Working Memory3–4 minBurden: MediumEMA: Medium

Mental Arithmetic

Timed arithmetic with typed answers: problems appear one at a time, and the participant types each result on an on-screen number pad. Measures the speed and accuracy of mental calculation (number facility).

Number facilityMental calculationProcessing speed
Category
Working Memory
Typical duration
3–4 min
Participant burden
Medium
EMA suitability
Medium

Try this task

Configure parameters and run an interactive preview exactly as participants will experience it. No data is recorded.

Configure preview

Adjust parameters below, then start the preview on the right.

Include practice trials

Shown with feedback before the main task

Task parameters

Problem levels

Default: levels 1 and 2, the four basic operations, as in the Number Facility tests of the ETS Kit of Factor-Referenced Cognitive Tests (Ekstrom, French, Harman & Dermen, 1976; addition, subtraction and multiplication, division) and the Tempo-Test-Rekenen (de Vos, 1992). Levels 3 and 4 have no published counterpart in this format (our choice). The families of the chosen levels take turns in shuffled rounds, so each gets the same share of the problems. Every answer is a whole number of 0 or more.

A 180 s block; the problem on screen when the time is up is kept but not scored. No time limit per problem: the next problem comes after an answer. Practice: 4 problems with feedback. About 4 min in all.

Default 180 s, no limit per problem and no feedback, as in the tablet task of Roy et al. (2025): problems one at a time, answers typed on an on-screen keypad with a delete key, 3 minutes. The Kit times each page of 30 problems at 2 minutes and the Tempo-Test-Rekenen each column of 40 at 1 minute; a smartphone version for daily life used 90 s (Price et al., 2017). For repeated use in daily life a shorter block works: in Roy et al. (2025) the first minute alone gave a split-half reliability of .92 (.97 for all 3 minutes). Practice length and feedback time have no published value (our choice).

Number ranges

Operands start at 2. Level 1: 1- and 2-digit numbers (up to 99), as in the Kit's addition and subtraction items; a − b is always positive. Level 2: 2-digit × 1-digit numbers, and division of a 2- or 3-digit number by a 1-digit number with a whole-number result, as in the Kit (the Kit details are from the description in Bermingham et al., 2013). Level 3: numbers up to 30 and multipliers and divisors as in level 2 (our choice); a number that is divided, or from which a product is subtracted, is built from these. Problems are built so that working strictly from left to right, or ignoring the brackets, gives a different answer. Level 4: nine word problems (three each for percentages, direct and inverse proportion) with numbers below 1000, each naming the unit it asks for.

Empty: every participant and session gets new problems, and the seed drawn is stored with the result, so the problems can be rebuilt. A number (0 to 4294967295) gives every participant the same problems.

Outputs: correct answers per minute (number correct ÷ time spent on the scored problems, the score of Roy et al., 2025, after Vandierendonck, 2017), correct answers, problems answered, accuracy per problem family and the median time to ✓ of correct answers. Per problem: level, family, the problem shown, the answer, the typed response, time to ✓ and to the first key, deletions, and for level 3 whether a wrong answer came from the order of operations or the brackets. Duration estimate: 5 s per level 1 problem, 7 s level 2, 10 s level 3, 20 s level 4 (our estimate). Responses are timed from the drawn problem to the touch-down, but this timing has not yet been validated on real devices.

3–4 minBurden: MediumEMA: Medium

This is a researcher preview. No participant data is recorded.

Simulated participant view

9:41

Mental Arithmetic

Timed arithmetic with typed answers: problems appear one at a time, and the participant types each result on an on-screen number pad. Measures the speed and accuracy of mental calculation (number facility).

No data is recorded

Participant experience on smartphone

One problem at a time above an on-screen number pad (digits, delete, ✓). The answer appears in a box as it is typed and is confirmed with ✓; the next problem follows after a short blank. In a fixed-time block a countdown shows the seconds left. Expressions use ×, ÷ and − (in German · and :); level 4 problems are short word problems in the participant's language.

When to use

Useful when a study needs arithmetic fluency with produced answers rather than true/false judgements: the speed and accuracy of mental calculation under time pressure, for example in studies of numeracy, fatigue or cognitive change, or as practice for selection tests that contain arithmetic. The levels keep a study to the four basic operations, as in established number-facility tests, or add two-step problems and word problems.

When not to use

Not a standardised test and not a source of norms: the problems are generated, and no published test form is reproduced. Less suitable where math anxiety or differences in schooling would confound the target construct. For a two-button task with true/false judgements, use Arithmetic Verification.

How to use in a study

Keep the default fixed-time block of 3 minutes with the four basic operations, no feedback and no limit per problem, as in the tablet task of Roy et al. (2025). For repeated short sessions the block can be shortened: in Roy et al. (2025) the first minute alone had a split-half reliability of .92. Leave the seed empty for a new form per participant and session, or set one to give everyone the same problems. Report the levels and number ranges with the results.

Researcher-configurable parameters

  • Levels: 1 addition and subtraction, 2 multiplication and division (default 1 and 2: the four basic operations of the Number Facility tests in the ETS Kit, Ekstrom et al., 1976, and of the Tempo-Test-Rekenen); 3 two steps (order of operations, brackets); 4 word problems (percentages, rule of three with direct and inverse proportion)
  • Block: a fixed time (default 180 s, Roy et al., 2025) or a fixed number of problems (default 20)
  • Time limit per problem: default none, as in Roy et al. (2025); a problem that runs out is an omission
  • Number ranges: level 1 operands up to 99, level 2 up to 99 × 9 (2-digit × 1-digit numbers, as in the Kit), level 3 numbers up to 30
  • Practice: 4 problems with feedback (no published value)
  • Seed: empty for a new form per participant, or a number for the same problems for everyone

Outputs collected

  • Correct answers per minute: number correct ÷ time spent on the scored problems (Roy et al., 2025)
  • Correct answers and problems answered
  • Accuracy per problem family: addition, subtraction, multiplication, division, order of operations, brackets, percentages, direct and inverse proportion
  • Median time to ✓ of correct answers
  • Every problem: level, family, the problem shown (expression, or word-problem template and numbers), the answer, the typed response, correct, omission, time to ✓, time to the first key, deletions, and for level 3 whether a wrong answer equals the result of ignoring the order of operations or the brackets
  • The seed, so the problems of a run can be rebuilt

Interpretation notes

Correct answers per minute combines speed and accuracy: the number correct divided by the time spent on the scored problems (from the problem's onset to the touch on ✓, or to its time limit), the correct responses per minute of Roy et al. (2025), after Vandierendonck (2017). In a fixed-time block the problem on screen when the time runs out is kept with what was typed but not scored. The families differ in difficulty, so compare a family across sessions or people rather than families with each other. Typing adds to the time per problem, so answers with more digits take longer. Differences from the published procedures: problems appear one at a time instead of on printed pages (Kit) or in columns (Tempo-Test-Rekenen); level 1 adds two numbers, not three as the Kit addition test does; levels 3 and 4, the practice and the feedback have no published counterpart in this format. Responses are timed from the drawn problem to the touch-down, but this timing has not yet been validated on real devices.

Scientific evidence

  • Number facility, the speed and accuracy of simple arithmetic, is measured with timed tests of the basic operations: the Number Facility tests of the ETS Kit (addition; subtraction and multiplication; division; two pages of 30 problems, 2 minutes each) and the Tempo-Test-Rekenen (five columns of 40 problems, 1 minute each).
  • A tablet task in the same format (one problem at a time, answers typed on an on-screen keypad, 3 minutes) gave a correct-per-minute score with a split-half reliability of .97 (.92 for the first minute) in 3rd to 7th graders, and it predicted their scores on a state mathematics test.
  • A 90-second smartphone arithmetic test with an on-screen keypad has been used in daily life to assess cognitive fatigue; in that study only the reaction time in a vigilance task, not the arithmetic score, was a significant predictor of self-reported mental fatigue.

Timing and session conditions

Each run of this task records how its times were measured and the conditions of the session, such as screen brightness, battery saving and interruptions. Runs with possible problems are flagged on the Quality tab and in the export.

How SMAAT times cognitive tasks

Read more

Related tasks

Add Mental Arithmetic to your ESM study on SMAAT — free to get started.