Free Autism Quotient Test

The Autism Quotient (AQ) is not a single test but a family of related questionnaires, all built by Professor Simon Baron-Cohen's team at the Autism Research Centre in Cambridge. If you have seen the term "AQ score" mentioned online or by a clinician, this page explains what that family actually consists of and lets you take the full version.

This page embeds the complete AQ-50, the original 50-question instrument the AQ family is built around. If you want the shortest possible screening instead, our dedicated AQ-10 page covers the same ground in about 2 minutes.

What "Autism Quotient" Actually Refers To

The term traces back to a single 2001 paper by Simon Baron-Cohen, Sally Wheelwright, Richard Skinner, Joanne Martin, and Emma Clubley in the Journal of Autism and Developmental Disorders, which introduced a 50-item questionnaire measuring how strongly an adult of average intelligence displays autistic traits. Eleven years later, in 2012, Carrie Allison, Bonnie Auyeung, and Simon Baron-Cohen published a follow-up paper that selected the 10 most discriminating items from that original 50 to create the AQ-10 for use where time is short. Both papers share the same underlying trait model, which is why people use "AQ" loosely to refer to either version, or to the concept of autistic-trait scoring in general.

The Five Trait Domains Behind Every AQ Score

Whichever version you take, the AQ family is built around five domains, each represented by 10 questions in the AQ-50 and by one or two questions in the AQ-10:

• Social Skills: Comfort in group settings, ability to read social atmospheres, and preference for social versus solitary activities.
• Communication: Conversational reciprocity, understanding of figurative language, and comfort with small talk.
• Imagination: Engagement with fiction, history with pretend play, and ability to imagine others' perspectives.
• Attention to Detail: Pattern recognition, noticing small changes, and focus on specifics, often a strength.
• Attention Switching: Ease of transitioning between tasks, tolerance for interruptions, and flexibility with changing plans.

How We Score the AQ-50 on This Page

The version embedded above is the full 50-question AQ, not the AQ-10. Each item is answered on a four-point scale from Definitely Agree to Definitely Disagree, and a response in the autistic-trait direction earns one point, regardless of whether that direction is agreement or disagreement for that specific item. Your overall total runs from 0 to 50, and each of the five domains above is also scored on its own out of 10. Our tool treats a total of 26 or above as meeting our threshold, and a total of 32 or above as an additional higher band. That second number, 32, matches the threshold used in the original 2001 AQ-50 paper; the 26 figure is our own earlier flag, not something drawn from the original research. If you want a full breakdown of that distinction and why we chose it, see our dedicated AQ-50 page and our method.

What the Original Research Found

The 2001 AQ-50 paper set 32 or above as the score indicating clinically significant traits, based on comparisons between adults with a diagnosis of Asperger syndrome or high-functioning autism and adults without one. The 2012 AQ-10 paper, built from the same item pool, set 6 or above as its own referral threshold for the shortened version, a recommendation that has since been incorporated into NICE guidance in the UK for adult autism referral.

Why the Name "Autism Quotient" Sticks Around

The word "quotient" invites comparison to an IQ score, a single number that seems to sum up something complex. That framing is partly why the AQ became so widely referenced outside academic research: it offers a concrete number people can share, compare, and search for online. But the researchers who built it were studying trait distribution across a population, not building a personal identity label. A number from either version describes where your self-reported answers fall relative to the items on that specific questionnaire, on that specific day. It is not a fixed quantity you carry around the way a birth date or a blood type is fixed, and it can shift if you retake the same questionnaire at a different point in your life, in a different mood, or with a different understanding of what a question is actually asking.

That said, the AQ's popularity is not accidental. Because both versions are short enough to complete in a browser and specific enough to point toward five concrete domains, they have become a common first step for people who are curious about autism but not yet ready to seek a formal evaluation. Used that way, as a prompt for further reading and conversation rather than a final answer, the AQ family does what it was designed to do.

What Neither Version Can Tell You

• Neither the AQ-50 nor the AQ-10 is a diagnosis. Both are self-report screening tools, and a score in either direction is a starting point, not a conclusion.
• Neither gathers developmental history. A formal evaluation looks at how traits have shown up since childhood; a questionnaire answered today cannot do that on its own.
• Neither distinguishes autism from overlapping conditions. Anxiety, ADHD, and OCD can each raise scores on some items without the person being autistic.
• Both depend on self-insight, which is exactly what masking can make difficult to access, particularly for people who have spent years learning to appear more typical than they feel internally.

Who Should Use This Page Versus the AQ-10

Use this page, and take the full AQ-50, if you want to see where your traits sit across each of the five domains individually, rather than a single combined number. This is especially useful if you already suspect that one specific area, such as attention switching or communication, is where most of your day-to-day friction comes from, since the AQ-50 gives that domain 10 questions of its own instead of one. If you would rather get a rough answer in two minutes and decide from there whether to go deeper, start with the AQ-10 instead. Neither the AQ-50 nor the AQ-10 was designed for screening young children; for toddlers, see the M-CHAT-R-based screening for kids.

Researchers and clinicians often talk about "AQ score" as shorthand without specifying which version they mean, and that can make it hard to tell whether a number you have seen quoted somewhere, in an article, a forum post, or a book, refers to a possible 0 to 10 or a possible 0 to 50. If you are trying to compare your own result against something you read elsewhere, check which version the source is describing before assuming the numbers line up. A 6 on the AQ-10 and a 26 on the AQ-50 are not directly comparable just because both are described loosely as "an AQ score above threshold."

The same caution applies to comparing your own results across time. If you take the AQ-10 today and the full AQ-50 next month, do not expect the two totals to convert into each other cleanly. They ask about the same domains but with very different numbers of questions per domain, so small differences in how you interpreted a given question can move the two scores in different directions even when your underlying experience has not changed.

What a "Meets" or "Below" Result Means

On the AQ-50 embedded here, a total of 26 or above meets our threshold, and 32 or above matches the original study's own cutoff. Either result means your answers describe a level of traits worth exploring further, whether through the longer RAADS-R or a conversation with a professional. A total below 26 means your answers did not reach our threshold on this occasion, which does not rule autism out on its own, particularly if a small number of items felt hard to answer honestly or if one domain stood out even while your overall total stayed low. See how we explain results for a walk-through of every band on every screening we offer.

Frequently Asked Questions