Does Abacus Increase IQ? What the Research Actually Says

"Will abacus increase my child's IQ?" is one of the most common questions we get, usually because a brochure somewhere has promised exactly that.

The honest answer has three parts: what IQ actually measures, what abacus training reliably changes, and why those two things overlap less than the marketing suggests.

What IQ tests measure

An IQ score is not a single quantity. Standard tests combine several sub-scores, typically including:

• Verbal comprehension — vocabulary, verbal reasoning
• Visual-spatial reasoning — mental rotation, pattern completion
• Fluid reasoning — solving novel problems without prior knowledge
• Working memory — holding and manipulating information
• Processing speed — how quickly simple tasks are completed

The composite is designed to be stable over time. That is the point of the measure — it is meant to reflect a durable capacity, not something that shifts after a term of classes.

That design goal is exactly why claims of large IQ jumps deserve scepticism.

What abacus training reliably changes

The research literature is clearest on these:

Visual-spatial working memory. Trained mental-abacus users hold and manipulate more visual information than untrained peers. This is the strongest and most replicated finding.

Numerical processing. Faster, more accurate arithmetic, including on problems not directly practised.

Sustained attention. Maintaining a mental image under load is attention work, and it improves with practice.

Where arithmetic happens in the brain. Imaging studies show experienced users recruiting visual-spatial regions rather than the language regions typically used for calculation.

Notice that two of these — working memory and processing speed — are IQ sub-components. So there is a genuine overlap.

So does the composite score move?

Here is the careful answer.

Training can improve performance on the specific sub-tests closest to what was trained, particularly working memory and processing speed. Some studies find modest gains on those.

But the composite IQ score is dominated by fluid reasoning and verbal comprehension, and those show little reliable change from abacus training. A child does not become better at vocabulary or at novel logical puzzles because they can visualise beads.

So: a small, domain-specific improvement, not a transformation. Anyone promising a 20-point jump is not describing research.

The transfer problem

This is the central issue in all cognitive-training research, and it is worth understanding because it applies to brain-training apps, memory courses and abacus alike.

Near transfer — improvement on tasks similar to the trained one — is common and well documented. Abacus training improves arithmetic and visual-spatial memory. That is near transfer, and it is real.

Far transfer — improvement on unrelated tasks — is rare and weakly supported. Abacus training improving reading comprehension or general reasoning would be far transfer, and the evidence for it is thin.

Most exaggerated claims are far-transfer claims dressed up as near-transfer findings.

Why the honest version is still a good case

Dropping the IQ claim does not weaken the argument for abacus. It sharpens it.

Working memory is one of the strongest predictors of academic performance across subjects. A child who can hold more in mind while working through a problem does better at maths, at science, at reading comprehension — not because those skills were trained, but because the bottleneck was capacity.

Attention is similar. A child who can stay with one task for fifteen minutes learns more in every lesson they attend.

Those are the real returns, and they are worth having. They are simply not the same thing as an IQ score.

What this means for choosing a programme

Be wary of any centre that:

• Promises specific IQ point gains
• Uses "whole-brain development" as its main claim
• Cannot tell you what your child will be able to do after 12 months
• Does not require daily practice

Be reassured by one that:

• Talks about concentration, working memory and confidence
• Has a clear level structure with specific skills at each stage
• Is honest that results need consistent practice
• Will tell you if your child is too young to start

Our science behind abacus and memory improvement goes deeper into the mechanism, and is abacus worth it covers the cost side.

Frequently asked questions

So abacus does nothing for intelligence? It improves specific cognitive capacities — working memory and attention — that support learning broadly. Whether you call that "intelligence" depends on your definition; it does not reliably move a standardised IQ composite.

Do the gains last? Calculation speed fades without practice. Working-memory and attention gains appear more durable, though they also decline over years without use.

At what age is training most effective? Roughly 6 to 12, when the visual-spatial system is most plastic.

Is more practice always better? Up to a point. Daily short sessions beat long weekend ones, and beyond about 20 minutes a day the returns fall off for most children.

Should I test my child's IQ before and after? We would not recommend it. The measure is not designed for that, the test-retest effect confounds it, and it puts a number on a child that is easily misread.

Judge it by what your child can do

The best evidence is in front of you: is your child calmer with numbers, more willing to attempt a hard question, holding instructions better?

Our abacus course is built around those outcomes, in live batches from age 6.

Book a free demo class and see for yourself.

Cover image: Edlow, Brian L., Massachusetts General Hospital Mareyam, Azma, Massach.webm) — CC0.