Free spatial reasoning test

What spatial reasoning is

The APA Dictionary defines spatial ability as reasoning about objects in space, and gives map reading and jigsaw puzzles as examples (APA Dictionary: spatial ability). The part most tests focus on is mental rotation: turning a figure away from its normal orientation in your head, a skill first measured with Roger Shepard's laboratory task (APA Dictionary: mental rotation). If you want the theory in full, our page on spatial intelligence covers it. This page is about the test and how to get better at the puzzles.

Spatial ability has a place in almost every model of intelligence. It was one of Thurstone's primary mental abilities (his factor “S”) and is one of Gardner's multiple intelligences (APA Dictionary: spatial intelligence). In the Cattell-Horn-Carroll (CHC) model it falls under visual processing, or Gv, which covers tasks such as mentally rotating a shape or picturing how a folded sheet of paper would look (APA Dictionary: Cattell-Horn theory). Gv is one of the broad abilities that sit alongside fluid reasoning (Gf), comprehension-knowledge (Gc), working memory and processing speed (McGrew et al., 2023). In a 2023 reanalysis of Woodcock-Johnson data, the Spatial Relations and Block Rotation tests were the stronger indicators of visualization, a major narrow ability within Gv (McGrew, 2023).

Spatial reasoning is not walled off from the rest of reasoning. One test battery's authors note that spatial reasoning is also associated with fluid reasoning (Primi et al., 2013), and a 2024 paper notes that better mental rotation performance goes with higher fluid intelligence and better mathematical thinking (Stark et al., 2024). That is why spatial items show up inside general IQ tests rather than only in specialist batteries.

Two classic findings explain what your brain is doing on a rotation item. Shepard and Metzler showed that the time people needed to decide whether two drawings showed the same 3D object rose in a straight line as the angle between them grew (Shepard & Metzler, 1971). In other words, people seem to turn the picture step by step, and bigger turns take longer. The same study found that turning an object in depth took no longer than turning it flat on the page. The ability also matters outside tests: in a group of 563 intellectually talented 13-year-olds followed for more than 30 years, SAT math and verbal scores explained 10.8% of the variance in later patents and refereed publications, and adding spatial ability explained another 7.6% (Kell et al., 2013).

How spatial reasoning is measured

Clinical IQ batteries give spatial ability its own score. The WAIS-5 has five primary index scales, one of which is Visual Spatial, measured by the Block Design and Visual Puzzles subtests (Pearson: WAIS-5 overview). That is new: the WAIS-5 split the old perceptual domain of the WAIS-IV into separate Visual Spatial and Fluid Reasoning index scores (Pearson: WAIS-5). The children's version, the WISC-V, also reports a Visual Spatial index (Pearson: WISC-V), and when 225 clinic children were retested after an average of 2.6 years, Visual Spatial was one of only three scores whose stability coefficient exceeded 0.80 (Watkins et al., 2022). The Stanford-Binet 5 measures Visual-Spatial as one of its five factors (PRO-ED: SB5).

Research and specialist tests go further. Vandenberg and Kuse turned the Shepard and Metzler figures into a paper-and-pencil Mental Rotations Test with an internal consistency of .88 and a test-retest reliability of .83, and its scores correlated strongly with other spatial tests and hardly at all with verbal tests (Vandenberg & Kuse, 1978). The ETS kit of reference tests splits the area in two: a Visualization factor measured by form board, paper folding and surface development tests, and a Spatial Orientation factor measured by card rotation and cube comparison tests (ETS: Kit of Factor-Referenced Cognitive Tests). The nonverbal UNIT2 includes a Cube Design subtest (PRO-ED: UNIT2).

Spatial items also appear in employment and selection tests. The CCAT mixes verbal, math and logic, and spatial reasoning questions (Criteria: CCAT prep), with 50 questions in 15 minutes, so speed on rotation items matters there too. For an overview of how all these batteries differ, see types of IQ tests.

The ten questions above use two of these formats. Six are mental rotation items: one figure, four options, and every wrong option is a mirror image. Four are odd-one-out sets: six figures, five share a hidden rule, one breaks it. The easy and medium rotations are flat figures turned in the page; the two hard ones are clusters of cubes turned in space.

Worked examples: easy, medium and hard

Start with an easy odd-one-out set. The figures differ in size, tilt and fill on purpose, and none of that matters here. Count the straight sides of each figure and look for the one count that does not match.

The medium rotation is a flat figure turned in the page. Do not try to turn the whole thing at once. Pick one landmark, such as the marked cell or the end of the longest arm, find where it lands in each option, and then check which side the neighboring cells sit on.

The hard rotation is a cluster of cubes seen from a new angle. Treat it as a chain. Count the cubes in each arm, then walk from one end and note whether each bend goes left or right, up or down. The mirror images match almost everywhere but bend the wrong way at one point.

Common mistakes on spatial puzzles

• Picking the option that looks closest. Every wrong rotation option is a mirror image, so it looks nearly identical. The question is whether a turn can reach it, not whether it resembles the original.
• Turning everything at once. Holding a whole figure in mind is slow and error-prone. Track one landmark, then confirm with one neighboring part.
• Mixing up clockwise and counterclockwise. A quarter turn clockwise sends the top to the right. If you lose track, go back to the landmark rather than guessing the direction.
• Stopping at the first match on cube clusters. Mirror images of a cube cluster share the same arm lengths. Only the direction of one bend gives them away, so check every bend before you choose.
• Reacting to size, tilt or fill in odd-one-out sets. These are distractors. The real rule is a side count, or, in the hardest sets, which region of two overlapping outlines the dot sits in.
• Choosing without stating the rule. Before you pick an odd one out, say the rule in a short sentence, such as “five have four sides.” If you cannot, you have probably found a figure that only looks different.
• Leaving a question blank. Each question has its own 30-second ring. Once three options are clearly mirror images, the last one is your answer even if you have not fully turned it.

Pacing matters too. Spend the first few seconds on the original figure, not the options: you cannot compare well until you know what you are looking for. Remember Shepard and Metzler's finding that bigger turns take longer to check; a half turn is worth a slower look than a quarter turn. If a question is still unclear at about 20 seconds, make your best choice, because a guess has a chance of being right and a blank does not.

How to practice spatial reasoning

Spatial skill responds to training more clearly than most abilities. A meta-analysis of 217 spatial training studies found an average effect of g = 0.47 compared with controls, and the gains were stable over delays and transferred to spatial tasks that were not trained (Uttal et al., 2013). A 2022 meta-analysis found spatial training improved spatial thinking by g = .49 and math performance by g = .28, and training with concrete materials worked better than training that did not use them, such as computer-only practice (Hawes, Gilligan-Lee & Mix, 2022). A practical plan:

1. Learn the method on easy items. Work through the easy rotation puzzles using the landmark method until it feels automatic.
2. Add the clock. Move to the medium rotations and give yourself 30 seconds each, the same limit as this test.
3. Train the hard format separately. The hard cube clusters need the bend-by-bend method. Do a handful in a row so the habit sticks.
4. Practice rule-finding. The odd-one-out library has easy, medium and hard sets; the hard ones use the dot-position rule.
5. Use physical objects too. Building with blocks, folding paper and doing jigsaw puzzles are hands-on spatial practice, and training with concrete materials did better than computer-only training.
6. Read the method once. The spatial reasoning guide explains both formats in more depth, and the full mental rotation library has worked answers for every item.
7. Mix it up. Try the non-verbal reasoning test, which adds sequences, matrices and an analogy to the spatial formats.

Be clear about what practice can and cannot change. Retaking a cognitive test raises scores by about a quarter of a standard deviation on average, more when the same form is reused (Hausknecht et al., 2007), and a large meta-analysis found no further gains after the third sitting (Scharfen, Peters & Holling, 2018). So a rising score on these same ten items mostly shows familiarity. Spatial training does carry over to other spatial tasks, but reviews of brain training find that effects shrink as study quality rises and that far transfer to general ability rarely occurs (Sala & Gobet, 2017). And shortcuts do not work: a meta-analysis of about 40 studies found no support for gains in spatial ability due to listening to Mozart (Pietschnig et al., 2010).

Spatial reasoning on the full IQ Octopus test

The full test has 25 items in five sections: Pattern Recognition (5), Abstract Reasoning (9), Spatial Reasoning (5), Quantitative Reasoning (3) and Logical Reasoning (3). The Spatial Reasoning section uses the same two formats you just practiced, mental rotation and odd one out, but the items are generated fresh for every attempt, so there is nothing to memorize. The other four sections cover sequences, matrices, number series and figure analogies.

Your practice score here is a count out of ten, nothing more. Real IQ scores are deviation scores that compare you with people your own age on a scale with a mean of 100 and a standard deviation of 15 (Wikipedia: Intelligence quotient), and ten items in one area cannot place you on that scale. The full test scores all 25 answers with a Rasch model and reports an estimated IQ with a confidence range. When you are ready, take the full IQ test.