Spatial Reasoning: The Ability Most Tests Ignore
- Written by
- IQCognify Editorial Team
- Reviewed for accuracy
- IQCognify Research Review Process
- Last updated
- Reading time
- 6 min
Quick answer
Spatial reasoning is the capacity to represent, transform and reason about objects in space — rotating a shape in your head, folding a net into a cube, reading a cross-section. It predicts who ends up in science and engineering better than the tests used to select them, and it is almost never assessed by the tests used to select them. It is also the one cognitive ability with solid meta-analytic evidence that training works.
What spatial reasoning is
Spatial reasoning is not one thing. It is a family of related capacities, and the distinctions between them matter because they train, predict and correlate differently.
| Component | What it involves | Classic task |
|---|---|---|
| Mental rotation | Turning a rigid object in imagination | Deciding whether two rotated shapes are identical or mirrored |
| Spatial visualisation | Multi-step transformation of a form | Folding a flat net into a solid; paper-folding tests |
| Spatial perception | Locating the horizontal or vertical despite distraction | Water-level and rod-and-frame tasks |
| Spatial working memory | Holding locations in mind while doing something else | Corsi block-tapping; spatial span |
In the Cattell–Horn–Carroll taxonomy, visual processing is one of the broad abilities beneath general intelligence. It appears in modern batteries as its own index — the WAIS-5 and WISC-V both report a Visual Spatial index — and it is measured by tasks such as block design and visual puzzles.
Fluid vs crystallized intelligence — where spatial ability sits in the broader structure of cognitive abilities.The neglected predictor of scientific achievement
In 2009 Jonathan Wai, David Lubinski and Camilla Benbow assembled fifty years of evidence around a single question: does spatial ability predict who becomes a scientist or engineer?
Their central analysis drew on Project Talent, a stratified random sample of roughly 400,000 US high-school students in grades 9 through 12, followed for eleven years. They aligned those results with pre-1957 findings, with contemporary Graduate Record Examination data, and with the Study of Mathematically Precocious Youth.
The finding
Spatial ability measured in adolescence was a salient attribute of the people who went on to earn advanced credentials and hold occupations in STEM — and it predicted this over and above mathematical and verbal ability, which were already in the model.
The word doing the work is over. Spatial ability is not merely correlated with STEM success because both correlate with general intelligence. It adds prediction that mathematical and verbal ability do not supply.
Why this is a problem
Consider what the gatekeeping tests measure. The SAT and ACT assess mathematical and verbal ability. Graduate admissions rest largely on the same two dimensions. University entrance in most countries follows suit.
A student with exceptional spatial ability and merely good verbal ability is therefore invisible to the selection system, in precisely the fields where spatial ability matters most. The authors described spatial talent as a sleeping giant for identification and development — an ability that predicts scientific achievement and is systematically not looked for. It is also the reason the aptitude test versus IQ test distinction still has any practical content left in it.
Spatial skills can be trained — and this is not the brain-training claim
Here the evidence diverges sharply from what we find elsewhere in cognitive training, and the distinction is worth drawing carefully.
Uttal and colleagues meta-analysed 217 training studies for Psychological Bulletin, examining the magnitude, moderators, durability and generalisability of spatial training. After removing outliers, the average effect of training relative to control was Hedges's g = 0.47 — a moderate effect, and a real one.
| Question | Answer |
|---|---|
| Studies analysed | 217 |
| Average effect size | g = 0.47 (SE = 0.04) |
| Did the gains persist? | Yes — effects were stable and unaffected by delay before post-testing |
| Did the gains generalise? | Yes — training transferred to untrained spatial tasks |
The distinction that must not be blurred
Uttal found transfer to other spatial tasks. That is within-domain transfer. It is not evidence that spatial training raises general intelligence, and it is not the far transfer that brain-training products claim. Both statements are true at once: spatial skills are genuinely malleable, and the fluid-intelligence gains promised by cognitive-training software remain undemonstrated.
Reading the two literatures together gives a coherent picture rather than a contradictory one. Practice reliably improves the specific capacity practised, and it generalises across tasks that share the underlying representation. What it does not do is climb the hierarchy of abilities to move g. Spatial training makes you better at spatial things — durably, and across spatial tasks you never trained on. That is a considerable and useful result, and it is a different claim from making you smarter.
Do brain training games work? — why near transfer is real, far transfer is not, and the difference between them is the whole argument.The sex difference in mental rotation
Mental rotation produces one of the larger and more reliably replicated average sex differences in cognitive psychology, favouring men. It is worth being precise about what that does and does not mean.
It is an average difference between two heavily overlapping distributions, on one narrow task within one component of one broad ability. It says nothing about any individual. And the malleability evidence is directly relevant: Uttal and colleagues examined sex as a moderator of training effects, and spatial skills respond to training in everyone.
Why this matters for the STEM finding
If a trainable ability predicts entry into a field, and that ability is unevenly distributed at the outset, and nobody is taught it explicitly — then the pipeline is selecting on something it could have been developing instead. That is the practical argument Wai's group and Uttal's group make together.
How spatial reasoning is measured
- Block design and visual puzzles form the Visual Spatial index of the WAIS-5 and WISC-V.
- Mental rotation tasks present pairs of three-dimensional figures and ask whether one is a rotation of the other or a mirror image.
- Paper-folding and surface-development tasks measure spatial visualisation over multiple transformations.
- Matrix reasoning tests, including Raven's, use visual material but load primarily on fluid reasoning rather than spatial ability specifically.
That last point is the most common confusion. A test being visual does not make it a spatial test. Raven's Progressive Matrices presents figures, but what it requires is rule induction, not mental transformation of an object through space.
Raven's Progressive Matrices — a visual test that is not a spatial test, and what it was built to measure instead.Myths and facts about spatial reasoning
| Myth | Fact |
|---|---|
| Spatial ability is just a facet of general intelligence | It predicts STEM achievement over and above mathematical and verbal ability, both already in the model. |
| Admissions tests capture spatial talent | The SAT and ACT assess mathematical and verbal ability. Spatial ability is largely unmeasured in selection. |
| Spatial skills are fixed | Uttal's meta-analysis of 217 studies found an average training effect of g = 0.47, durable and generalising to untrained spatial tasks. |
| Spatial training raises IQ | It produces within-domain transfer to other spatial tasks. Far transfer to general intelligence has not been demonstrated. |
| Any visual test is a spatial test | Raven's Matrices is visual but loads on fluid reasoning. Spatial ability specifically involves transforming objects in space. |
| The sex difference in mental rotation means men are better at STEM | It is an average difference on one narrow task between heavily overlapping distributions, on an ability that responds to training. |
Frequently asked questions
What is spatial reasoning?+
Spatial reasoning is the ability to represent, transform and reason about objects in space — mentally rotating a shape, folding a flat net into a solid, or interpreting a cross-section. It comprises several related components, including mental rotation, spatial visualisation, spatial perception and spatial working memory, and appears in the Cattell–Horn–Carroll model as the broad ability of visual processing.
Does spatial ability predict success in science and engineering?+
Yes, and it does so beyond mathematical and verbal ability. Wai, Lubinski and Benbow (2009) followed a stratified random sample of 400,000 US high-school students for eleven years and found spatial ability measured in adolescence was a salient attribute of those who later earned advanced STEM credentials and held STEM occupations, with mathematical and verbal ability already accounted for.
Why is spatial ability called a neglected dimension?+
Because the tests used to select students for scientific fields do not measure it. The SAT and ACT assess mathematical and verbal ability, as do most graduate and university admissions systems. A student with exceptional spatial ability and merely good verbal ability is therefore invisible to selection in exactly the fields where spatial ability predicts success.
Can spatial reasoning be improved?+
Yes. Uttal and colleagues (2013) meta-analysed 217 training studies and found an average effect of Hedges's g = 0.47 relative to control. The gains were durable — unaffected by delay before post-testing — and they transferred to spatial tasks that had not been directly trained.
Does spatial training increase IQ?+
No, and it is important not to blur this. Uttal's evidence shows transfer to other spatial tasks, which is within-domain transfer. It is not the far transfer to general intelligence that brain-training products advertise, and that claim remains unsupported. Spatial skills are genuinely malleable; general intelligence has not been shown to move with them.
Is Raven's Progressive Matrices a spatial reasoning test?+
Not primarily. Raven's uses visual material, but the task is rule induction — inferring abstract relationships that govern a pattern — rather than mentally transforming an object through space. A test being visual does not make it a spatial test, and this is the most common confusion in the area.
How is spatial ability measured on an IQ test?+
The WAIS-5 and WISC-V both report a Visual Spatial index, derived from tasks such as block design and visual puzzles. Dedicated spatial measures used in research include mental rotation tasks, paper-folding and surface-development tests, and the Corsi block-tapping task for spatial working memory.
Are men better at spatial reasoning than women?+
On mental rotation specifically, there is a reliably replicated average difference favouring men — between two heavily overlapping distributions, on one narrow task within one component of one broad ability. It tells you nothing about any individual, and spatial skills respond to training regardless of sex.
Sources
This guide draws on standard psychometric references and peer-reviewed research:
- 1.Wai, J., Lubinski, D., & Benbow, C. P. (2009). “Spatial ability for STEM domains: Aligning over 50 years of cumulative psychological knowledge solidifies its importance.” Journal of Educational Psychology, 101(4), 817–835.
- 2.Uttal, D. H., Meadow, N. G., Tipton, E., Hand, L. L., Alden, A. R., Warren, C., & Newcombe, N. S. (2013). “The malleability of spatial skills: A meta-analysis of training studies.” Psychological Bulletin, 139(2), 352–402.
- 3.Carroll, J. B. (1993). Human Cognitive Abilities: A Survey of Factor-Analytic Studies. Cambridge University Press.
- 4.McGrew, K. S. (2009). “CHC theory and the human cognitive abilities project.” Intelligence, 37(1).
- 5.Melby-Lervåg, M., Redick, T. S., & Hulme, C. (2016). “Working memory training does not improve performance on measures of intelligence or other measures of ‘far transfer’: Evidence from a meta-analytic review.” Perspectives on Psychological Science, 11(4).
- 6.Pearson — Wechsler Adult Intelligence Scale, Fifth Edition (WAIS-5), 2024.
- 7.Deary, I. J. (2020). Intelligence: A Very Short Introduction (2nd ed.). Oxford University Press.
- 8.American Psychological Association (APA)
Sources are provided for further reading. Organization links point to official sites; academic works are cited in full. See our research standards and editorial team.
Find out your IQ
Take the free IQ test and get your score, percentile, and a full cognitive breakdown in about 12 minutes.
Start Free Test