Reasoning

Matrix Reasoning Test

Measure nonverbal reasoning with 3 by 3 visual matrices generated for each session and scored by the number of rules each puzzle needs.

Useful for reasoning, education, aging, intervention, and cross-language cognitive research batteries.

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.

Executive function7-10 minUntimed
?
Pattern completion
Complete the visual pattern

What this task measures

Measures nonverbal reasoning with 3 by 3 visual matrices generated for each session. Shapes change in number, size, rotation, fill and color by row and column rules.

Core constructs

  • Fluid reasoning
  • Rule induction
  • Nonverbal problem solving
  • Relational reasoning

Research fit

  • Fluid reasoning research
  • Education and development studies
  • Aging and intervention studies
  • Within-study group comparison

Paradigm overview

Matrix Reasoning measures nonverbal reasoning. Each puzzle is a 3 by 3 grid of pictures with the bottom-right picture missing. The pictures change by rules across the rows and down the columns, and the participant chooses the picture that completes the grid.

Items are generated for each session from a seed saved with the trial data. Each cell holds one to four identical shapes whose number, size, rotation, fill and color follow three kinds of rule: a value that changes across each row, a value that changes down each column, or three values that each appear once in every row and column. Items need one, two or three rules to answer and run from easiest to hardest. One- and two-rule items have 4 options and three-rule items have 6. Every wrong option breaks at least one rule, and every one-rule distractor breaks exactly that rule. Option sets are balanced, so the answer cannot be found from the options alone. There is no time limit, and two practice puzzles show the right answer.

Scores include accuracy, total correct, median response time on correct puzzles, a difficulty-weighted score in which each puzzle counts by its number of rules, accuracy at each rule count, and the share of answers at the most-chosen option position.

Key scoring outputs

Accuracy

proportion

Share of scored puzzles answered correctly.

Higher is better

Difficulty-Weighted Score

proportion

Correct puzzles weighted by the number of rules each needs (1 to 3), divided by the total weight.

Higher is better

Total Correct

count

Number of scored puzzles answered correctly.

Higher is better

Three-Rule Accuracy

proportion

Accuracy on puzzles that need three rules. One- and two-rule accuracy are also reported.

Higher is better

Median Correct RT

ms

Median response time on correctly answered puzzles.

Informational

Most-Chosen Position

proportion

Share of answers given at the most frequently chosen option position. A high value can mean a fixed answering pattern.

Informational

Task design as built

The trial sequence participants see, as the test runs in ConductCognition. Settings a researcher can change are listed below.

Grid
3 by 3 pictures with the bottom-right picture missing.
Rules
Number, size, rotation, fill and color change across rows, down columns, or once in every row and column.
Difficulty
Items need one, two or three rules and run from easiest to hardest.
Options
4 options for one- and two-rule items and 6 for three-rule items, balanced so the answer cannot be found from the options alone.
Items
Generated for each session from a seed saved with the data.
Timing
Untimed. Two practice puzzles show the right answer.
Input
Mouse, touch or keyboard.

The items are generated by ConductCognition and do not come from a published matrix test, so scores are compared within a study.

Running this test

Settings you can change

SettingStandardAllowed
Number of items2412 to 36

Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.

How test settings work

Need a setting this test does not have?

Devices and languages

  • Untimed test: runs on desktops, laptops, tablets and phones. Participant devices
  • Task text: English and Italian.
  • Available on: Every plan, including the free plan.
  • Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.

Cite this test

ConductCognition versions: task version 2.0.0, scoring version 2.0.0. Every export records the versions each session used, for a methods section.

  • Carpenter, P. A., Just, M. A., & Shell, P. (1990). What one intelligence test measures: A theoretical account of the processing in the Raven Progressive Matrices Test. Psychological Review, 97(3), 404-431. https://doi.org/10.1037/0033-295X.97.3.404Rule types in matrix problems.
  • Chierchia, G., Fuhrmann, D., Knoll, L. J., et al. (2019). The matrix reasoning item bank (MaRs-IB). Royal Society Open Science, 6(10), 190232. https://doi.org/10.1098/rsos.190232Open matrix item bank, a different item set.

Source papers

Chierchia G, Fuhrmann D, Knoll LJ, et al. (2019). The matrix reasoning item bank (MaRs-IB). Royal Society Open Science, 6, 190232.

N 10618-33

Carpenter, Just and Shell (1990): rule types in matrix problems. Arendasy and Sommer (2013): option sets that limit answering from the options alone.

Paradigm sourceAdults

Research FAQ

Common questions about this online cognitive test

How long does the Matrix Reasoning Test take?

About 7 to 10 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?

Accuracy, Difficulty-Weighted Score, Total Correct, Three-Rule Accuracy, Median Correct RT and Most-Chosen Position. Each session is also marked valid, needs review or invalid, with the reason.

Which settings can a researcher change?

Number of items (12 to 36). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.

Which languages can participants use?

The test's own text is in English and Italian. Italian participant pages come with yearly plans.

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.