When Autism and Extremism Get Tangled Up

Every few months, a headline lands that makes autistic people brace for impact. A young man does something terrible. Somewhere in the story, the word "autistic" appears. And the public conversation drifts, again, toward the idea that there's something inherently dangerous about how our brains work.

In this episode, I walk through a 2026 study from Autism in Adulthood that pushes back on that drift. Researchers interviewed family members, friends, and clinicians who watched autistic people slide into extreme ideologies up close. What they found is more complicated, more compassionate, and more useful than the headlines suggest. 

Topics covered include: 

• Why autism alone does not explain extremism 
• What "negative autistic identity" means and why it matters 
• Hyperfixation as both superpower and liability 
• How algorithms exploit the gaps left by under-resourced autistic people 
• What real prevention looks like — and why it isn't surveillance 

Watch this episode on YouTube.

If you'd like to know more about topics discussed in this episode, check out: 

"Family and Clinician Perspectives About How Autism and Extremism Intersect" by Sachindri Wijekoon et al.

"Neurodivergence and the Rabbit Hole of Extremism: Uncovering Lived Experience" by Sachindri Wijekoon et al.

"Autism in England: Assessing Underdiagnosis in a Population-Based Cohort Study of Prospectively Collected Primary Care Data" by Elizabeth O'Nions et al.

"Understanding the Use of the Term 'Weaponized Autism' in an Alt-Right Social Media Platform" by Christie Welch et al.

"Social and Individual Grievances and Attraction to Extremist Ideologies in Individuals With Autism: Insights From a Clinical Sample" by Cécile Rousseau et al. 

Theme music: "Everything Feels New" by Evgeny Bardyuzha.

All episodes written and produced by Kristen Hovet.

Send in your questions or thoughts via audio or video recording for a chance to be featured on the show! Email your audio or video clips to otherautism@gmail.com through WeTransfer.

Buy me a coffee!

Buy The Other Autism merch. Use code FREESHIP for free shipping on orders over $75 USD!

The views, opinions, and experiences shared by guests on this podcast are their own and do not necessarily reflect those of the host or production team. The content is intended for informational purposes only and should not be taken as medical or professional advice. Please consult with a qualified healthcare provider before making any decisions related to your health, fitness, or wellness. 

Transcript

Kristen Hovet (00:00)
Welcome to The Other Autism podcast. I'm going to be doing things a little bit differently today. So normally you would expect an intro here with thanks to certain listeners.  I'm switching things around a little bit and doing something a little bit different. So there is a particular flavor of headline that shows up every few months: A young man, often white, often described by a neighbor as quiet or a loner, does something terrible. And buried somewhere in paragraph six or seven of the news article, sometimes earlier, sometimes in the headline itself is the word autistic or with autism. And every time it happens, autistic people, well, we brace ourselves because we know what comes next — the think pieces, the pop psychology, the slow drift in public conversation toward the idea that there's something inherently dangerous about how our brains work. 

Here's the thing though, the actual research on this is so very small. The actual research is also way more interesting and way more compassionate than the headlines would have you believe. And there's a new paper out that I want to walk you through today because I think it really matters for autistic people, for our families, for the clinicians who work with us, and honestly, for the broader conversation about who gets pulled into extreme ideologies and why. So today we're going somewhere a lot of autistic people do not actually love going. We're going into the research on autism and extremism. And I think you're gonna be surprised by what's actually in it.

Today's episode is about a study that came out in Autism in Adulthood in 2026, this year. It's called Family and Clinician Perspectives About How Autism and Extremism Intersect. The lead author is Sachindri Wijekoon at Western University, and the senior author is Melanie Penner at Bloorview Research Institute in Toronto. There's a long list of co-authors, including the autistic advocate, John Elder Robison, and the former extremist turned de-radicalization specialist, Christian Picciolini. I'll come back to those two names later because their involvement actually shaped the study in important ways. But before we get into what they found, I want to set the table because if we just jump straight into the findings without the context, it's really easy for any of this to get misread. So let's talk about what we knew before this study came out.

First, the basics. Extremism, in research terms, is generally defined as the defense or promotion of in-group supremacy, thinking that the in-group, the group that they belong to, is better than everyone else in some way. It's grounded in rigid ideological commitment and it's expressed through beliefs, social influence, or actions that undermine the rights, dignity, or inclusion of out-groups or people who are not part of the in-group. And extremism, like a lot of things, exists on a continuum or a spectrum. So you've got people who passively consume extreme content on the one hand or on the one side, all the way through to people who actively enact violence in the name of their belief system and their in-group philosophy. 

The research on extremism more broadly tells us something really important. It tends to grow in soil that's been prepared by systemic discrimination, marginalization, and social inequity. People who feel like the world has no place for them are way more vulnerable to groups that promise belonging and a sense of justice. That's true regardless of whether someone's autistic or not. Now when you narrow the lens to autism specifically, the picture gets more complicated and frankly more frustrating because most of what's been driving the conversation about autism and extremism has not been research. It's been media, which is true for a lot of topics, many topics. High profile cases get a lot of attention, the autism label gets attached, and a narrative starts to build and it runs away from reality basically.

The actual prevalence data on autism in extremist populations is really murky. So the paper we're looking at today says that diagnosed autism in people involved in extremism is roughly comparable to autism prevalence in the general population. So somewhere around 1 % in the formal prevalence data. And I want to pause here because that number deserves a real look. 1 % is a diagnostic prevalence figure and it only counts people who've made it through the system, gotten in front of the right clinician and walked out with a formal autism diagnosis. It does not count the rest of people who have not gotten diagnosed, who've missed, who've gone, who've fallen through the diagnostic cracks. There's now a substantial body of research telling us that the diagnostic numbers are a serious undercount. A landmark study published in the Lancet Regional Health Europe in 2023, led by Elizabeth O'Nions at University College London, looked at primary care records for over six million people in England and compared them against community-based prevalence research. 

What they found was striking. They estimated that more than nine in 10 autistic adults aged 50 and older in England are not diagnosed.

Among adults aged 20 to 49, somewhere between 150,000 and 500,000 people were estimated to be autistic and undiagnosed. The midpoint of those estimates suggests roughly 750,000 undiagnosed autistic adults in England alone. An earlier study out of Cambridgeshire estimated that around 40 % of all autistic individuals go undiagnosed.

That was based on data that's now over a decade old before we really understood how autism can present in women, those born female, in racialized communities, in people who are high masking, and of course, in late diagnosed adults. More recent work from the Global Burden of Disease Study estimated about one in 127 people globally are autistic.

But that figure also relies heavily on identification systems that we know are leaky or just really outdated, really old. So when you read the 1 % number in the autism and extremism paper, the honest read is this. The people we're comparing... the autistic people in extremism research are also being measured by diagnosis, which means both sides of the comparison are missing the same large group of undiagnosed autistic people.

We don't actually know whether autism is overrepresented in extremist populations once you account for everyone who's autistic but flying under the diagnostic radar. There may be some overrepresentation in specific subgroups like incel forums and very socially isolated online communities, but the data is limited. I want to be clear about why this all matters. So it's not because I think more autistic people are secretly extremists. I don't think that, and the research does not support that. It matters because we should be skeptical anytime anyone, including the researchers I generally trust, uses a 1 % prevalence figure as if it's the truth about how many autistic people are actually out there. It's the truth about how many of us the system has caught. That's a different question.

What earlier research has suggested, and I want to be careful with how I say this, is that certain autistic traits can increase someone's vulnerability to radicalization when other risk factors are also present. Things like difficulty forming meaningful social attachments, a strong need for structure and predictability, intense and specific interests, and sometimes differences in perspective taking.

The key phrase there is when other risk factors are also present. Autism on its own doesn't get anyone radicalized. We have to keep saying that because the public conversation keeps forgetting that. The research team behind today's paper has actually been working on this question for a while. Their previous study published in 2024 was the first to directly interview autistic people who had been involved with extreme ideologies. And what they found in that earlier study was that autism alone wasn't responsible. Shocker! 

The autistic participants had experienced a whole range of formative disruptions: early exclusion, ongoing marginalization, social rejection. And the extreme ideology was something they slid into, not something their autism delivered them to. But there was a gap. Some of the autistic participants in that first study were unaware of the full picture of their own involvement. Some were embarrassed. Some were afraid to share specifics. So the researchers went back and asked, who else has eyes on this? And the answer was, well, the people closest to them: family, friends, and the clinicians who'd been working with them. So how did they do the study? The team interviewed two groups. The first group was close contacts: seven family members, mostly parents, and one friend. The second group was five clinicians, psychologists, a social worker, a behavior analyst, and a physician. Most were based in Canada or the United States. Recruitment happened through Christian Picciolini's de-radicalization network and through social media.

Interviews ran 60 to 90 minutes over Zoom. They used something called reflexive thematic analysis, which is a qualitative method that lets researchers identify patterns of meaning or themes across interviews while staying transparent about their own perspectives shaping the work. And here's a piece I really appreciated about how this study was set up. The team explicitly states their position: they note that they're guided by a neurodiversity affirming perspective. They state, and I'm paraphrasing, that autism itself is not a driver of hate or violence and that autistic people exercise agency in their choices. They had John Elder Robison, an autistic advocate, and Christian Picciolini, or I've also heard it's pronounced peach-ee-oh-leenie, a former extremist on the team helping shape interview guides and review the analysis.

That kind of transparency about the lens you're using really matters, especially in research that could be so easily weaponized. So what did the researchers find? They identified four big themes across both the family and clinician interviews. I'm going to walk you through each one and I'll add some commentary as we go because honestly, some of this is going to feel familiar even to autistic listeners who have nothing to do with extremism.

The vulnerabilities they identified are vulnerabilities a lot of us have brushed up against or are currently dealing with now. Theme one, experiences of social vulnerability. The first theme is about how isolated and unsupported these autistic people had been, often for years before they got pulled in. Family members and clinicians described autistic people who didn't have real life friendships. Some had jobs and even got positive feedback at work, but still felt fundamentally like they didn't fit. They felt totally excluded from life, from social life. They spent enormous amounts of time online by themselves, and the lack of in-person friendships meant there was no one to push back when they encountered something extreme. Family dynamics came up a lot too.

Several parents described what the researchers called fragmented households with disrupted support systems. Some parents named their own guilt, wondering if they'd missed something, wondering if more support earlier could have changed things. One mother in the study described people coming to her house, eating her cookies, drinking lemonade she'd served, and only later realizing those people had been grooming her son.

Then there were the system level failures, and this is where it gets really layered, because the parents and the clinicians had different perspectives on whose fault that was. So some clinicians said parents had dropped the ball, that autism had been flagged in childhood and the family hadn't followed up. Some parents said clinicians had missed the diagnosis, dismissed their concerns, or in one case actually refused to treat their kid because the clinicians politics didn't align with the kids beliefs. The researchers don't pick a side. They make a point I think is really important, that the blame goes back and forth because the system itself is reactive instead of proactive. By the time anyone notices a problem, the problem's already had years to kind of grow and fester. 

And there's this thing called aging out of the system, when childhood autism services end, often right around the transition to adulthood. You've got someone who's suddenly more autonomous, less supported, and potentially more isolated. It's a very vulnerable time. Theme two, autistic and neurodivergent characteristics in the context of risk. Okay, so this is the section that requires the most care because this is where it would be easy to misread the findings as autism causes extremism.

That's not what they found. What they found is that certain autistic traits when combined with other risk factors — isolation, idle time, online immersion, lack of digital literacy, and exposure to a particular kind of content — can make extremist content stickier than it might otherwise be. They identified three sub themes here. So the first is cognitive and social processing differences.

Some of the autistic people described in the study had difficulty with perspective taking. So difficulty understanding views and feelings that differ from their own. And some of them tended toward black and white thinking. Now, extremist ideologies are essentially black and white thinking dressed up as a worldview. They explain complexity away. They give us an easy us and an easy them.

If you're already someone who finds nuance exhausting and craves clear answers, then that simplification can feel like relief, like a breath of fresh air. One social worker in the study made a point I want to highlight. She said that autistic people may be more vulnerable to taking information at face value if they hadn't had explicit instruction in how to distinguish fact from misinformation or fact from conspiracy theory. That's not a deficit. That's a gap in education that we could address if we cared to. The second sub-theme is seeking structure in an uncertain world. Several participants pointed out that engagement with extreme ideologies often started during a period of crisis. Loss of a job, the COVID-19 pandemic, the end of school services, et cetera.

All of these were characterized by routine collapsing. And then the ideology came in and provided a sense of certainty. One clinician noted that hate groups essentially showed up during the pandemic with a message of, here's the answer, here's the black and white of what's going on. If you're autistic and you've ever clung to a routine through a hard time, you probably feel the shape of this pretty easily.

The pull toward predictability is real. The danger is when the predictability you find is being offered by a group whose entire architecture is built on hating other people.

The third sub theme is what the researchers called tunnel vision interests, hyperfixation. And I think this is where a lot of autistic listeners are going to feel something uncomfortable because hyperfixation is also one of our superpowers. The friend who participated in the study described two of his autistic friends who hyperfixated on European history and ancient European warfare. That interest on its own, not a problem.

Loads of people are into military history, but for those two specific people in the context of everything else in their lives, that fixation became a doorway. It led to certain books, certain forums, certain communities, certain people. One clinician in the study said something that stuck with me. She said, quote, that stickiness with a topic and pursuing it for a long time can be such an asset in so many autistic people when it's applied to certain things. But in this case, I think it can really become a liability. End quote. 

Hyper fixation is morally neutral. What it gets pointed at sometimes is not. Theme three, negotiating a complex identity. This is the theme I found the most psychologically interesting and I think the one that has the most to say to autistic listeners broadly, even those who would never go anywhere near an extremist group. The researchers found that autistic people who'd been engaged with extreme ideologies had wildly different relationships with their own autistic identity. Some, often diagnosed early, had integrated autism into their sense of self in a relatively positive way. Others, often diagnosed later, outright rejected the label. And the clinicians attributed that rejection to the deficit-based model of autism that a lot of us grew up with. If the only autistic people you ever met or heard about as a kid were people who fit a 1980s stereotype, which is a stereotype that erased basically every autistic woman, every late diagnosed adult, every high masking person, then of course you're not going to want to identify with that.

The model just didn't include you. So you concluded the label wasn't yours. One clinician in the study said that people who are flirting with extremism are much less likely to have a positive autistic identity. They might claim ADHD or some adjacent diagnosis that they feel more comfortable with, but autism itself, they kept at arm's length. Again, not all of them, but a lot of them.

And listeners, I want you to sit with that for a second. So a negative autistic self image in this study was identified as a risk factor for radicalization.

Not because being autistic is dangerous, but because hating yourself creates a wound that extreme ideologies are remarkably good at exploiting. The second piece of this theme is even more painful. The researchers describe autistic people with multiple marginalized identities. So racialized, queer, gender diverse, who tried to distance themselves from those identities by aligning with groups that opposed them. There's a quote in the paper from a mother whose grandfather had been imprisoned in a Nazi concentration camp during the Second World War. Her son rewrote that family history to claim that the grandfather had been a Nazi sympathizer. He started doing Second World War reenactments as a Nazi.

The clinicians frame this through a concept called reaction formation. There's something about yourself you can't tolerate. So you flip it inside out and externalize it. You push it away. You're like, that's not me. You become aggressively the opposite of what you feel it hurts to be. The researchers also identified two specific identity strategies. One was gaming: autistic people taking on power roles in fantasy worlds, often in games that lets you play as fascists or in dictatorships. The other was the troll identity, like provoking people, wearing that inflammatory hat. We've all encountered them online, making the inflammatory salute, getting strong reactions. The reasoning, as one clinician put it, was that this allowed these autistic people to enact their own rejections, basically, to make rejection something they were choosing instead of something being done to them, imposed on them. It was a way to feel powerful instead of preyed on. And this is heartbreaking. And it's also, if you read between the lines, an extremely autistic logic. So if something's going to happen to me anyway, at least let me be the one who decides when and how.

Theme four, a slippery rabbit hole. This last theme is about how the engagement actually unfolds, the trajectory, not the cause. So clinicians described a really consistent pattern. It almost always started with passive consumption: lurking, reading, learning vocabulary, hanging out on forums, no relationships yet, just taking it in, just absorbing. And then because of the way social media algorithms are designed, the more time someone spent in that space, the more that space the algorithm fed them. One clinician described a client who started with a lot of failed first dates, just normal dating frustration. And then over time, his explanations for those failures shifted and incel material started appearing in how he described his life. Eventually he couldn't describe a single interaction outside of that framework. That's the rabbit hole. It's not a single identifiable bad decision. It's a slow narrowing of available explanations until there's only one left. I also wanted to add  a definition here before we move on in case anyone needs it. And I know I used this term earlier and I forgot to define it. So the term incel comes up a few times in the study.

It's short for involuntary celibate. It refers to a primarily online subculture of men, usually men, who identify as unable to find romantic or sexual partners despite wanting them, and who tend to blame women, feminism, and more conventionally attractive men for that. The community didn't start out this way. It was actually founded in the 1990s by a queer woman as a support forum, and she's since publicly distanced herself from what it became. But over the last decade or so, parts of it have radicalized into deeply misogynistic territory, and incel ideology has been linked to several mass killings, including Elliot Rodger's attack in California in 2014, and Alek Minassian's van attack in Toronto in 2018. Some governments now classify it as a form of violent extremism.

So when you hear the clinicians in this paper talking about incel forums or incel material, that's the world they're describing. The researchers note that for some autistic people, the extreme ideology isn't where the journey starts. It's where the journey ends. They got there through hyper fixation, through algorithmic reinforcement, through having time alone, idle time, lots of time alone, through unmet need for structure, and through wanting to belong.

Once they were inside the community and getting validation, the ideology became identity. And from there, they often moved from passive consumption to active participation: going to rallies, joining organizations, becoming, as one clinician put it, foot soldiers. There are a few things I want to flag from the discussion section of the paper because I think they're really important for how we metabolize all of this.

First, the researchers are extremely clear that autism alone does not lead to extremism. Okay, we have to be really, really clear. They say it explicitly and they say it more than once because they don't want their work to be misused or push the violent autistic stereotype. That stereotype is not supported by the evidence. The vast majority of autistic people will never go anywhere near hate groups.

What this paper describes is a very small population of autistic people whose vulnerabilities line up with predatory ideologies in a particular way. Second, the clinicians in this study described their autistic clients as disproportionately young, white, cisgender, heterosexual men. The researchers are honest about why that pattern shows up. Some of it has to do with who has access to clinical care in the first place.

Some of it likely reflects the demographic patterns of extremism more broadly, which definitely skew young, white, and male, regardless of neurodivergence. There's an intersection of social privilege and neurodivergent marginalization that the authors think deserves a lot more research. And I totally agree. Third, and this is the part I want autistic listeners to really hear: the paper points to identity as a critical leverage point. A negative autistic self-image makes someone more vulnerable in general and specifically here. A positive autistic self-image is, in this framing, protective. The work we do to understand ourselves, to find our community, to push back on the deficit model, to talk about our actual lived experiences, that work is not just personal.

It's preventative. It's saving lives. It builds people who don't need to outsource their identity to a group that will hate on their behalf. Fourth, the paper takes a hard look at the role of digital platforms. Algorithms designed to maximize engagement push vulnerable users, autistic and otherwise, into ecosystems that reward sensational, polarizing, emotionally activating content. The authors call on social media companies to take responsibility, on policy makers to regulate, and on educators to provide neuro inclusive digital literacy training that meets autistic people where we actually are. Fifth, and this one I sat with for a while: the paper documents how parents and clinicians blame each other when an autistic person ends up radicalized.

The researchers don't take sides here. They name it as a system failure, a reactive system, a system that lets families fall through the cracks and then asks them why they didn't catch their kid. If you're a parent listening to this and feeling some prickly feelings, the paper's on your side. It's also on the clinicians' side. It is not on the side of the institutions that left both groups holding the bag. So what this study can't tell us.

So quick note on limitations here. This is a small qualitative study, right? So eight close contacts and five clinicians. It's very small, common for a qualitative study, but you know, the findings are about meaning and pattern, not prevalence. We cannot use this study to say anything about how common any of this is. It's very, very early research in this area. The participants are reporting their perceptions.

Parents in particular may be motivated consciously or not to frame their child as a victim and downplay the agency their child actually exercised. The researchers acknowledge this. They specifically tell readers to interpret parent quotes very carefully. The clinician sample is people who saw clients who had access to professional help. That tilts the data toward people with resources. Autistic people who are radicalized and never make it into a clinical setting are not represented here. And the whole study is North American. Sociopolitical context matters enormously for extremism research. So we can't assume any of this maps onto, say, autistic people in Europe or Asia or anywhere else. This research is the kind of research that scares me a little because it's the kind that can be cherry picked by people who don't read past the abstract. The headlines write themselves. Study finds autistic traits linked to extremism.

That headline would be wrong, but it would definitely get clicks. What the study actually says is more uncomfortable and more useful. It says that some of the things that protect autistic people — community, identity, structure, predictability, deep interests, online connections — can also become entry points for harm when they're not paired with support. 

It says that the people most likely to slide into a hate group are people who never got a real shot at building a life that fits them. 

It says that timely diagnosis matters. 

It says that continuous support across the transition to adulthood matters. 

It says that a positive autistic identity is, among other things, a kind of armor. 

It says that the platforms we spend our time on are not neutral, and that the algorithms shaping our feeds are working on us whether we notice it or not and whether we like it or not. 

And it says quietly but clearly that prevention is not about surveilling autistic people. 

Prevention is about building a world where autistic people don't need to look for belonging in the worst possible places. That world is built in part by autistic people talking to one another, by the kind of conversations we have on this podcast, by community, by telling the truth about what it's like to be us, and refusing the deficit narrative that's been handed to us and forced down our throats, to be honest. So if you've made it this far, thank you for sitting with a hard topic. If anything in this episode resonated in a way that surprised you or unsettled you, that's worth paying attention to.

And if you know an autistic person — maybe a younger one, maybe a teenager, maybe an adult who's been spending a lot of time alone online — this might be a good week to send them a text. I'll link the full study in the show notes. It's open access, so you can read it yourself. Before I sign off, I'd like to thank Yolande and Emma for each buying me three coffees. Thank you both so, so much for your generous support.

If you would also like to buy me a coffee to help this podcast keep going and growing, there's a link in the show notes that says buy me a coffee. Well, that's all I have for you today. Thank you so much for being here. Until next time, bye.