Three cognitive assessments grounded in a century of research. Free, honest, and built on the same instruments psychologists actually use.
The phrase "IQ test" carries more cultural weight than it probably deserves. Popular imagination treats it as a single number summarizing a person's cognitive worth — a tidy verdict on the mind. The reality is messier and more interesting.
The statistical foundation was laid by Charles Spearman in 1904. He noticed that people who did well on one mental task tended to do well on others, even when the tasks seemed unrelated. He proposed a general factor, g, that contributes across cognitive domains. A century of research has refined this picture without overturning it.
Well-constructed assessments predict a narrow but real set of outcomes: academic achievement, performance in cognitively demanding work, and the speed at which novel material is learned.
Creativity, wisdom, emotional intelligence, conscientiousness, ethical judgment, motivation — none of these are captured by a pattern-recognition test. A score is one slice of one dimension of one kind of capability.
Intelligence is what the tests test. This is not a tautology if we recognize that the tests measure something useful — but only something. — Edwin Boring, Harvard, 1923
40 items across eight cognitive domains — matrix reasoning, sequences, number series, analogies, mental rotation, set logic, and Latin squares. Difficulty rises sharply; the final items discriminate beyond the 99th percentile. Returns an estimated IQ score with normal distribution placement.
BeginSyllogisms, conditional inferences, and propositional reasoning drawn from classical Aristotelian logic and modern propositional calculus. A focused test of analytical capacity — detecting valid arguments and spotting where conclusions don't follow.
BeginThree subtests after the classical paradigms — forward digit span, backward digit span, and spatial sequence (Corsi block-tapping). Measures the cognitive system that holds and manipulates information across seconds. A strong predictor of complex reasoning.
BeginDeveloped by John C. Raven in 1936. Among the most extensively researched cognitive instruments in history, with strong loadings on g and minimal cultural bias. Our main test borrows its item structure for pattern recognition tasks.
The dominant contemporary framework for cognitive abilities, integrating Cattell's fluid/crystallized distinction with Carroll's hierarchical model. Informs how we categorize results.
Alan Baddeley's multi-component model distinguishes the phonological loop, visuospatial sketchpad, and central executive. Our three memory subtests target each component separately.
Items in the logical reasoning test draw on syllogistic forms formalized by Aristotle and on propositional logic developed by Frege and Russell. Performance reflects formal reasoning, not domain knowledge.
After completing an assessment you'll receive a score and a percentile estimate. Two things are worth understanding first.
A clinical percentile means your performance equals or exceeds that of X% of a carefully normed sample. Our percentiles come from the population that has taken our tests — self-selected and not representative. Treat them as orientation, not placement.
Performance varies with sleep, stress, mood, motivation, time of day, and familiarity with the format. A single score is a data point, not a verdict.
Repeating the same test improves your score — not because you're getting smarter, but because you're getting better at the test. Expect inflated results on retakes.
Cortextest assessments are educational. They are not designed to identify learning disabilities, giftedness, or any medical condition. If you have concerns about cognitive function, consult a licensed psychologist.
Yes. All three assessments are free, with no registration, no email collection, and no paywall for results. The site is supported by advertising.
The main test contains 40 items and runs for 45 minutes. The logical reasoning test is shorter (15 items, ~15 minutes) and the working memory test is shorter still (~10 minutes). All three can be completed in roughly an hour.
Our scores indicate roughly how your performance compares to others who have taken the same test. They are not equivalent to clinical instruments like the WAIS-IV, which require controlled administration and representative norms. The main test is calibrated to discriminate up to approximately IQ 150.
No personal data is collected during the test. Anonymized performance data may be aggregated to improve item calibration. See our Privacy Policy for details.
Only the main 40-item test returns an estimated IQ score. The logical reasoning and working memory tests are focused on specific cognitive domains and report domain-specific results rather than a general IQ. Producing a defensible IQ score requires broad coverage of cognitive abilities, which only the main test provides.
The pattern recognition and memory tests use non-verbal materials specifically to minimize cultural and linguistic dependence. The logical reasoning test requires reading proficiency in English. No test is entirely culture-free.
Long-form articles on intelligence, cognition, and how psychometric measurement actually works. Clear explanations grounded in a century of research.
Spearman's 1904 discovery — the statistical pattern behind general intelligence — explained clearly, with diagrams.
The small, fragile mental workspace behind reasoning, mental arithmetic, and language. Baddeley's model explained.
From Binet's humane 1905 test, through Wechsler's modern scales, to the Flynn effect — and the controversies along the way.