Discursive constructions of autism on social media: a multilingual analysis across five languages

Introduction: Language choices in autism-related social media discourse reflect ideological models and shape public understanding. Although neuroaffirmative and identity-first language is increasingly visible, cross-linguistic patterns on social media remain poorly understood. These platforms provide a valuable setting to observe how autistic identity and competing medical and neuroaffirmative frameworks circulate internationally. Methods: This observational multilingual study analyzed 678 public posts collected through stratified keyword sampling in English, Spanish, French, Norwegian, and Georgian. Posts were normalized, anonymized, and classified using multilingual lexicons and sentence-embedding-assisted disambiguation. Variables included theoretical frame (medical, neuroaffirmative, both, neutral), linguistic style (identity-first, person-first), sentiment (-5 to +5), and engagement metrics. Descriptive statistics and longitudinal trend analyses were conducted. Results: Medical and neuroaffirmative discourse coexisted across languages. Spanish showed the highest medical framing (45.1%), while English (43.4%) and French (36.4%) favored neuroaffirmative discourse; Norwegian and Georgian leaned neutral. Identity-first language predominated overall, with person-first usage most frequent in Norwegian and minimal in Georgian. Sentiment was largely neutral. Engagement varied by language, with identity-first posts generating the highest interaction in English. Neuroaffirmative discourse increased across all languages after 2023. Discussion: Autism discourse on X/Twitter reflects a global but uneven shift toward neuroaffirmative and identity-first language shaped by linguistic and cultural context, positioning social media as a key arena for renegotiating autistic identity and expertise.