Processing Speed
Fraction Arithmetic Task
Also called: fraction computation task, fraction operations task, rational number arithmetic, whole number bias in fraction arithmetic
Measure fraction arithmetic: participants add, subtract, multiply and divide fractions with equal or unequal denominators and type each answer as a fraction, scored as accuracy by operation and the strategies behind wrong answers.
4-8 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 85
- Evidence grade
- D
- Registered clinical studies
- 1
What it measures
Measures fraction arithmetic. Participants add, subtract, multiply and divide two fractions, with equal or unequal denominators, and type each answer as a fraction; equivalent fractions count. The main score is accuracy, also scored by operation and kind of denominators, and wrong answers are sorted by the strategy behind them, such as adding numerators and denominators.
Constructs
- Fraction arithmetic
- Rational number understanding
- Whole number bias
- Procedural knowledge
Research uses
- Mathematics education research
- Numerical cognition research
- Research on strategy use and errors
- Comparing operations and kinds of denominators within a study
How the paradigm works
The fraction arithmetic task measures how accurately people add, subtract, multiply and divide fractions, and which strategies their wrong answers come from (Siegler and Pyke, 2013; Braithwaite, Pyke and Siegler, 2017). Each problem pairs two proper fractions with denominators from 2 to 9, either sharing a denominator or with denominators that have no common factor and a product of at most 40, and the participant types the answer in two boxes stacked as a fraction, with no time limit and no calculator. A round has one problem for each operation with equal and with unequal denominators; the standard 2 rounds give 16 problems, generated from the session's seed, and subtraction and division put the larger fraction first. Any answer equal in value to the right one counts, so unsimplified and improper fractions are right. Wrong answers are sorted into independent whole numbers (adding or subtracting numerators and denominators separately), a procedure from another operation (such as keeping the common denominator when multiplying), execution slips named by the strategy model, and other errors; every problem is chosen so these give different values.
Scores
- Accuracy%
- Answers equal in value to the right answer, over the problems shown.
- Higher is better
- Addition Accuracy%
- Accuracy on the addition problems.
- Higher is better
- Subtraction Accuracy%
- Accuracy on the subtraction problems.
- Higher is better
- Multiplication Accuracy%
- Accuracy on the multiplication problems.
- Higher is better
- Division Accuracy%
- Accuracy on the division problems.
- Higher is better
- Equal Denominators Accuracy%
- Accuracy on the problems whose fractions share a denominator.
- Higher is better
- Unequal Denominators Accuracy%
- Accuracy on the problems whose denominators differ.
- Higher is better
- Accuracy by Operation and Denominators%
- Accuracy in each of the 8 cells, such as add-equal or divide-unequal.
- Higher is better
- Whole Number Answers%
- Answers that add or subtract numerators and denominators separately, over the addition and subtraction problems where that gives a fraction.
- Higher means more whole number bias
- Wrong Operation Answers%
- Answers that use another operation's procedure, such as keeping the common denominator when multiplying, over the multiplication and division problems.
- Higher means more borrowed procedures
- Whole Number Errorscount
- Answers from adding or subtracting numerators and denominators separately.
- Informational
- Wrong Operation Errorscount
- Answers from another operation's procedure.
- Informational
- Execution Errorscount
- Answers that leave the numerators unconverted over a common denominator, or invert the first fraction when dividing.
- Informational
- Other Errorscount
- Wrong answers that match no named error.
- Informational
- Unanswered Problemscount
- Problems submitted with both boxes empty.
- Quality check
- Median Answer Timems
- Median time from a problem appearing to its answer, for answers of at least 100 ms.
- Informational; the task has no time limit
- Answers Under 100 mscount
- Answers submitted less than 100 ms after their problem appeared.
- 0 expected
- Problems Answeredcount
- Problems with an answer.
- Fewer than half the planned problems invalidates
- Problemscount
- Problems shown in the scored run.
- Compare with planned problems
- Planned Problemscount
- 8 a round: 16 at the standard 2 rounds, 8 to 32 by the setting.
- Informational
- Most Repeated Answercount
- The largest number of problems given one identical answer.
- Half the planned problems or more is flagged
Use and evidence
85 papers from 2021 to 2025 name this task in their title or abstract, up 42% compared with 2016 to 2020. That ranks it 316 of 711 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 26 |
| 2024 | 18 |
| 2023 | 22 |
| 2022 | 10 |
| 2021 | 9 |
| 2020 | 11 |
| 2019 | 14 |
| 2018 | 9 |
| 2017 | 15 |
| 2016 | 11 |
| 2015 | 12 |
| 2014 | 6 |
| 2013 | 7 |
| 2012 | 7 |
| 2011 | 4 |
| 2010 | 2 |
| 2009 | 2 |
| 2008 | 2 |
| 2007 | 2 |
| 2006 | 1 |
| 2005 | 1 |
| 2004 | 0 |
| 2003 | 1 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 1 |
Use grade D
30 to 99 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 20
- Registered clinical studies using it as an outcome
- 1
Measured on ConductCognition
Pooled across the studies that feed our platform norms, by device group. A figure needs 50 or more people.
No pooled figure yet.
Note on the counts: includes classroom studies.
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 0.58.0, counted 13 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Problems
- Two proper fractions with denominators from 2 to 9, sharing a denominator or with denominators that have no common factor and a product of at most 40. Subtraction and division put the larger fraction first.
- Rounds
- A round has one problem for each operation with equal and with unequal denominators, 8 problems. The standard 2 rounds give 16 problems; the setting runs 1 to 4 rounds.
- Answer
- Two boxes stacked as a fraction, the top number over the bottom number, or the top box alone for a whole number. Any answer equal in value to the right one counts, simplified or not.
- Order
- Problems generated from the session's seed, without repeats, in an order drawn from it. One practice problem with its answer comes first; then no time limit and no feedback.
- Errors
- Each wrong answer is matched to independent whole numbers, another operation's procedure or an execution slip, or counted as other. Every problem is chosen so these give different values.
- Scores
- Accuracy (the main score), accuracy by operation, by kind of denominators and by cell, and the rates of whole number and wrong operation answers.
The problems are generated by this library's code. The scores describe answers to these problems, support research and make no school-readiness or diagnostic claim.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: fraction_arithmetic_problem from the run with the most of them, latest on ties, one typed fraction per problem. Typed answers have no chance level, so accuracy is not judged against chance, and problems have no time limit, so none can be missed. Fewer than half the planned problems answered invalidates. An incomplete run is flagged, and so is one answer typed for half the problems or more. Answers under 200 ms on at least half of four or more answered problems invalidate. A hidden tab is flagged.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. A score over fewer than half the problems cannot be compared with a score over all of them, so it invalidates. Wrong answers, and the strategies behind them, are what the test measures and never invalidate. The problems have different answers, so one answer typed for half of them cannot follow them; it is flagged. 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. Problems stay on screen until they are answered, so a hidden tab does not change what was shown, but it can mean a calculator or help was used; it is flagged.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Rounds of 8 problems | 2 | 1 to 4 |
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 Fraction Arithmetic Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Siegler & Pyke, 2013; Braithwaite, Pyke & Siegler, 2017), with the standard settings (rounds of 8 problems: 2). The main score was accuracy (answers equal in value to the right answer, over the problems shown).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Siegler, R. S., & Pyke, A. A. (2013). Developmental and individual differences in understanding of fractions. Developmental Psychology, 49(10), 1994-2004. https://doi.org/10.1037/a0031200The four operations crossed with equal and unequal denominators, and the independent whole numbers and wrong fraction operation strategies.
- Braithwaite, D. W., Pyke, A. A., & Siegler, R. S. (2017). A computational model of fraction arithmetic. Psychological Review, 124(5), 603-625. https://doi.org/10.1037/rev0000072A strategy model of fraction arithmetic that sets execution errors apart from strategy errors and counts an answer right by its value.
Source papers
Questions
How long does the Fraction Arithmetic Task take?
About 4 to 8 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, Addition Accuracy, Subtraction Accuracy, Multiplication Accuracy, Division Accuracy, Equal Denominators Accuracy, Unequal Denominators Accuracy, Accuracy by Operation and Denominators, Whole Number Answers, Wrong Operation Answers, Whole Number Errors, Wrong Operation Errors, Execution Errors, Other Errors, Unanswered Problems, Median Answer Time, Answers Under 100 ms, Problems Answered, Problems, Planned Problems and Most Repeated Answer. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Fraction Arithmetic Task used in research?
85 papers from 2021 to 2025 name it in their title or abstract, up 42% compared with 2016 to 2020. That ranks it 316 of 711 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.58.0, counted 13 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Fraction Arithmetic Task have?
In the published research, 0 papers report its test-retest reliability and 1 registered clinical study uses it as an outcome (evidence grade D in the ConductCognition test census).
Which settings can a researcher change?
Rounds of 8 problems (1 to 4). 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.