Adult Autism Screening
Understand yourself
a little better.
This is the AQ-10 — a validated ten-question screening tool used by clinicians worldwide. Your answers are scored against a calibrated model. The result is an indicator, not a diagnosis.
Step 1 of 11
A little about you
Used only to add demographic context to your score. Not stored anywhere.
I often notice small sounds when others do not.
Sensory sensitivityWhen I'm reading a story, I find it difficult to work out the characters' intentions.
Theory of mindI find it easy to "read between the lines" when someone is talking to me.
Social inference · reverse-scoredI usually concentrate more on the whole picture, rather than the small details.
Attention to detail · reverse-scoredI know how to tell if someone listening to me is getting bored.
Social awareness · reverse-scoredI find it easy to do more than one thing at once.
Attention switching · reverse-scoredI find it easy to work out what someone is thinking or feeling just by looking at their face.
Facial affect · reverse-scoredIf there is an interruption, I can switch back to what I was doing very quickly.
Cognitive flexibility · reverse-scoredI like to collect information about categories of things (e.g. types of cars, birds, trains).
Restricted interestsI find it difficult to work out people's intentions.
Theory of mind–
–
Calibrated logistic model · AQ-10 clinical threshold = 6/10
Question breakdown
About this screening tool
This quiz is built around the Autism Quotient 10-item (AQ-10) questionnaire, a validated brief screening instrument developed at the Autism Research Centre, Cambridge. It is used worldwide by clinicians as a first-stage screen for adults who may be on the spectrum.
What the questions measure
- Sensory sensitivity — noticing sounds or details others miss
- Theory of mind — reading intentions, understanding characters in stories
- Social awareness — reading faces, knowing when someone is bored
- Attention flexibility — switching tasks, handling interruptions
- Restricted interests — collecting categorical information
Several items are reverse-scored per the AQ-10 convention — for those, disagreeing contributes to the score.
How the model works
Each AQ item contributes equally. The underlying model is a logistic regression:
P = sigmoid(1.5 × AQ_score + demographic_adjustments − 9.0).
With no demographic adjustments, a score of exactly 6 produces a 50% probability — matching the clinical threshold precisely.
Small adjustments are added for jaundice at birth (+0.30) and family history of autism (+0.50).
Interpreting your result
- Low (<25%) — AQ 0–4. Few ASD-like traits detected.
- Moderate (25–50%) — AQ ~5. Borderline; some traits present.
- High (50–75%) — AQ 6–7. Above clinical threshold; consider a referral.
- Strong (>75%) — AQ 8+. Strong indicator; discuss with a specialist.
Full analysis
EDA, model comparison, and feature importance generated from the dataset.