Autism and Mapping the Evidence on Online Risk: A Scoping Review and Narrative Synthesis

Background: Digital platforms have increased exposure to potentially harmful online content, including misinformation, conspiracy theories, hateful material, and extremist or radicalization-related material. Research examining how autism relates to these domains remains sparse, fragmented, and poorly integrated across disciplines. Methods: We examined quantitative, qualitative, and adjacent literature to map how autistic characteristics have been studied in relation to harmful online content, with findings synthesized narratively. Results: Quantitative studies were limited in number and relied largely on cross-sectional belief measures, providing little insight into online exposure or change over time. Qualitative empirical research offered detailed clinical and forensic accounts but was mostly case-based, retrospective, professional, online discourse, lived-experience, or family/clinician evidence, limiting empirical testing of proposed explanations and author reflexivity. Adjacent empirical literature examined relevant cognitive and social processes without linking them to autistic populations. Conclusions: Existing frameworks and clinical interpretations continue to be ahead of the available empirical evidence. Current findings do not support a simple pathway from autism to harmful online engagement, but suggest that risk may emerge under particular combinations of personal, social, cognitive, and online conditions. Progress requires integrated quantitative and qualitative research that uses ethically appropriate evidence sources, tests these conditional pathways, and refines frameworks in line with empirical findings, clinical realities, and safeguarding responsibilities. Community Brief Why is this an important issue? Social media can provide a valuable environment for connection, information sharing, and community participation. For many autistic people, online spaces can offer greater flexibility than face-to-face settings because they can choose when to respond, how to respond, and connect around shared interests. At the same time, these spaces can pose risks. Social media varies in the quality of the information presented, often providing false content, hostile comments, and harmful posts. It can also present information without a clear context, or in ways that make it hard to judge what is trustworthy. By understanding autistic adults’ experiences of social media, we can develop approaches that support online safety and protection while respecting autonomy, promoting informed decision-making, and preserving the benefits and opportunities that online communities provide. What was the purpose of this review? Researchers have studied autism and harmful online content across a range of fields. This shows the topic matters to many different researchers, but it also makes it hard to see the gaps in our current knowledge about autistic adults’ experiences, and how findings from different studies relate to one another. There was a need to provide a clearer understanding of autistic adults’ experiences of harmful online content, identify areas where further investigation is needed, and highlight priorities for future research, policy, and supports that promote both online safety and meaningful participation in digital spaces. What did the authors do? The authors conducted a scoping review. A scoping review maps the existing research on a topic and shows where researchers still lack clear evidence. This review focused on research that examined autism or autistic traits in relation to misinformation, extremist content, and other harmful online material. The authors brought together the existing research to identify common themes, key patterns, and important areas that future research should address. What studies did the authors find? We included 25 studies and four additional related studies from other areas of research that are useful to this area of work. Of the 25 main studies, 16 used surveys or experiments to collect statistics. Nine used interviews, case reports, or other methods that collected people’s own words and experiences. The survey-based studies mostly looked at beliefs and how people process information in either the general population or clinical groups. The interview-based studies focused on clinical, legal, or practitioner experiences, often by looking back at past cases. The four related studies looked at thinking and social factors that may relate to autistic traits, even when they did not directly include autistic people. In summary, what did those studies show? Some studies used surveys or tasks to look at people’s beliefs and how they judge information. These studies suggest that autistic traits may relate to how some people judge information, how confident they feel, and how they respond when they see new information. However, most of these studies did not look at people’s real online behavior. Other studies looked at real-world examples, professional experience, family views, or online communities. These studies suggest that harmful online engagement is rarely about autism alone. It may also involve loneliness, exclusion, support needs, identity, personal interests, grievance, and online content that feels clear, certain, structured, or linked to belonging. The review did not find evidence that autism, on its own, makes someone more likely to engage with harmful online content. Instead, online risk seems to depend on the person, the content, and the wider situation. What are the remaining gaps in the research? Most studies focused on what people believe or how they thought they behave, rather than their actual behavior. We still know relatively little about how autistic adults engage with potentially harmful online content in practice, including whether they search for such content, create posts, share material, comment on discussions, join online groups, or interact with others around harmful content. Some studies asked people to look back and describe past events, or reviewed past cases, rather than following people over time. This makes it hard to know what actually caused what. In addition, some related studies did not directly include autistic participants, meaning their findings should be applied to autistic communities with caution. Also, some related studies did not directly include autistic participants, meaning their findings should be applied to autistic communities with caution. What we do and experience online is also likely to be affected by what we do offline, such as social connection, exclusion, interpersonal conflict, stress, relationships, and unmet support needs. However, most studies did not look closely enough at how a person’s online and offline life affect each other. This review also has several limitations. As a scoping review, its purpose was to bring together and map the findings of the research available, rather than judge how reliable or high quality each study’s findings are. The included studies also varied considerably in their methods, participant samples, and types of evidence, making direct comparisons difficult. Together, these gaps highlight the need for future research to be more specific about who is being studied, the types of online content being examined, and the forms of support that may help autistic adults engage safely and meaningfully in online spaces. Based on this review, what do the authors recommend? Future research should examine in greater detail how autistic adults engage with online spaces, how their experiences change over time, and what makes some online content more difficult or harmful to navigate. Researchers should also be clearer and explain exactly what they are studying. Support should not just focus on reducing risk. It should also help people feel connected, feel less alone, get clear and reliable information, and take part in online spaces in ways that are both safe and meaningful. To make this happen, researchers, clinicians, community organizations, autistic people, and policymakers will need to work together. This will help make sure future research is useful in real life, and can help guide better support and clinical decisions—including for more complex situations.