Cognitive Reflection Test: See How Often You Check Your First Answer in Ten Problems
This cognitive reflection test gives you ten original problems built on the logic of the published CRT, where a fast answer feels right and the correct one needs a second look. Get an instant reflection range, a breakdown by problem type, and an optional PDF.
Reviewed by Michael Callans, MSW 
Cognitive Reflection Test: Intuitive or Analytical? is a free, confidential 10-question self-assessment, original CRT-style problems, not the published CRT. It takes about 5 minutes, runs entirely in your browser, and gives you a plain-language score across 4 interpretation bands plus an optional in-depth report. It is an educational screening, not a diagnosis.
What this cognitive reflection test actually measures
The Cognitive Reflection Test, introduced by Shane Frederick in 2005, measures one specific habit: whether you accept the answer that springs to mind or pause to check it. The three original problems and the later seven-item version are now so widely shared that many people already know the answers, so this page uses ten problems of our own in the same style. The result is a range of reflective answers out of ten, for education and self-reflection, not a measure of intelligence. It is untimed and unproctored, and it is never an IQ score.
Ten fresh problems, not the famous three
Every problem here is original and written for this page. The bat and ball, the lily pads, the widgets, and the seven-item extension by Toplak and colleagues are not reproduced, to reduce direct recall of familiar answers. Prior exposure can still affect performance, and changing a problem's surface details does not establish that the new set is an equivalent validated measure. The new problems use the same traps: a sum-and-difference, doubling growth, rates, averages, and off-by-one counting.
The tempting answer is always on the list
Each problem offers four choices. One is correct, one is the intuitive answer most people reach first, and two are other plausible errors. Typical online versions ask for a typed number, which penalizes slips and typos as if they were lapses in reflection. A multiple-choice format with the lure visible tests the thing the CRT is about: noticing that the obvious answer deserves a second look.
A breakdown by trap type
Your result groups the problems into five pairs: sums and differences, growth and doubling, rates and averages, percentages and proportions, and counting and physical setups. These are our own descriptive groupings, not validated subscales. They show your correct-answer count in each pair, which can guide review. They do not establish why an answer was wrong or which reasoning habit caused it.
| Feature | Typical free quiz | Psychology.com |
|---|---|---|
| Which problems you get | The three 2005 items or the CRT-7, which many people have already seen | Ten original problems in the same logic families, written for this page |
| What the score means | Often presented as an intelligence or IQ result | A range of reflective answers out of ten; no IQ, percentile, or intelligence claim |
| Answer format | Typed numbers, so typos count as errors | Four choices per problem with the intuitive lure always listed |
| Breakdown | A single total | Five problem-type pairs, labeled as our own unvalidated groupings |
| Honesty about familiarity | Rarely mentioned | Explained on the page, with the research on prior exposure |
| Clinician-reviewed interpretation | Rarely | Yes, reviewed for framing and limits |
| Answers stay on your device | Often collected | Scored in your browser; only summary scores if you request a PDF or report |
How scoring works
Answers are scored 0–10 in total. Your result falls into one of 4 bands:
| Score | Band | What it suggests |
|---|---|---|
| 0–3 | The first answer mostly won: a reflection range | On most of these problems the quick answer went unchecked. That describes this attempt, not your intelligence. |
| 4–6 | A mix of quick and checked answers: a reflection range | You caught some of the traps and missed others. This is the middle scoring range on these ten problems. No representative norms have been established for this particular set. |
| 7–8 | Mostly checked answers: a reflection range | You answered seven or eight of the ten problems correctly, leaving two or three incorrect answers. Review those explanations without treating one attempt as a measure of your thinking style. |
| 9–10 | Checked on nearly every problem: a reflection range | You caught the trap on almost every problem. The quiz reaches its ceiling here and cannot distinguish further. |
What the Cognitive Reflection Test measures
Shane Frederick published the Cognitive Reflection Test in 2005 in the Journal of Economic Perspectives. It had only three problems, each with an answer that comes to mind immediately and is wrong, and a correct answer that is easy to confirm once you stop to check. The test does not measure how much maths you know; the arithmetic in every item is primary-school level. It measures whether you notice that the first answer deserves scrutiny, and whether you are willing to spend the small effort to scrutinize it. Frederick found that scores related to patience in financial choices and to some risk preferences, and that they correlated only moderately with other cognitive measures, which is why the test became popular as a quick index of a thinking disposition rather than ability.
Maggie Toplak, Richard West and Keith Stanovich extended the test to seven items in 2014 and argued that it captures what they call miserly information processing: the tendency to accept the cheap, fast answer rather than pay for a more careful one. The ten problems on this page are not the original three or the CRT-7 items. They are new problems that set the same kind of trap, scored as the number of reflective answers out of ten.
Dual-process theory in plain words
Psychologists often describe thinking as two broad modes. The first is fast, automatic, and effortless: it recognizes faces, finishes familiar sentences, and produces the number that pops up when you read a word problem. The second is slow, deliberate, and effortful: it checks, calculates, and holds several pieces of information in mind at once. Daniel Kahneman popularized these as System 1 and System 2 in Thinking, Fast and Slow (2011). Neither is better. The fast mode is what lets you drive, talk, and read without exhausting yourself; the slow mode is what catches the fast mode when it guesses.
The CRT works because its problems are built so the fast mode offers a confident wrong answer. A reflective answer means the slow mode was engaged long enough to notice the error, not that the person is incapable of intuition. Pennycook and colleagues (2016) concluded that CRT performance was a valid measure of reflection but not of intuition. That finding concerns the studied CRT measures and does not validate this page's new ten-item set. Someone who reaches the correct answer directly, without ever feeling the lure, scores the same as someone who felt it and corrected it. The score cannot tell those two people apart.
The familiarity problem
The original three problems have been printed in newspapers, bestselling books, and countless online quizzes. Stieger and Reips (2016) found that 44 percent of their participants had encountered the CRT or similar tasks before, and those people scored higher. Haigh (2016) found that 51 percent of a volunteer sample had seen at least one of the problems, and prior exposure raised scores. Other researchers, including Bialek and Pennycook (2018), found that repeated exposure did not destroy the test's ability to predict other reasoning measures, so the debate is unresolved. What nobody disputes is that a person who remembers that the ball costs five cents is recalling, not reflecting.
This is why every problem here is new. If you have seen the bat and ball before, you can still take this test fairly, because the specific answers cannot be remembered. What may carry over is the general lesson: slow down, the first number is a trap. Carrying that lesson is itself a kind of reflection, so the result still means something. It is also why a retake of this page is not an independent estimate; once you know a problem's twist, you remember it.
What the score does and does not predict
In research samples, higher CRT scores go along with lower susceptibility to several reasoning errors, more patience in choices between a smaller reward now and a larger one later, and lower belief in some kinds of unsupported claims. These are average associations across groups, usually modest in size. They do not mean that a particular person with a lower score is impulsive, gullible, or less intelligent, and a higher score does not make someone wise or immune to error. Reflection is a habit that depends on the situation: fatigue, time pressure, mood, and whether you care about the question all change how often the slow mode gets switched on.
The score on this page has not been validated against anything. We have no norms, no reliability estimate, and no evidence that it predicts the outcomes found for the published CRT. Treat your result as a snapshot of this attempt. If you are interested in the broader picture of how you think, the Thinking Style Test looks at preferences, the Critical Thinking Test at evaluating arguments, the Logical Reasoning Test at deduction, the Intuition Test at how much you trust gut feelings, the IQ Test at a wider sample of reasoning, and the Aptitude Test at verbal, numerical, logical and abstract questions against a clock. None of them converts into this score, and this score converts into none of them.
Methodology & sources
This is an original educational quiz informed by the Cognitive Reflection Test (Frederick, 2005) and its seven-item extension (Toplak, West and Stanovich, 2014). It does not reproduce any item from either instrument. We wrote ten new problems that share the CRT's defining structure: a fast, tempting answer that is wrong, and a correct answer that is simple to verify on reflection. Each problem has four choices, one correct and scored 1, the intuitive lure scored 0, and two further distractors scored 0. The total is the number of reflective answers out of ten. There are no reversed items, no time limit, no speed bonus, and no correction for guessing. Fixed editorial cutoffs of 0 to 3, 4 to 6, 7 to 8, and 9 to 10 assign broad reflection ranges; they are not clinical thresholds, percentiles, or intelligence bands, and we have no validation data, norms, or reliability estimate for this set. The five breakdown pairs are our own descriptive groupings, not validated subscales.
Every answer was verified arithmetically. Problem 1: a pastry costs p and the coffee p plus 5, so 2p plus 5 equals 6.50 and p equals 0.75, not 1.50. Problem 2: a number that doubles daily and covers the whole office on day 12 covered half on day 11, not day 6. Problem 3: the inlet adds one tenth of the tank each minute and the drain removes one fifteenth, so the tank gains one thirtieth of its volume a minute and fills in 30 minutes, not the 5 minutes you get by subtracting 10 from 15. Problem 4: average speed is total distance over total time; for equal distances at 10 and 30 km/h it is 2 times 10 times 30 divided by 40, which is 15 km/h, not 20. Problem 5: a 50 percent rise followed by a 50 percent cut gives 1.5 times 0.5, which is 0.75, so 25 percent lower, not the same. Problem 6: a 100 meter path with a lamppost every 10 meters and one at each end has 100 divided by 10 plus 1, which is 11 posts, not 10. Problem 7: 20 kg of produce at 99 percent water holds 0.2 kg of solids; at 98 percent water the solids are 2 percent of the total, so the total is 0.2 divided by 0.02, which is 10 kg, not 19.8. Problem 8: the boat floats and rises with the tide, so the same 10 rungs stay above the water, not 6. Problem 9: s plus 3s equals 12, so the shorter rope is 3 meters, not 4. Problem 10: two colonies at noon equal one colony's size at 1 pm, so the dish fills one hour earlier, at 9 pm, not 5 pm.
The references explain the science and its limits; none supplies these problems or validates this quiz. Research on familiarity (Stieger and Reips, 2016; Haigh, 2016) shows that prior exposure to CRT problems raises scores, which is the reason this page uses new items and the reason a retake of this page is not an independent estimate. Reflection also varies with fatigue, distraction, time pressure, and interest, so a single attempt describes a moment rather than a fixed trait. Do not use this result for employment, education placement, or any decision about another person.
References
- Frederick S. Cognitive reflection and decision making. Journal of Economic Perspectives. 2005;19(4):25–42. doi.org/10.1257/089533005775196732
- Toplak ME, West RF, Stanovich KE. Assessing miserly information processing: An expansion of the Cognitive Reflection Test. Thinking & Reasoning. 2014;20(2):147–168. doi.org/10.1080/13546783.2013.844729
- Stieger S, Reips UD. A limitation of the Cognitive Reflection Test: familiarity. PeerJ. 2016;4:e2395. doi.org/10.7717/peerj.2395
- Haigh M. Has the standard Cognitive Reflection Test become a victim of its own success? Advances in Cognitive Psychology. 2016;12(3):145–149. doi.org/10.5709/acp-0193-5
- Kahneman D. Thinking, Fast and Slow. Farrar, Straus and Giroux; 2011. ISBN 9780374275631.
- Pennycook G, Cheyne JA, Koehler DJ, Fugelsang JA. Is the cognitive reflection test a measure of both reflection and intuition? Behavior Research Methods. 2016;48(1):341-348. doi.org/10.3758/s13428-015-0576-1
- Bialek M, Pennycook G. The cognitive reflection test is robust to multiple exposures. Behavior Research Methods. 2018;50(5):1953-1959. doi.org/10.3758/s13428-017-0963-x
Cite this source
Psychology.com. (2026, October 5). Cognitive Reflection Test: Intuitive or Analytical? Psychology.com. https://psychology.com/tests/cognitive-reflection
Cognitive Reflection Test: Intuitive or Analytical? FAQ
Is this the real Cognitive Reflection Test?
No. The Cognitive Reflection Test is Shane Frederick's three-problem measure from 2005, and the CRT-7 is the seven-item extension by Toplak, West and Stanovich. Both are named here nominatively and neither is reproduced. This page uses ten original problems that set the same kind of trap, so people who already know the bat and ball answer can still take a fair attempt. The result is a reflection range on our problems, with no validation data behind it.
What does a low score mean?
It means that on this attempt, the fast answer won more often than the checked one. That is a description of ten problems on one day, not a verdict on your intelligence or your judgment. Tiredness, reading quickly, or not expecting a trick all push scores down. Most people who see what the trap was can avoid it next time, which is the whole point of the exercise. If you want to practice, our Logical Reasoning Test and Critical Thinking Test exercise related habits.
Does a high score mean I am analytical rather than intuitive?
It means you checked the tempting answer on most of these problems. It does not mean you lack intuition; a strong intuition that reaches the right answer directly scores the same as a corrected one. Everyone uses both fast and slow thinking all day. If the question of how much you rely on gut feeling interests you, the Intuition Test and the Thinking Style Test look at that preference directly rather than at accuracy.
I have seen the bat and ball problem before. Does that spoil the result?
Not on this page, because none of the published problems appear here. What may carry over is the general lesson to slow down, and research shows that prior exposure to CRT-style problems raises scores. That carried lesson is itself reflection, so the result still describes something real. A retake of this exact page is different: once you know each problem's twist you are remembering it, so a second attempt is not an independent estimate.
Is cognitive reflection the same as IQ?
No. The CRT correlates moderately with cognitive ability measures in research, but it was designed to capture a disposition to check answers, not raw capacity, and the arithmetic in every problem is deliberately simple. This page makes no IQ claim and has no conversion to one. Our IQ Test page is a separate untimed reasoning quiz with its own limits, and it also does not produce a certified IQ score.
Is the test really confidential?
Yes. It runs entirely in your browser. Your individual answers are scored in your browser and are never sent to us or linked to you. They stay in this tab so you can return to your result, and closing or clearing the tab clears them. If you ask us to email the PDF, we receive your summary scores with your email address, never your individual answers. No account is needed.
Related tests
Abstract Reasoning Test
24 original pattern and inductive reasoning questions in text, with a section breakdown.
Emotion Recognition Test
Name the emotion behind 20 described faces, voices and postures. A text analog, not the RMET.
Giftedness Test (Adult)
Reflect on intensity, curiosity, and overexcitability traits often linked to giftedness.
Hyperphantasia Test
Explore the vividness of your imagined sights, sounds, and other sensations.
IQ Test: Free Reasoning Quiz
30 original reasoning questions with an estimated performance range, not a certified IQ score.