IIF Certified Assessment
Start IQ Test
IIF Certified Assessment
The largest synthesis of the question pooled 88 imaging studies and put the association between brain volume and IQ at roughly 0.24. That is about six per cent of the variation in scores, which is enough to rule out the strong claim and the flat denial at the same time.
Of all the folk claims about intelligence, "bigger brain, smarter person" is the one that sounds most obviously testable. It has been tested — for over a century by measuring skulls, and for the last thirty years by measuring living brains directly with magnetic resonance imaging. There is an answer. It is a number, it is positive, and it is small enough that the people who repeat the claim and the people who wave it away are both wrong in interesting ways.
In 2015 a team led by Jakob Pietschnig published in Neuroscience and Biobehavioral Reviews the largest synthesis of the question available. They gathered 88 studies reporting effect sizes across 148 samples and more than 8,000 individuals, restricted to brain volume measured in living people by imaging rather than inferred from head or skull size, and pooled the correlations. The headline figure was a correlation of about 0.24, and it held across children and adults, across full-scale, verbal and performance scores, and across sexes.
A correlation is a single number between 0 and 1 describing how tightly two measurements move together — 0 meaning no relationship at all, 1 meaning one is perfectly predictable from the other. A correlation of 0.24 is real and positive: brain volume and IQ do travel together. The way to feel its size is to square it. That gives roughly 0.06, which is conventionally read as brain volume corresponding to something in the region of six per cent of the variation in scores across people. The remaining ninety-four per cent is accounted for by everything else, measured and unmeasured.
A correlation of 0.24 is a fact about a population. It is not a fact about anyone in it, and it was never capable of becoming one.
The 2015 figure did not settle the field. Writing in Intelligence in 2017, Gignac and Bates argued that the quality of the intelligence measurement itself moderates the result: studies that administered fuller, better-constructed IQ batteries recovered noticeably stronger associations than studies relying on a brief screening measure, and on their reading the corrected estimate sits closer to 0.4. That is a substantially bigger effect — around sixteen per cent of the variance rather than six.
A 2022 paper in Royal Society Open Science took the disagreement seriously enough to run it systematically, working through a large space of defensible analytic choices — which samples to include, how to correct for measurement error, how to handle clinical groups — and showing how far the pooled answer travels depending on which combination you adopt. The honest summary is that the true value lies somewhere in a band, that the band excludes zero, and that it also excludes anything strong enough to justify the popular version of the claim.
Total volume is an extremely coarse summary of an organ whose function depends on organisation. As Deary, Penke and Johnson set out in their 2010 review in Nature Reviews Neuroscience, the imaging correlates of intelligence differences are distributed rather than local, and involve properties gross volume cannot see: the integrity of the white-matter tracts connecting regions, cortical thickness in particular areas, the efficiency with which networks are recruited under load. Two brains of identical volume can differ on all of it.
That last point connects to a habit worth having generally. The reason a single striking imaging study should never move your view very far is the same reason a single test sitting should not — both are estimates with error attached, and our explainer on why a score is a range rather than a number is the short version of the argument. The pooled result is the one that matters, and pooled results in this area are consistently modest.
It licenses a general statement: across large samples, larger measured brain volume is associated with slightly higher scores on tests of general ability, and the association is robust enough to have survived repeated meta-analysis. It does not license an inference about you, your child, or a public figure whose head looks a certain size in photographs — the same category error that produces the numbers we examined in where celebrity IQ figures come from. And it says nothing about what any individual can do, because a correlation between two measurements is silent on what would happen if either one changed.
If the question underneath the curiosity is where your own reasoning actually sits, that is a measurable thing and it does not require a scanner. A properly normed assessment places a score against a reference sample, and a percentile converts that position into the share of people who scored lower — a far more informative number than any anatomical one. Our IQ percentile calculator does the conversion, and you can take the assessment to get a score to convert. Reading the result as a band rather than a point is the discipline that makes it worth having.
Anatomy is associated with about six per cent of the differences between people on these tests. Your own result tells you rather more. Take a properly normed assessment and read the percentile it produces.
Find your IQ score now! →The century-long appeal of the brain-size claim is that it promises a physical fact underneath a contested psychological one. The measurements now exist and the promise did not survive them. What survived is a small, real, much-argued-over association that tells you something about populations and almost nothing about persons — which is, in the end, the shape of most honest findings about human intelligence, and the reason the interesting work has moved to how brains are wired rather than how large they are.
Across large samples there is a positive association, but a modest one. The largest meta-analysis, published in Neuroscience and Biobehavioral Reviews in 2015, pooled 88 studies and reported a correlation of about 0.24 between in-vivo brain volume and IQ, corresponding to roughly six per cent of the variation in scores. Later work has argued the figure may be higher when intelligence is measured well, but no estimate in the literature is strong enough to predict an individual.
It means the two measurements tend to rise together, faintly. Squaring the correlation gives about 0.06, conventionally read as brain volume corresponding to around six per cent of the differences in IQ scores between people. Ninety-four per cent of the variation is associated with other things, so the relationship describes a broad tilt across a population rather than a rule you could apply to two individuals.
No. Average brain volume is larger in men by roughly a tenth, but well-constructed adult IQ batteries are built and normed to show little or no mean difference in full-scale scores between men and women. That mismatch is one of the standard reasons researchers treat the volume and IQ correlation as something other than a simple causal chain.
The existence of a positive association is well established; its size is not. The 2015 meta-analytic estimate of about 0.24 was challenged in 2017 by work arguing that better intelligence measurement yields a stronger figure nearer 0.4, and a 2022 multiverse analysis in Royal Society Open Science showed the pooled result shifts considerably with defensible changes to the analysis.
Corrections: spotted an error? Email corrections@iqmetrics.org and we will update this story and note the change here.
Scores on tests that look nothing alike still correlate positively with one another. g is the factor pulled out of that pattern — which is what makes a single summary number defensible, and also what makes it far less than the whole story.
The correlation between them is one of the sturdier findings in cognitive psychology. How strong it is, and what it implies about training either one, is where the agreement stops.
A full-scale IQ is not one thing a test measures directly. It is a composite built from several index scores, and the theory that decides which ones is called Cattell-Horn-Carroll — CHC for short.
Our IIF-certified assessment reports your score with its scale, percentile and confidence range — and a breakdown of the cognitive domains behind it.
Start IQ Test →
