How does working memory affect mental math — and can I expand it?
Working memory is the small mental workspace where you hold information while you act on it. In mental math it is where the running total, the carry, and the next step all sit at once, and it is easy to overload.
Its limits are real and fairly fixed, but that does not doom you to slow arithmetic. The trick is not to expand the workspace; it is to put far less in it, by making facts automatic and choosing methods that hold fewer pieces.
What is working memory, in plain terms?
Working memory is your short-term holding space for active information. It keeps a few items available and lets you manipulate them, but only for a short time and only a few at once.
When you compute 38 × 7 in your head, working memory is what holds the 210 while you work out 56 and then add them. If anything bumps that 210 loose, the whole calculation collapses.
Why does working memory limit mental calculation?
Every partial result you must remember occupies a slot, and you have only a handful. Long methods with many intermediate values fill the space fast, and once it is full you start dropping pieces and making errors.
This is why a calculation that looks simple on paper feels hard in your head. On paper the page holds the partials; in your head you must hold them yourself, against a clock and a limited buffer.
What is chunking and how does it help?
Chunking means grouping small pieces into one larger, meaningful unit so it takes a single slot instead of several. The phone number 1-4-9-2 is four items, but '1492' as a year is one chunk.
In arithmetic, knowing 25 × 4 = 100 as one fact lets you treat it as a single chunk rather than a multi-step sub-problem. The more facts you own as whole chunks, the less your workspace has to juggle.
How do I reduce the load of a given calculation?
Pick the method that leaves the fewest partial results alive at once, and work left-to-right so you can release each piece as you speak or settle it. Round and adjust when it shortens the path.
For 38 × 7, hold the size first: 40 × 7 = 280, then subtract 2 × 7 = 14 → 266. You only ever held one number (280) and one small correction (14), instead of stacking several partials.
Does it help to say the numbers out loud or picture them?
Both can lighten the load, and the right one is personal. Saying a partial result out loud or subvocalizing it can hold it steady while your hands and eyes are free. Visualizing the digits works better for some people.
Try each and keep whichever frees up the most space for you. The aim is the same: park a value somewhere reliable so your active workspace is open for the next step.
Can I actually expand my working memory by training it?
Be honest with yourself here: general working-memory training tends to make you better at the training task, but the gains rarely transfer to unrelated skills like arithmetic. Chasing a bigger buffer is mostly a dead end.
The real win is offloading load. When your facts are automatic, recalling 7 × 8 = 56 costs almost no working memory, leaving the whole workspace for the parts that genuinely need thought.
So train the facts, not the buffer. SIXTY builds automaticity through short timed retrieval and replays the facts you miss, which is the path that actually frees up mental room.
- Memorize core facts so recall replaces in-head calculation.
- Choose methods that keep the fewest partials alive at once.
- Work left-to-right and release each piece as it settles.
- Park values by speaking or picturing them, whichever holds better.
- Practice short and daily so automaticity keeps growing.
Reading is review. Recall is what sticks.
Start a 60-second sprint →