The gap
The instruments used to assess autism, ADHD, sensory processing, and other neurodevelopmental conditions were overwhelmingly developed in English, normed on Western (often American) populations, and validated within the cultural assumptions of the societies that produced them. Translating these instruments into other languages is difficult. Validating those translations properly is expensive and slow. The result: large parts of the world’s population are assessed using tools that have not been tested for their language, their culture, or their norms of typical behaviour. Where validated translations do exist, they are concentrated in a small number of European and East Asian languages. Most of the world’s 7,000 languages have no validated neurodevelopmental assessment instruments at all.
This is not a minor methodological footnote. It shapes who gets identified, who gets missed, what “autism” or “ADHD” looks like in the research literature, and whose experience counts as evidence.
What proper validation requires
Translating an assessment instrument is not the same as validating one. The gold-standard process, described by Beaton et al. (2000) and updated by the International Test Commission (2017), involves multiple stages: independent forward translations by at least two bilingual translators, synthesis of those translations, independent back-translations by translators unfamiliar with the original, expert committee review for semantic, idiomatic, experiential, and conceptual equivalence, cognitive debriefing with target-population participants, and a field test of 30–40 people minimum. After all that, the translated instrument needs its own psychometric validation: internal consistency, test-retest reliability, inter-rater reliability, and criterion validity, ideally with both neurodivergent and neurotypical samples from the target population. Cut-off scores may need recalculation, because what counts as “atypical” depends on what is typical in a given culture.
A full validation programme for a single instrument in a single language can take years and cost tens or hundreds of thousands in research funding. Most instruments are protected by copyright, adding licensing costs. For clinician-administered tools like the ADOS-2, validation also requires training clinicians in the target language to a reliable standard. The barriers are cumulative.
The current landscape
Diagnostic instruments (autism)
The ADOS-2 and ADI-R are considered the “gold standard” for autism diagnosis globally. The ADI-R has been translated into roughly 17 languages including Dutch, German, French, Japanese, Korean, Spanish, and Swedish, with published validation studies in several of these. The ADOS-2 has fewer validated translations. Published validation studies exist for Polish, Korean, Brazilian Portuguese, and Afrikaans versions, but many language versions in clinical use lack formal psychometric validation. The ADOS-2 also presents a specific cultural problem: its tasks assume particular social conventions (pointing, showing, turn-taking games) that may not be universal. A 2022 study found that while the ADOS showed high sensitivity across cultures, specificity was poor, producing high false-positive rates in non-Western populations.
The Social Responsiveness Scale (SRS-2) is the most widely translated screening instrument, with over 20 official foreign-language versions including German, Spanish, Vietnamese, Malay, Urdu, Mandarin, Farsi, and a British Sign Language adaptation. Validation quality varies: the German version has extensive psychometric data from over 1,400 participants; the Vietnamese version was one of the first adaptations for a low- or middle-income country.
Sensory processing instruments
The Sensory Profile 2, the most widely used sensory processing assessment globally, has formally validated versions in Greek, Persian, Polish, Malay, Hebrew, Arabic, Brazilian Portuguese, and Korean, among others. Several of these are adaptations of the original Sensory Profile or the Short Sensory Profile rather than the SP-2 itself. Crucially, the Dutch-speaking world (the Netherlands, Flanders, Suriname) lacks a formally validated Dutch version with published psychometric data, despite the Netherlands’ strong tradition of autism research and the existence of well-developed Dutch practice frameworks like the prikkelprofiel. Practitioners use informal translations without published reliability or validity data.
The Sensory Processing Measure (SPM-2), the Sensory Experiences Questionnaire (SEQ), and the Multidimensional Assessment of Interoceptive Awareness (MAIA) have fewer validated translations. The SEQ, designed specifically for autism populations, has no validated non-English versions to our knowledge.
ADHD instruments
A landmark 2024 study by Lewczuk et al. tested the Adult ADHD Self-Report Scale (ASRS) Screener across 42 countries with over 72,000 participants. It found the instrument had good reliability but only partial measurement invariance across languages and cultures. Scores were systematically highest in English-speaking Western countries and lowest in East Asian countries, raising the question of whether the instrument measures the same construct cross-culturally, or whether it measures something shaped by the cultural context in which it was developed.
Most ADHD behaviour rating scales are normed on US populations and rarely adapted for other cultural settings. Different cultures interpret the frequency and intensity of behaviour captured by rating scale descriptors differently. A child’s activity level that triggers a clinical score in one culture may fall within normal range in another.
Where the gaps are widest
The pattern is stark. Instruments have been validated primarily in Western European languages, plus Korean, Japanese, and Brazilian Portuguese. The regions most underserved are sub-Saharan Africa, South and Southeast Asia, the Middle East, and Latin America outside Brazil.
A 2022 review by Bauer et al. found that the majority of validated autism tools present “a multitude of barriers” for use in low- and middle-income countries, including cultural inappropriateness, prohibitive copyright and licensing costs, and the absence of trained professionals. Divan et al. (2021) reviewed data from over fifty countries and found that the lack of culturally appropriate screening and diagnostic instruments was a primary barrier to autism identification in these regions. The consequences are direct: autism prevalence estimates from low- and middle-income countries are substantially lower than from high-income countries, not because autism is rarer, but because the tools to find it do not work there.
Why this matters beyond methodology
The cross-cultural validation gap reinforces a pattern described on the intersectional neurodiversity page: the Western, white, male, English-speaking default in neurodevelopmental research. When instruments are developed in one culture and exported without proper validation, they carry their assumptions with them. What counts as atypical eye contact, atypical social reciprocity, or atypical sensory response is culturally mediated. A child who avoids eye contact with adults may be autistic in one cultural context and well-mannered in another. An instrument that cannot distinguish between these is not measuring autism; it is measuring deviation from one culture’s norms.
This has practical consequences at every level. Clinicians working in non-English-speaking settings face a set of poor choices: administer English-language tools to people who may not be fully proficient; use informal translations without psychometric backing; rely on clinical observation alone; or import cut-off scores from populations that may not share the same behavioural norms. Each workaround introduces risk of misidentification, missed needs, and poorly targeted support.
For autistic people with intellectual disability, who are already underserved by existing assessment tools in any language, the cross-linguistic gap compounds an already significant barrier.
What would help
No single study will close this gap. What is needed is structural.
The copyright and licensing costs protecting commercial instruments like the ADOS-2 and Sensory Profile 2 are a major barrier to cross-cultural adaptation, particularly in low-resource settings. Open-source instruments offer one way forward. The CHATA project (Children with Autism Technology Enabled Assessment), which is developing an open-source diagnostic pathway designed for cultural and linguistic adaptability, represents a model for how this could work.
Cross-cultural adaptation is labour-intensive and unglamorous. It rarely attracts the same funding as novel discovery research. National research councils and international bodies could treat validation studies as research infrastructure rather than an afterthought.
Translation alone is insufficient. A properly adapted instrument accounts for different social norms, different developmental expectations, different relationships between child and adult behaviour. This requires collaboration with local researchers and communities, not just bilingual translators.
If resources are limited, validation efforts should prioritise languages spoken by the largest underserved populations. Arabic, Hindi, Mandarin Chinese, Swahili, and Bengali would collectively serve billions of people currently without access to validated neurodevelopmental assessment in their own language.
Key sources
- Beaton, D.E. et al. (2000). Guidelines for the process of cross-cultural adaptation of self-report measures. Spine, 25(24), 3186–3191.
- International Test Commission (2017). ITC Guidelines for Translating and Adapting Tests (2nd ed.).
- Lewczuk, K. et al. (2024). Cross-cultural adult ADHD assessment in 42 countries using the Adult ADHD Self-Report Scale Screener. Journal of Attention Disorders, 28(4), 569–583.
- Bauer, K. et al. (2022). Autism assessment in low- and middle-income countries: feasibility and usability of western tools. Journal of Autism and Developmental Disorders.
- Divan, G. et al. (2021). Annual research review: achieving universal health coverage for young children with ASD in low- and middle-income countries. Journal of Child Psychology and Psychiatry, 62(5), 514–531.
- Tafla, T.L. et al. (2024). Autism spectrum disorder diagnosis across cultures: are diagnoses equivalent? Autism & Developmental Language Impairments, 9.
- De Leeuw, A. et al. (2024). Screening tools for autism in culturally and linguistically diverse paediatric populations: a systematic review. BMC Pediatrics, 24, 644.