Education

Mental Math Trainer

Practise mental arithmetic against a timer, with difficulty you set, immediate feedback, and a session summary showing accuracy and median time per question.

Last reviewed by the Radiatus Cloud team

Press start to begin.
A summary appears when the session ends.

Need this done properly for your business?

Radiatus delivers secure cloud, DevOps & compliance engineering.

Book a free consult

Retrieval, not review

Arithmetic fluency comes from retrieval practice: producing an answer from memory, being told immediately whether it was right, and repeating. Reading through worked examples feels more productive and is measurably worse, because recognising a correct answer is a different and easier task than generating one. This is one of the better-replicated findings in learning research and it applies as much to a fourteen-year-old practising times tables as to anyone else.

Speed matters because working memory is small

The point of automatic recall is not speed for its own sake. Working memory holds only a few items, so a student who must reconstruct 7 × 8 by counting has spent that capacity before reaching the actual problem. Fluent recall frees the capacity for the reasoning the question was about. That is why a multi-step word problem collapses for students whose arithmetic is accurate but slow.

Track the median, not the mean

Timing per question is noisy: a single interruption produces a thirty-second outlier that drags a mean upward and hides real improvement. The median is stable against that, which is why this tool reports it alongside accuracy. Improvement usually shows as the median falling while accuracy holds, and if accuracy falls as the median does, the difficulty is set too high.

Related tools

Frequently Asked Questions

Why does answering from memory beat reviewing worked examples?

Because recognising a correct answer is an easier task than generating one, so review feels productive while producing weaker recall. Retrieval practice is one of the better-replicated findings in learning research.

Does speed matter or just accuracy?

Both, because working memory is small. A student reconstructing 7 × 8 by counting has spent that capacity before reaching the actual problem, which is why multi-step questions collapse for students whose arithmetic is accurate but slow.

Why report the median time rather than the average?

A single interruption produces an outlier that drags an average upward and hides real improvement. The median is stable against that.

How long should a session be?

Short and frequent beats long and occasional. Five minutes daily produces more durable recall than half an hour once a week, because spacing is what drives retention.

What if accuracy drops as speed improves?

The difficulty is set too high. Reduce the range until accuracy sits above roughly 90 percent, then let the median time fall before raising it again.

Privacy & Security

Everything runs in your browser; nothing is uploaded.

Data: None
Client-side-Side
Active
v1.0

How to Use

Choose the operations and difficulty, then answer each question.

Disclaimer: This tool is provided "as is" without warranty of any kind. Results are for educational and utility purposes.