Practice and reflection. An educational model-based estimate, not a clinical or standardized intelligence score.
You have a test in progress
Before we begin
Your reasoning profile.
Explore five reasoning domains through an adaptive 25–35 item session. Your responses influence which reviewed items are selected, while a content blueprint protects domain coverage. Read carefully and use your own reasoning.
25–35Reasoning questions
5Reasoning domains
~15Minutes
Find a quiet moment. Once a question is answered it moves on — you can't go back, by design.
AurorIQ is an educational self-reflection tool, not a clinical assessment.
A few details first
Tell us who's taking the test.
These optional details personalize your report. Age, gender and education do not change your score; this assessment has no representative age norms.
Question 1 · adaptive length 25–35
Computing your cognitive profile
Estimating ability
Mapping reasoning domains
Summarizing your answers
Resolving your archetype
Use the profile as a starting point
Each session samples five domains with at least five items per domain and usually 25–35 items total. The bank is larger than a single session, so retakes can use different subsets and prefer unseen items on this browser. Difficulty values are developer-seeded estimates, not empirical calibration. The AurorIQ reasoning index is not an IQ score, percentile or population norm.
This is an educational task or self-assessment, not a diagnosis. Read the instructions above for the format and approximate time. Scores describe this task or your reported preferences; research on the underlying concept does not validate this implementation.
The engine calculates results in your browser. Completed report snapshots are saved locally in My AurorIQ; planning tools require an explicit save. You can export or delete saved data at /passport/. Separate hosting, analytics, advertising and optional sharing have separate data implications. Read the privacy notice, methodology, or report an error.
Under the hood
How the adaptive engine works
AurorIQ uses a bounded Bayesian/EAP-like model to update an internal performance estimate and choose among eligible reviewed items. Selection occurs inside a content blueprint that requires coverage of pattern, numeric, verbal, spatial and working-memory tasks. The item parameters are developer-seeded routing values, not empirically fitted psychometric calibration.
In practice, the engine updates a bounded internal estimate after each scored response. It then prefers reviewed items near the current modeled level while enforcing domain minimums, item exposure penalties and retake unseen-item preference. Sessions can end after the minimum coverage is met and the estimate becomes sufficiently stable, or continue to a hard maximum. This is an engineering adaptation strategy, not a validated computerized adaptive test.
Some items may feel easier or harder. The engine can change their order, but this small bank cannot keep every question matched to your level. Difficulty is not evidence that the test is accurate.
The engine tracks posterior spread as an internal uncertainty signal. Its real-world measurement precision has not been established, so AurorIQ does not present population confidence intervals. Interpret the reasoning index together with observed accuracy, domain evidence and limitations.
What gets measured
Five reasoning domains, one composite
Each session is content-balanced across five reasoning domains. Domain interpretations are withheld unless at least five items were served in that domain:
Pattern recognition — abstract sequences and rule discovery: given a series, what governs it, and what comes next? This is the closest cousin to the matrix-style items used in classic fluid-intelligence measures.
Numeric reasoning — quantitative relationships, proportions, and number series. Not arithmetic speed, but the ability to see the structure inside numbers.
Verbal reasoning — analogies, relationships between meanings, and logical inference expressed in language.
Spatial reasoning — mental rotation, folding, and visualizing forms in space without physically manipulating them.
Working memory — holding and transforming information over short spans, the mental scratchpad that underpins every other domain.
The balance between your five domain scores is what determines your cognitive archetype. Two people can land on the same composite score with completely different profiles — one carried by spatial and pattern reasoning, another by verbal and memory strength — and the archetype is designed to capture that difference in shape, not just height. For the deeper theory behind the domains, our guides on fluid versus crystallized intelligence and working memory cover the research in plain language.
The reveal
What your result includes
Your result includes the number answered correctly, an estimated AurorIQ reasoning score, relative domain patterns and a descriptive archetype. The detailed report connects this session to practical experiments. An educational model-based estimate, not a clinical or standardized intelligence score. AurorIQ does not have representative population norms. No population percentile, age ranking or rarity is reported.
Conditions matter
Getting an accurate result
Cognitive measurement is sensitive to state. The same person can score noticeably differently depending on sleep, stress, and setting — an effect large enough that we wrote a full guide on it. To give the engine a clean signal:
Take it in one sitting. The test takes about 15 minutes, and mid-test progress isn't saved. Wait for a window where you won't be interrupted.
Don't look things up or use paper. Try the items unaided and note any assistance when interpreting your session.
Answer honestly, not strategically. There's no penalty scheme to game and no pattern to exploit. The engine assumes each answer reflects a genuine attempt.
Reasonably rested beats late-night curiosity. Sleep deprivation measurably suppresses working memory and fluid reasoning — the exact things being measured.
Honest by design
What this test is — and isn't
AurorIQ is an educational self-reflection tool, not a validated clinical assessment. Online delivery alone does not determine validity: evidence must support the particular instrument, population and use. Do not use AurorIQ for diagnosis, placement, hiring, legal decisions or credentials.
This exercise can describe your performance on its questions and suggest a practice focus. Five items per domain cannot establish fixed strengths or weaknesses. Read the methodology for the scoring model and its limitations.
Questions
Frequently asked questions
Do I need to prepare for the IQ test?
No preparation is needed, and practicing beforehand would only distort the estimate. What helps is the setting: a quiet room, a clear 15-minute window, and reasonable alertness. Fatigue, stress, and interruptions are the biggest sources of underperformance on any cognitive test.
Why can't I go back to a previous question?
Answers are locked to preserve the response sequence and a consistent test flow. This is an interface rule, not evidence that revising an answer would make scoring mathematically invalid.
Why do the questions feel difficult?
The next item responds to your earlier answers within this fixed bank. There is no established target percentage correct for this implementation; difficulty alone cannot establish test quality.
What happens if I close the tab mid-test?
Mid-test progress isn't saved — you'd start fresh on your next visit. Once you complete the test, your full result is stored on your device so you can revisit it any time from the results page.
Is my data stored or shared anywhere?
The assessment engine processes answers in your browser and saves supported results in local storage. Optional sharing and website services have separate data implications. There's no account, no email, and your responses are never uploaded to a server.
Can I retake the test, and will my score change?
You can retake it whenever you like. Expect some movement between sittings — results can move between sittings because item subsets, familiarity and testing conditions differ. Retakes prefer unseen items stored on this browser, but overlap can still occur; a score change does not establish a change in intelligence.
How certain is the model estimate?
The engine tracks posterior spread as an internal uncertainty signal. Its real-world measurement precision has not been established, so AurorIQ does not present population confidence intervals. Interpret the reasoning index together with observed accuracy, domain evidence and limitations.
