Discursive Constructions of Autism on Social Media: A Multilingual Analysis Across Five Languages

The language used to describe autism is far from neutral; it carries implicit assumptions about identity, agency, and societal norms, profoundly shaping public perception. This study delves into the complex linguistic landscape of autism discourse on X (formerly Twitter), comparing how medical and neuroaffirmative frameworks are represented across English, Spanish, French, Norwegian, and Georgian. The findings reveal a global yet uneven shift towards neuroaffirmative and identity-first language, with social media platforms emerging as crucial arenas for renegotiating autistic identity and expertise.

Introduction: The Evolving Language of Autism

The terminology surrounding autism is a dynamic reflection of evolving societal understanding. For decades, a medical model dominated, framing autism primarily as a disorder characterized by deficits requiring diagnosis and treatment. This perspective often led to the use of person-first language, such as "person with autism," which linguistically separates the individual from the condition. However, a growing movement, rooted in the neurodiversity paradigm, reframes autism as a natural and valuable form of human variation. This approach champions identity-first language, like "autistic person," emphasizing autism as an integral part of an individual’s identity rather than a detachable impairment.

Social media platforms have become pivotal spaces for these linguistic and ideological battles. They democratize discourse, allowing autistic individuals, their families, researchers, and advocacy groups to share experiences, challenge medical narratives, and influence public understanding. While professional bodies and journals increasingly advocate for neuroaffirmative and identity-first language, global patterns of its adoption remain under-examined. This research addresses this gap by analyzing multilingual autism discourse on X, aiming to understand how competing frameworks circulate internationally and how language choices shape public perception and the very definition of autistic identity.

Methodology: A Multilingual Snapshot of Online Discourse

This observational study employed a stratified keyword sampling strategy to collect 678 public posts from X in English, Spanish, French, Norwegian, and Georgian. The selection of languages aimed to represent diverse linguistic families and cultural contexts. The research team focused on posts containing terms associated with either the neuroaffirmative model (e.g., "autistic," "be autistic") or the medical/deficit model (e.g., "with autism," "suffering autism"), alongside a neutral term ("autism").

Data collection involved retrieving the last 25 posts (as of September 17, 2025) for each keyword in each language. This approach, while not aiming for statistical representativeness of all X content, allowed for a controlled, cross-linguistically comparable corpus for detailed discursive analysis. Posts were then normalized, anonymized, and classified using multilingual lexicons and sentence-embedding assisted disambiguation. Key variables analyzed included:

• Theoretical Frame: Whether posts primarily reflected a medical model, a neuroaffirmative model, both, or were neutral.
• Linguistic Style: Use of identity-first language, person-first language, both, or neither.
• Sentiment: A score from -5 (highly negative) to +5 (highly positive), assessing the affective tone of the post.
• Engagement Metrics: Number of replies, reposts, likes, and views, indicating the reach and interaction generated by the content.

Descriptive statistics and longitudinal trend analyses were conducted to identify patterns and evolutions in the discourse.

Results: Divergent Narratives Across Languages

The analysis revealed a complex interplay of medical and neuroaffirmative discourses across the five languages, with notable variations in framing and linguistic style.

Discourse Models: A Spectrum of Perspectives

• Spanish exhibited the highest proportion of posts framing autism through a medical lens (45.1%), aligning with previous research suggesting a continued prevalence of deficit-based metaphors in Spanish-language autism coverage.
• English (43.4%) and French (36.4%) showed a stronger leaning towards neuroaffirmative discourse, reflecting the growing influence of autistic self-advocacy movements in these linguistic communities.
• Norwegian and Georgian discourse leaned more towards neutral framing, suggesting a public conversation that is perhaps less polarized between medical and affirmative ideologies, or still in earlier stages of adopting these nuanced discussions.
• Interestingly, the "Both" category, where medical and neuroaffirmative frames coexisted, was more prominent in French (9.2%) and Georgian (12.0%) than in English or Spanish. This could indicate a more complex negotiation of these ideas within those specific online communities.

Linguistic Styles: Identity-First on the Rise

Overall, identity-first language predominated across the corpus. This aligns with the strong preference reported by many autistic adults for language that integrates autism into their identity.

• Person-first usage was most frequent in Norwegian (28.7%) and French (26.6%), suggesting a lingering influence of traditional person-first conventions in these contexts.
• Georgian showed a striking prevalence of posts categorized as "None" (66.7%), indicating a significant portion of discourse that refers to autism without explicitly employing either identity-first or person-first formulations, perhaps relying on more general or descriptive language.
• The "Both" category for linguistic styles was minimal, appearing only in English, Spanish, and French posts, suggesting that while conceptual frames can coexist, explicit mixing of identity-first and person-first phrasing within a single post is less common.

Sentiment and Engagement: Mixed Signals

Sentiment analysis indicated that posts across all models and styles largely clustered around neutrality. However, some variations emerged:

• The "Both" model in English registered the highest positive sentiment.
• The "Medical" model in French leaned slightly negative, while its counterpart in Georgian remained strictly neutral.
• Identity-first language in English also showed a positive tendency, while its Georgian equivalent leaned slightly negative.
• Person-first language across most languages displayed a slightly positive sentiment.

Engagement metrics presented a varied picture:

• In English, identity-first posts generated the highest interaction (likes, reposts), underscoring the resonance of self-affirming language within that community.
• In French, person-first language posts, particularly those with a medical or neutral framing, garnered substantial engagement, suggesting that institutional or informational content may hold significant sway.
• Spanish discourse showed higher engagement with medical and neutral frames, while neuroaffirmative content also saw notable interaction.
• Norwegian and Georgian generally exhibited lower interaction levels across models and styles, with no consistently dominant pattern, though identity-first language showed a slight advantage in Norwegian.

Temporal Evolution: A Shifting Landscape

A temporal analysis, examining posts from 2011 to 2025, revealed a discernible trend across all languages:

• Erosion of Medical Hegemony: The historical dominance of the medical model is gradually giving way to greater discursive diversity. While medical framing remains significant, especially in Spanish and Georgian, the neuroaffirmative model has steadily gained ground since approximately 2023.
• Rise of Neuroaffirmative Discourse: The neuroaffirmative perspective is demonstrably expanding, becoming a more visible and consolidated frame in recent years across English, French, Spanish, and Georgian.
• Consolidation of Identity-First Language: Parallel to the shift in theoretical models, identity-first formulations have progressively increased their prominence, although they continue to coexist with person-first expressions and other linguistic styles.
• Pluralization of Discourse: The overall trend points towards a more pluralistic discursive space, where mixed conceptual frames and neutral or descriptive approaches also find a place.

This temporal evolution suggests that public discourse on autism is actively being reshaped by evolving advocacy efforts and the increasing visibility of autistic voices online.

Discussion: Social Media as a Crucible for Identity and Understanding

The findings of this study underscore the profound impact of language on the public understanding of autism and the negotiation of autistic identity. The coexistence of medical and neuroaffirmative frameworks, alongside the rise of identity-first language, reflects a global, albeit uneven, paradigm shift. Social media platforms like X are not merely passive conduits of information but active sites where these competing ideologies are debated, disseminated, and internalized.

Cross-Linguistic Nuances and Cultural Contexts

The observed differences in discourse models across languages highlight the influence of local contexts. The continued strength of medical framing in Spanish, for instance, may be linked to specific cultural norms, media ecosystems, or the maturity of autistic self-advocacy movements within Spanish-speaking regions. Conversely, the stronger presence of neuroaffirmative discourse in English and French reflects the well-established influence of these movements in Anglophone and Francophone spheres. The more neutral stance in Norwegian and Georgian suggests a different stage in the public discourse, potentially characterized by a greater focus on awareness and inclusion rather than a strong ideological divide.

The preference for identity-first language, while dominant overall, also shows linguistic variations. Its stronger presence in Norwegian alongside person-first language suggests a nuanced adoption of preferred terminology. The significant proportion of "None" in Georgian discourse warrants further investigation, potentially pointing to unique linguistic structures or a different cultural approach to describing conditions.

Implications for Wellbeing and Social Inclusion

The language used to discuss autism has tangible implications for the wellbeing of autistic individuals. Deficit-based language and medicalized framings can contribute to stigma, misunderstanding, and poorer mental health outcomes. Conversely, identity-affirming language and neurodiversity-centric narratives can foster self-acceptance, a sense of belonging, and collective empowerment. Social media, by providing a platform for these self-advocacy narratives, plays a crucial role in creating supportive online communities and challenging dominant, often pathologizing, representations. The higher engagement observed for identity-first posts in English, for example, suggests that such language resonates strongly and effectively circulates within that linguistic community.

A Global but Culturally Translated Neurodiversity

While the study points to a global trend towards neuroaffirmative discourse, it is essential to recognize that neurodiversity is not a monolithic concept. Its interpretation and adoption vary significantly across cultures. The way neurodiversity is translated and implemented in different linguistic and socio-political contexts shapes its meaning and impact. This "cultural translation" means that while the underlying principles may be similar, the manifestations and reception of neurodiversity can differ substantially worldwide.

Methodological Considerations and Future Directions

This study, while providing valuable insights, is subject to certain limitations. Social media data is inherently influenced by sampling biases, algorithmic filtering, and unequal platform access. The use of specific keywords means that the findings represent patterns within the defined search space, rather than an exhaustive depiction of all autism discourse on X. Future research could benefit from triangulating findings across multiple platforms (e.g., TikTok, Reddit) and incorporating qualitative methods like interviews to gain a deeper understanding of individuals’ conscious language choices and their motivations. Further exploration into the role of AI systems in mediating autism discourse is also warranted, given their increasing prevalence.

Conclusion: A Continual Redefinition

This multilingual analysis of autism discourse on social media demonstrates that language is a powerful tool in reshaping our understanding of autism. The observed shift from predominantly medical framings to a more prominent neuroaffirmative and identity-first discourse across various languages signifies an ongoing redefinition of autism. Social media platforms serve as critical battlegrounds and incubators for these evolving narratives. Recognizing and respecting the linguistic and cultural nuances of these discourses is paramount for fostering a truly inclusive, cross-cultural understanding of neurodiversity and promoting the wellbeing of autistic individuals globally.