HealthScorer

Autism Test (AQ-10) — Adult Screening

Validated AQ-10 autism screening test for adults. Allison, Auyeung & Baron-Cohen (2012). 10 items, 2 minutes, NICE-recommended brief screen with.

Last updated: Sources verified:

How AQ-10 is calculated

AQ-10 contains ten short statements about everyday experiences. You respond on a four-point scale: definitely agree, slightly agree, slightly disagree, definitely disagree. Each item scores either 0 or 1 — a binary point if your response matches the autistic-trait direction.

Items 1, 7, 8 and 10 score 1 if you agree (slightly or definitely). Items 2, 3, 4, 5, 6 and 9 score 1 if you disagree (slightly or definitely). The total range is 0 to 10. NICE recommends referral at a score of 6 or above.

The AQ-10 was developed in 2012 by Carrie Allison, Bonnie Auyeung and Simon Baron-Cohen at the Autism Research Centre in Cambridge. It is a shortened, validated version of the original 50-item Autism Spectrum Quotient (AQ-50, Baron-Cohen 2001). The brief version preserves diagnostic accuracy while taking only two minutes.

What your score means

A score below 6 means clear autistic traits did not show up in this short check. Autism remains less likely — but is not ruled out, especially for women, non-binary people and adults who have learned to mask traits.

A score of 6 or 7 is a positive screen. NICE recommends referral for full assessment at this level. About 1 in 40 adults meets criteria for autism — you are not unusual.

A score of 8 or higher is a strongly positive screen. Most people in this band who proceed to full assessment receive a confirmed diagnosis. The probability of confirmation is high, but only an assessment can confirm it.

When AQ-10 is useful — and when it is not

AQ-10 works well as a first-step screen for adults curious about autism. It works best if you have not yet spoken to a professional but want a structured starting point.

It works less well for late-diagnosis adults whose patterns are subtle. Female autism patterns can look like anxiety, perfectionism or “always been a bit shy” — and get missed. Non-binary people and adults who learned to mask traits often score lower than fits their experience. If your gut says something is going on but the score is low, trust your gut and consider AQ-50 or RAADS-R.

It is not designed for people with intellectual disability, severe mental illness in acute phase, or anyone under 16. For under-16s, the AQ-Child or AQ-Adolescent are better fits.

Autism vs ADHD or anxiety

Autism, ADHD and anxiety frequently overlap. About half of autistic adults also have ADHD. Many also have anxiety, depression or OCD as co-occurring conditions. Sensory differences run across all of them.

A positive AQ-10 does not mean ADHD or anxiety are absent. It also does not mean autism is the only explanation for your experience. A specialist can sort out which conditions are present and how they interact. Many adults receive a combined autism plus ADHD diagnosis after assessment.

PHQ-9 (depression screen) and GAD-7 (anxiety screen) are useful companions if you suspect overlap. PSS-10 measures perceived stress, which often runs high in undiagnosed autistic adults. None of these replaces a specialist assessment — they help you bring better questions to one.

Privacy

All calculations run in your browser. We never see, log, or store your individual answers. Only an anonymous event is sent to a privacy-respecting analytics service.

Licence and source

AQ-10 is freely available for clinical and research use under the original Cambridge Autism Research Centre release. Source: J Am Acad Child Adolesc Psychiatry 2012;51(2):202-12. PubMed: 22265366. NICE recommendation: CG142.

Frequently asked questions

What does AQ-10 measure?
AQ-10 is a brief 10-item screening tool for autistic traits in adults of average or above-average intelligence. It checks five core areas: attention to detail, attention switching, communication, imagination, and social skills. NICE (UK) recommends it as a first-step screen. A score of 6 or above suggests referral for full assessment.
Is AQ-10 a diagnosis?
No. AQ-10 is a screen, not a diagnosis. A positive screen means traits are present at a level that warrants specialist assessment. Diagnosis requires a clinical interview by a psychiatrist or psychologist trained in adult autism, often using ADOS-2, DISCO or developmental history with informants.
What is a positive AQ-10 score?
Six or higher. NICE recommends referral for full assessment at a score of 6 or above. Scores of 8–10 are strongly positive — most people in this band who proceed to full assessment receive a confirmed diagnosis. A score below 6 makes autism less likely but does not rule it out, especially in women, non-binary people and adults who have learned to mask.
Can I use AQ-10 if I am a woman or non-binary?
Yes — but be aware female and non-binary autism patterns can score lower than they should. The AQ-10 was validated mostly on adults who already suspected autism. If your score is below 6 but you recognise yourself in many items, consider the longer AQ-50 or RAADS-R, both of which capture some patterns AQ-10 misses.
Does my data leave the device?
No. All calculations run entirely in your browser. Only an anonymous event (locale, severity band) is sent to a privacy-respecting analytics service. Your individual answers never leave the device.

More health calculators

Calculators →

Sources

  1. Toward Brief 'Red Flags' for Autism Screening: The Short Autism Spectrum Quotient and the Short Quantitative Checklist for Autism in 1,000 Cases and 3,000 Controls — Allison, Auyeung & Baron-Cohen, J Am Acad Child Adolesc Psychiatry 2012 (peer reviewed, retrieved 2026-05-02)
  2. The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians — Baron-Cohen et al., J Autism Dev Disord 2001 — origin of the AQ-50 (peer reviewed, retrieved 2026-05-02)
  3. Autism spectrum disorder in adults: diagnosis and management (CG142) — NICE — UK National Institute for Health and Care Excellence (guideline, retrieved 2026-05-02)