Reasoning
Insight Problems Test
Also called: matchstick arithmetic, matchstick problems, insight problem solving task
Measure insight problem solving: participants move one matchstick to make a false Roman numeral equation true, within a time limit, and say whether the answer came suddenly.
8-12 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Preview any test free after you sign up.
- Papers, 2021 to 2025
- 173
- Evidence grade
- C
- Registered clinical studies
- 1
What it measures
Measures insight problem solving with matchstick arithmetic. Each of 8 problems is a false equation in Roman numerals made of matchsticks, which moving one stick makes true; some need a move that few people think of at first, such as changing a sign. The main score is the number solved within the time limit. The problems use no words.
Constructs
- Insight problem solving
- Constraint relaxation
- Chunk decomposition
Research uses
- Creativity and insight research
- Individual differences in problem solving
- Within-study group comparison
How the paradigm works
Insight problems are solved by a sudden change in how the problem is seen, after a stage of being stuck. Matchstick arithmetic makes that change measurable (Knoblich, Ohlsson, Haider and Rhenius 1999). Each problem is a false equation in Roman numerals made of matchsticks, and moving one stick makes it true. Most people first move sticks between numerals, which is easy. Changing a sign, turning a plus into a second equals sign so that three equal numbers stand in a row, or sliding a stick to turn a V into an X needs a constraint or a tight perceptual chunk to be let go, and Knoblich and colleagues found those problems solved less often and more slowly, and much faster on a second meeting.
More about the paradigm
The 8 problems were written for this test and checked by code, so each has exactly one solution and that solution needs the move of its type: 2 that move a numeral's stick, 2 that change a sign, 2 that make three equal sides and 2 that turn a V into an X or back. Two blocks each hold one problem of every type, as in the source's two blocks, and the session's random seed orders each block; the order is recorded. Two practice problems of the easiest type come first, so practice does not teach the harder moves. The problems are drawn as matchsticks and use no words, so they are the same in every language.
Participants type the new equation on an on-screen keypad of I, V, X, +, - and = (or the same keys), with 90 seconds per problem by default and a countdown bar. A wrong answer is marked and they keep trying until the time runs out, as in the source, which then shows the solution. After each solved problem a question asks whether the answer came suddenly (Aha!) or step by step, the self-report of Jung-Beeman and colleagues (2004); Danek, Wiley and Öllinger (2016) found that many matchstick solutions come without an Aha. Scores support research and do not diagnose.
Scores
- Problems Solvedcount
- Problems solved with a correct equation within the time limit.
- Higher means more problems solved
- Percent Solved%
- Problems solved as a percent of those presented.
- Higher means more problems solved
- Numeral Problems Solvedcount
- Solved problems among the 2 that move a stick between numerals (Knoblich's type A).
- Usually the easiest type
- Sign Problems Solvedcount
- Solved problems among the 2 that change a sign (type B).
- Higher means more sign changes found
- Three-Equal Problems Solvedcount
- Solved problems among the 2 that make three equal sides (type C).
- Usually the hardest type on a first meeting
- V and X Problems Solvedcount
- Solved problems among the 2 that turn a V into an X or back (type D).
- Higher means more tight chunks broken up
- First Block Solvedcount
- Solved problems among the 4 of the first block, one of each type.
- A first meeting with each type
- Second Block Solvedcount
- Solved problems among the 4 of the second block.
- Compare with the first block for transfer
- Median Solution Timems
- Median time from a problem's appearance to its correct equation, on solved problems, leaving out answers under 100 ms.
- Lower means faster solutions; it covers solved problems only
- Wrong Answerscount
- Submitted equations that were not a solution.
- Describes how readily answers were tried
- Time-outscount
- Problems not solved when the time ran out.
- A normal result on hard problems
- Aha Share%
- Of the solved problems rated, the share whose answer came suddenly.
- Describes how solutions felt; empty when nothing was rated
- Sudden Solutionscount
- Solved problems whose answer the participant said came suddenly.
- Empty when nothing was rated
- Answers Under 100 mscount
- Equations submitted within 100 ms of a problem's appearance.
- Expected to be 0
- Problems Attempteditems
- Problems with a submitted equation, or with keys pressed when the time ran out.
- Fewer than 20% of those presented is invalid
- Problems Presenteditems
- Problems shown in the scored run.
- Fewer than 75% of those set is invalid
- Problems Setitems
- Problems the session set out to show.
- 8 in a standard session
Use and evidence
173 papers from 2021 to 2025 name this task in their title or abstract, down 28% compared with 2016 to 2020. That ranks it 136 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 32 |
| 2024 | 28 |
| 2023 | 43 |
| 2022 | 34 |
| 2021 | 36 |
| 2020 | 44 |
| 2019 | 62 |
| 2018 | 63 |
| 2017 | 35 |
| 2016 | 37 |
| 2015 | 34 |
| 2014 | 33 |
| 2013 | 43 |
| 2012 | 39 |
| 2011 | 35 |
| 2010 | 24 |
| 2009 | 33 |
| 2008 | 20 |
| 2007 | 24 |
| 2006 | 21 |
| 2005 | 18 |
| 2004 | 19 |
| 2003 | 11 |
| 2002 | 4 |
| 2001 | 7 |
| 2000 | 8 |
Use grade C
100 to 299 papers from 2021 to 2025.
Evidence grade C
At least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies.
- Papers on test-retest reliability
- 1
- Papers on reliability or validity
- 46
- Registered clinical studies using it as an outcome
- 1
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Problems
- 8 matchstick equations written for this test, 2 of each type: moving a stick between numerals, changing a sign, making three equal sides, and turning a V into an X or back. Each has exactly one solution, checked by code.
- Order
- 2 blocks, each with one problem of every type; the session's seed orders each block, and the order is saved.
- Rules
- Move exactly one stick, turned or not; no stick is added or taken away, and a single slanted stick is not an I. Roman numerals I, V and X with +, - and =, drawn as matchsticks with no words.
- Answer
- The new equation, typed on an on-screen keypad of I, V, X, +, - and = (or the same keys). A wrong answer is marked and the participant keeps trying; when the time runs out, the solution is shown.
- Time limit
- 90 seconds per problem by default, 60 to 180 as a setting, with a countdown bar. Knoblich and colleagues (1999) allowed 5 minutes.
- Aha question
- After each solved problem: did the answer come suddenly or step by step? On by default; a setting turns it off.
- Practice
- 2 problems outside the set, of the easiest type, each followed by its solution.
The equations were written for this test; none uses the numerals of a published matchstick problem or solution, so published solution rates do not apply.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: insight_problems_problem from the scored run, one problem each, solved by a typed equation within its time limit. Typed equations have no chance level, so accuracy is not judged against chance, and a problem not solved in time is a normal result on a hard problem. Fewer than 75% of the problems set presented invalidates, and so does fewer than 20% of those presented attempted. Answers under 200 ms on at least half of 4 or more answered problems invalidate. A hidden tab withholds the percentile.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. A low score is the result the test measures and never invalidates, but with almost nothing typed, a low score cannot be told from no effort, so too few attempts invalidate. A score over a much shorter session cannot be compared with a full one. An answer under 200 ms is faster than a simple visual reaction, so it cannot reflect the item. Following response-time effort screening for untimed tests (Wise and Kong 2005), the session is invalid only when at least half the answers are that fast. A hidden tab lets the study, delay or response timers run without the participant, so a percentile is withheld after one.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Time per problem (seconds) | 90 | 60 to 180 |
| Aha question | On | On or off |
Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.
Need a setting this test does not have?
Devices and languages
- Untimed test: runs on desktops, laptops, tablets and phones. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Insight Problems Test in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Knoblich, Ohlsson, Haider & Rhenius, 1999), with the standard settings (time per problem (seconds): 90; aha question: On). The main score was problems solved (problems solved with a correct equation within the time limit).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Knoblich, G., Ohlsson, S., Haider, H., & Rhenius, D. (1999). Constraint relaxation and chunk decomposition in insight problem solving. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25(6), 1534-1555. https://doi.org/10.1037/0278-7393.25.6.1534
- Jung-Beeman, M., Bowden, E. M., Haberman, J., Frymiare, J. L., Arambel-Liu, S., Greenblatt, R., Reber, P. J., & Kounios, J. (2004). Neural activity when people solve verbal problems with insight. PLoS Biology, 2(4), e97. https://doi.org/10.1371/journal.pbio.0020097
- Danek, A. H., Wiley, J., & Öllinger, M. (2016). Solving classical insight problems without Aha! experience: 9 dot, 8 coin, and matchstick arithmetic problems. The Journal of Problem Solving, 9(1). https://doi.org/10.7771/1932-6246.1183
Source papers
- Knoblich G, Ohlsson S, Haider H, Rhenius D (1999). Constraint relaxation and chunk decomposition in insight problem solving. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25(6):1534-1555.Paradigm source, Adults
Questions
How long does the Insight Problems Test take?
About 8 to 12 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops, tablets and phones. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Problems Solved, Percent Solved, Numeral Problems Solved, Sign Problems Solved, Three-Equal Problems Solved, V and X Problems Solved, First Block Solved, Second Block Solved, Median Solution Time, Wrong Answers, Time-outs, Aha Share, Sudden Solutions, Answers Under 100 ms, Problems Attempted, Problems Presented and Problems Set. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Insight Problems Test used in research?
173 papers from 2021 to 2025 name it in their title or abstract, down 28% compared with 2016 to 2020. That ranks it 136 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Insight Problems Test have?
In the published research, 1 paper reports its test-retest reliability and 1 registered clinical study uses it as an outcome (evidence grade C in the ConductCognition test census).
Which settings can a researcher change?
Time per problem (seconds) (60 to 180) and aha question (On or off). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.