Adaptive questioning
Each question is chosen in real time based on your previous answer using a Computerised Adaptive Testing algorithm — the same approach used in clinical assessments.
25 adaptive questions. IRT scoring engine. Real percentile data against a 50,000-person norm group. No inflated scores, no email required, no data collected. Just honest results.
Why AurorIQ
Most free IQ tests inflate your score to make you feel good and share results. We built AurorIQ to do the opposite.
Each question is chosen in real time based on your previous answer using a Computerised Adaptive Testing algorithm — the same approach used in clinical assessments.
Item Response Theory — the same statistical framework used in the WAIS-IV and SAT. Your score reflects question difficulty, not just how many you answered correctly.
All scoring runs entirely in your browser. No data is sent to our servers. No account, no email, no tracking — your results exist only on your device.
We publish our confidence intervals, known biases, and what the test cannot measure. No online test is a clinical assessment — we say so, plainly.
Percentiles calculated against 50,000+ real test-takers — not a theoretical distribution. When we say 87th percentile, we mean 87th percentile.
Pattern recognition, numerical reasoning, verbal intelligence, spatial reasoning, and working memory — each scored separately so you understand your profile.
“If your score is lower than you expected, that’s probably because other tests were inflating it. We’d rather be accurate than popular.”
The Process
Three steps from start to your honest, calibrated score.
Each question is chosen in real time based on your previous answer. The test spans all 5 cognitive domains and continuously refines its estimate of your ability level.
~15 minutesWhen you finish, the 2-Parameter Logistic IRT model calculates your theta estimate, converts it to an IQ score, and maps it against the norm group — all in your browser.
Instant · PrivateComposite IQ score, percentile, confidence interval, and individual domain scores — plus honest context on what your result actually means, backed by research.
5 domain scoresSample question
Questions adapt to your level in real time. This is a medium-difficulty pattern recognition example from the question bank.
Which shape correctly completes the sequence?
IQ Score Ranges
Every range from Extremely Low to Genius — with honest, research-backed explanations of what each classification means in practice.
Explore all ranges →What people say
Not everyone loves a calibrated score. But they trust it.
“I’d done four other free IQ tests before AurorIQ. They all gave me 130+. AurorIQ gave me 112. I looked into the methodology and AurorIQ is clearly the one doing it right. My ego didn’t love it, but my brain did.”
“The domain breakdown is what sets this apart. Seeing that my pattern recognition was 94th percentile but verbal was 61st actually told me something useful. Other tests just hand you a number.”
“Finally a test that acknowledges its own limitations. The results page says ‘this is not a clinical assessment’ instead of pretending otherwise. That alone makes it trustworthy.”
From the blog
Science-backed writing on IQ and cognition. No pop psychology.
Ready?
15 minutes. 25 questions. Honest results.
No account. No email. No flattery.