Creativity Tests: How to Measure Your Creative Thinking

A creativity test attempts to measure what standard intelligence tests ignore: the cognitive abilities associated with generating novel, valuable ideas. That's a harder measurement problem than it first appears, and the field has produced several distinct instruments over the past 70 years — each with different theoretical underpinnings, different scoring methods, and different predictive validity.

This guide covers the main creativity tests in use, what they actually measure, how scores are interpreted, and the ongoing debate about whether any test can meaningfully capture creative potential.

What Creativity Tests Actually Measure

Most creativity tests operationalize divergent thinking — the ability to generate multiple, varied, and unusual responses to an open-ended prompt. The psychologist J.P. Guilford formalized this framework in his 1967 "Structure of Intellect" model, distinguishing divergent production (generating many possible answers) from convergent production (identifying the single correct answer).

Divergent thinking tests assess four core dimensions:

• Fluency: the number of relevant responses generated within a time limit
• Flexibility: the range of conceptual categories represented across responses
• Originality: the statistical rarity of responses compared to a norm group
• Elaboration: the level of detail and specificity added to responses

These dimensions capture something real — they correlate moderately with real-world creative achievement — but they're not the whole story. Divergent thinking is one component of creativity, not a synonym for it.

The Major Creativity Tests

The Torrance Tests of Creative Thinking (TTCT)

The Torrance Tests of Creative Thinking are the most widely used and most extensively validated creativity assessment in existence. E. Paul Torrance developed the test at the University of Minnesota in 1958, and the longitudinal research backing it spans five decades.

The TTCT comes in two forms:

Figural TTCT: Visual tasks where the test-taker draws and completes images. Tasks include Picture Construction, Picture Completion, and Repeated Lines. Nonverbal by design, which reduces language barriers.

Verbal TTCT: Written tasks including Asking (generating questions about a stimulus), Guessing Causes and Consequences, Product Improvement, and Unusual Uses.

The TTCT is scored on fluency, flexibility, originality, and elaboration. A newer Streamlined Scoring system for the Figural form also identifies 13 creative strengths including emotional expressiveness, humor, and resistance to premature closure.

Predictive validity: Torrance's 50-year longitudinal study found that childhood TTCT scores predicted adult creative achievement better than IQ scores — particularly for people who made inventions, started companies, or received recognition for creative work. Originality and elaboration scores were stronger predictors than fluency.

The Alternative Uses Task (AUT)

The Alternative Uses Task, developed by Guilford in 1967, is the most widely used research instrument for divergent thinking. The task is simple: generate as many uses as you can for a common object — a brick, a paperclip, a newspaper — within a time limit, typically two minutes per object.

Despite its simplicity, the AUT has been used in hundreds of studies examining divergent thinking, creativity, and incubation. A 2012 study by Baird et al. in Psychological Science used the AUT to demonstrate that mind-wandering during low-demand tasks improves creative performance — subjects who completed an undemanding task between AUT attempts scored higher on originality than those who rested or worked on a demanding task.

The AUT's limitation is scope. It measures a specific type of divergent thinking (object repurposing) and may not generalize cleanly to creative domains requiring different cognitive operations.

The Remote Associates Test (RAT)

The Remote Associates Test measures convergent creative thinking rather than divergent thinking — making it a useful complement to the AUT and TTCT. Developed by Sarnoff Mednick in 1962, the RAT presents three words (e.g., "pine," "crab," "sauce") and asks for a single word that connects all three ("apple").

The RAT operationalizes Mednick's associative theory of creativity: creative thinking involves forming distant associations between concepts, and the ability to find remote connections predicts the capacity to generate creative ideas in other domains.

RAT performance correlates with problem insight and sudden "aha" moments. Research by Mark Jung-Beeman and colleagues using fMRI found that RAT solutions preceded by a burst of gamma-band activity in the right anterior temporal lobe — a neural signature associated with sudden insight. This makes the RAT one of the few creativity instruments with a documented neural correlate.

The Consensual Assessment Technique (CAT)

The CAT, developed by Teresa Amabile in 1982, takes a fundamentally different approach. Instead of measuring divergent thinking on standardized tasks, the CAT has independent domain experts rate actual creative products — poems, drawings, collages, stories — on overall creativity.

Amabile's insight was that creative quality is ultimately a social judgment: a poem is creative to the extent that experts in poetry find it creative. The CAT produces reliable inter-rater agreement across a wide range of domains, validating expert consensus as a meaningful metric.

The limitation is practical: the CAT requires domain experts, real creative products, and substantial time. It's not a test you can administer in 20 minutes.

What Creativity Tests Don't Measure

Every major researcher in this area acknowledges that no test captures the full picture. What's consistently left out:

Domain knowledge: Actual creative work requires deep expertise. A physicist who scores low on the AUT can still make creative discoveries; an art student who scores high may produce derivative work. Tests measure cognitive process, not accumulated knowledge.

Motivation: Amabile's intrinsic motivation research found that external evaluation pressure reduces creative performance on production tasks. How someone performs on a timed, scored test may not predict performance when they're genuinely engaged.

Context and environment: Organizational structure, collaborative relationships, and access to resources shape creative output in ways no individual assessment can capture.

Domain specificity: Evidence for domain-general creativity is weak. Someone who generates highly original marketing ideas may produce conventional solutions in engineering, and vice versa.

How to Interpret Creativity Test Scores

A few interpretive cautions apply when reading creativity test results:

Fluency scores are least predictive. Raw idea count is the easiest dimension to inflate (just generate more) and the one least associated with real-world creative achievement in longitudinal data. Don't over-weight it.

Originality scores are most informative. Statistical rarity of responses is the dimension most strongly correlated with adult creative achievement in Torrance's longitudinal follow-up.

Scores reflect current skill, not permanent ceiling. Divergent thinking is trainable. Studies using structured practice consistently produce gains across all four scoring dimensions, including originality. A current score is a baseline, not a verdict.

Context effects are real. Mood, sleep quality, and prior activity affect divergent thinking performance. A test taken under suboptimal conditions will understate typical performance.

Creativity Tests vs. Intelligence Tests

Above an IQ threshold of roughly 120, creativity test scores and intelligence test scores are nearly uncorrelated. This finding — the "threshold hypothesis," established by Getzels and Jackson (1962) and replicated across multiple samples — means that higher IQ does not reliably predict higher creative performance within the normal-to-high ability range.

Convergent thinking, which intelligence tests measure well, is necessary but not sufficient for creative output. Beyond a threshold of general cognitive ability, what differentiates creative performance is divergent thinking ability, domain knowledge, and motivation — none of which standard intelligence tests capture.

Training vs. Testing

The most useful thing creativity research has established is that divergent thinking ability responds to practice. Studies comparing trained and untrained groups consistently find that structured practice improves AUT and TTCT scores. This makes creativity tests more useful as baselines for tracking improvement than as assessments of fixed ability.

If you're using a creativity test, the most productive approach is longitudinal: take the test, practice deliberately, test again. The change over time is more informative than the initial score.

Ready to train your creativity? Try science-backed exercises that measure and improve your creative thinking. Start a Free Exercise