focusing narrowly on law enforcement interventions.

Given the multiple intersecting vulnerabilities of clients and their trauma histories, systems and organisations need to be more agile in how they respond to young people caught up in extremism. There are calls for the criminal justice system to become more trauma-informed in its approach (Campion and Colvin, 2024; McLachlan, 2024) and we recommend all organisations interacting with young people at risk of or caught up in extremism to be responded to in a trauma-informed manner. Overarchingly, this requires a whole of system shift and adoption of successful, existing programs and approaches on a more widespread scale. For example, intermediaries to aid communication with vulnerable people, diversionary options, reduced punitive interventions and therapeutically agile responses in legal decisions. An example of this was raised in the data collection: A boy facing serious charges relating to online behaviour had been given the bail condition not to access the internet. This child had poor socialisation and communication skills. However, one positive area was his online gaming connections. So, while his online behaviour, on the one hand, had resulted in alleged serious offending, his online experiences were not exclusively harmful or unlawful. The bail condition left him cut off from positive communication and positive communities, which may be protective factors that could help him with the problematic behaviour that led to the criminal justice interventions.

A trauma-informed therapeutic response might consider a bail condition that balances preventing further unlawful activity while still fostering positive online connections as a preventive measure. This, in part, reflects some aspects of the risk-needs-responsivity (RNR) model of risk assessment. While we did not conduct a risk assessment, and acknowledge the differences between our cases and the offenders noted in Bonta and Andrews (2006), the RNR model seeks to highlight tailored interventions or treatments which “can be enhanced if the treatment intervention pays attention to personal factors that can facilitate learning” (p. 7). This includes consideration of client strengths and socio–biological–personality aspects in the treatment plan, in addition to engagement in community settings.

While our findings identify the vulnerability and susceptibility of children to engagement, we reiterate that in addressing these engagement strategies, we cannot just focus on labelling children as “at risk” and possibly further alienating or isolating vulnerable children. We must be aware of the community stereotypes, stigma and resistance young people may face as they undertake pathways out of extremism. We also must keep to the forefront the reality that, as illustrated by our interview participants, these young people have been “dealt a bad hand”, put in the “too hard basket” or otherwise been failed at some level by the families or communities who should have prioritised their safe development.

Young people become involved in extremism due to a confluence of factors that interact to augment their susceptibility to engagement and radicalisation. This is contingent upon existing vulnerabilities, such as trauma and adverse experiences, cognitive and developmental impairment, mental health, home and school environment and family dysfunction. These intersecting vulnerabilities may render some young people more susceptible to engagement approaches – both intentional and unintentional, via direct or online relationships. In some cases, these young people are already victims at the time of their grooming or guidance into extremism: simply fostering resilience in the young person overlooks the essential point that they need the protection of adults – from other adults. We need to acknowledge the role that adults play in the grooming process and move to ensure that there exist protections to prohibit this behaviour, such as specific legislation and offences that target predatory recruiters.

We make several recommendations to improve practice and policy to counter violent extremism engagement by children and young people. The first relates to legislative reform, such as an expansion or addition of an offence making it illegal to groom a young person for extremism. With respect to trauma-informed/responsive systemic approaches, not only should professionals and agencies that interact with young people involved in extremism be trained on trauma and its impact, but system responses should be improved to be more appropriate for these people. An example might be bail conditions that consider the young person’s cognitive development and the conditions within their ecosystem that may inhibit compliance. The use of intermediaries or professionals specifically trained to communicate complex information to vulnerable people in court settings or criminal justice programs is a valuable consideration here. Liaising with occupational therapists and speech pathologists about comprehension and communication for people presenting with multiple intersecting vulnerabilities, such as this cohort, is another pathway forward.

Recording the presence of ACEs and other vulnerability factors is critical to understanding more about who gets caught up in extremist activity and potential crossovers with criminal justice involvement. Given the similarities in characteristics between the studied cohort and young people involved in the criminal justice system in Australia, a better awareness of crossover is necessary. Care should be taken to ensure that this does not lead to net-widening but rather that information sharing is helpful, and that a more comprehensive understanding of the needs of vulnerable individuals caught up in extremism is achieved.

The practice currently applied by the studied program appears to be a model of best practice where young people can receive support without stigma. Staff have experience and education that demonstrates an understanding of both extremism and the multiple intersecting vulnerabilities that may render young people susceptible to engagement. More support is required from the government and community to further augment and refine CVE practice in this context. Furthermore, there is the potential that this model could have inter-relatability in other areas where complex case management is required.

While this study was conducted in Australia, the findings and recommendations provide insights and understanding that can be examined more generally in all jurisdictions facing the challenges of youth engagement in extremism. Indeed, there is evidence that similar challenges exist in comparable jurisdictions, such as England and Wales, Canada and the USA, as demonstrated in our background and literature review. Our findings can aid criminal justice practitioners and lawmakers by providing an evidence base for best practice policy in this area.

The engagement of young people in extremism – whether religiously or ideologically motivated – is often informed by a confluence of factors. While these factors are well understood by a specialist community of practice, there remains a gap in both community knowledge and academic scholarship about the engagement of young people in extremism. It is essential that young people engaged in extremism can receive experienced, trauma-informed and needs-based support as they begin their disengagement pathways. This is not simply a task for the government or its associated teams and departments, but an approach that requires the greater support of young people’s communities.