Spatial reasoning tests in two and three dimensions

The skill

What is a spatial reasoning test?

A spatial reasoning test measures how accurately you can manipulate shapes and objects mentally: turning them, reflecting them, folding them, viewing them from somewhere else, or combining them. You are given a figure and asked what it becomes, or which of several figures could be the same object seen differently.

The distinction to settle first is between a rotation and a reflection. Turning an object within its plane preserves the order of its features; mirroring it reverses that order. No amount of turning will convert a shape into its mirror image, which is what makes a reflected figure a usable wrong option.

Pearson describes its DAT Space Relations module as testing the ability to visualise two-dimensional shapes in three dimensions, which is the folding and unfolding half of the family. Criteria describes the spatial questions inside the CCAT as asking candidates to rotate or flip images in their head, recognise patterns and identify outliers, which is the comparison half.

Try it before anything else

Free spatial reasoning practice questions

Six original questions covering the formats above: nets, rotation against reflection, cube views and matching a view to an arrangement. Commit to an answer before opening the solution — each one shows the check that settles the question and names why every wrong option is on the list.

What employers are looking at

What do spatial reasoning tests measure?

Employers use spatial items for roles where physical or diagrammatic representations carry real work: engineering, design, manufacturing, construction, surveying, laboratory and technical apprenticeship roles. Saville groups spatial reasoning with mechanical and diagrammatic reasoning inside its technical battery for exactly that reason.

Two abilities separate here. Visualisation accuracy: can you hold an object stable while you turn it, rather than letting it drift. Rule use: do you fall back on reliable checks — face adjacency on a cube, handedness on a reflection, the two-apart rule on a net — instead of trying to picture the whole thing when the picturing gets hard.

Reading your result

How spatial reasoning tests are scored

A spatial result is rarely a simple count of right answers, and often it is not reported on its own at all. What the employer sees depends on the provider, and on whether your spatial questions stood alone or sat inside a wider battery.

• No provider reviewed here publishes a pass mark. Criteria sends the employer a raw score and a percentile for the CCAT as a whole, and any threshold belongs to the hiring organisation rather than the test. If a recruiter has disclosed a target, that is the only target worth working to.
• A battery total can hide the spatial section. The CCAT reports no separate spatial sub-score, and Saville reports one total plus individual sub-scores for its combination tests. A respectable overall figure can conceal a weak spatial section entirely, so record spatial accuracy separately when you practise — it is the only way to know whether the section deserves more of your remaining preparation time.
• On an adaptive test, difficulty is not feedback. Pearson states that the DAT Next Generation range uses computerised adaptive testing, so a correct answer leads to a harder question, and SHL's Verify Spatial Ability is computer-adaptive with 22 questions as a cap rather than a fixed count. Under either, you cannot read your score from how hard the questions feel, so treat rising difficulty as normal behaviour rather than as a verdict.
• Fixed-length tests save the hardest questions for the end. Saville states that its fixed-length format gives a gradual increase in difficulty, so the final questions are the most demanding ones. Pacing that spends everything early meets the hardest part of the test with the least time left, which is why protecting time for the end matters on this format.
• Completion is not the target. Criteria reports that fewer than 1% of candidates answer all 50 CCAT questions, with spatial questions interleaved among verbal and maths ones. Unanswered questions are designed into that format, and accuracy at a sustainable pace beats rushed completion.

Question formats

What do the questions look like?

Spatial assessments use a small number of shapes, and each has a check you can apply without visualising the whole object.

• Mental rotation. A figure is shown, then several candidates, one of which is the same figure turned. Reliable check: pick one asymmetric feature and track only that.
• Rotation against reflection. The odd option is a mirror image rather than a turn. Reliable check: read the order of three features clockwise. Rotation keeps that order; reflection reverses it.
• Cube nets and folding. A flat net to be folded into a solid, or a solid to be matched to its net. Reliable check: on any straight run of squares, faces two apart end up opposite each other.
• Cube views and adjacency. Several views of one marked cube, with a question about which view is possible. Reliable check: rotation never changes which faces touch, so any option that makes two adjacent faces opposite is out.
• Shape assembly and space relations. Pieces to be combined into a target shape, or a target to be decomposed. Pearson's DAT Space Relations sits in this area, testing the ability to visualise two-dimensional shapes in three dimensions.
• Matching different views. A three-dimensional arrangement seen from the front, side or above, with the question asking which plan or elevation belongs to it. Count along one axis at a time rather than judging the whole picture.

Launch coverage

Which providers use spatial reasoning tests?

Four launch families assess spatial ability, and only two of them do it as a standalone test. The rest fold it into a wider battery.

• DAT Space Relations. Pearson describes Space Relations as testing the ability to visualise two-dimensional shapes in three dimensions. It lists each DAT Next Generation module at 15 items and around 12 minutes, delivered untimed and unsupervised, and states that the range uses computerised adaptive testing so that a correct answer leads to a harder item.
• Spatial Reasoning. Saville publishes Spatial Reasoning as a single aptitude test with its own free practice test, and inside two batteries: Swift Technical Aptitude at 10 minutes with mechanical and diagrammatic reasoning, and Swift Apprentice Aptitude at under 20 minutes across six areas. Saville states that its fixed-length format gives a gradual increase in difficulty.
• CCAT spatial reasoning. Criteria lists spatial reasoning as one of three question kinds in the CCAT, and describes the task as rotating or flipping images in your head, recognising patterns and identifying outliers. The items are interleaved with verbal and maths questions inside 50 questions in 15 minutes, with no calculators allowed.
• Verify Spatial Ability. Verify Spatial Ability is one of the nine SHL tracks in Cognivy's launch scope. SHL's own assessment fact sheet gives a maximum of 22 questions in 15 minutes, computer-adaptive and designed for an unproctored environment, with a follow-up Verification test. Because it adapts, 22 is a cap rather than a fixed length, so the number you face may be lower. Your invitation is still the figure to trust if it differs.

Provider names identify the assessment format. Cognivy is independent and is not affiliated with or endorsed by these providers.

What varies, and what does not

Timing, delivery and scoring

Spatial items are rarely a test on their own. Most candidates meet them inside a battery, which is where the pacing decisions come from.

• Pearson DAT Space Relations: 15 items, about 12 minutes. Pearson lists those figures for each DAT Next Generation module and describes the delivery as untimed and unsupervised, with computerised adaptive testing selecting the next item from your last answer.
• Saville Swift Technical Aptitude: 10 minutes for three aptitudes. Spatial, mechanical and diagrammatic reasoning share a single 10-minute combination test, so the spatial section is a matter of minutes rather than a paper of its own.
• Criteria CCAT: spatial items inside 50 questions in 15 minutes. Criteria reports that fewer than 1% of candidates answer all 50 questions, and that no calculators are allowed. Spatial items are not grouped; they appear between verbal and maths questions.
• Fixed-length and adaptive behave differently. Saville states that its fixed-length format increases difficulty gradually, so the last items are the hardest and protecting time for them matters. Pearson's adaptive DAT selects difficulty from your answers instead. Under either model you cannot read your score from how hard the items feel, so treat rising difficulty as normal behaviour rather than as feedback.

A repeatable approach

A method that survives the clock

Rules settle the hard items that visualisation struggles with, and there are only a few worth knowing. Five steps.

1. 1

   Choose one anchor feature. A notch, an arrow, an asymmetric corner, one marked face. Track that feature through the transformation and let the rest of the object follow. Rotating a whole object mentally is where accuracy leaks away.

2. 2

   Check handedness before anything else. Read three features in clockwise order on the original and on the option. If the order reverses, the option is a reflection, and no rotation can produce it.

3. 3

   Use adjacency to eliminate. On any cube question, faces that touch keep touching however the cube is turned. Any option that shows two known-adjacent faces as opposite, or two known-opposite faces as adjacent, is out with no visualisation at all.

4. 4

   Apply the two-apart rule on nets. In a straight run of squares on a net, squares two apart become opposite faces. Squares that share an edge on the net can never be opposite on the cube. Those two facts settle most folding questions.

5. 5

   Eliminate first, visualise last. Work through the rules until at most two options survive, then do the one piece of mental rotation you actually need. The order matters: eliminating is cheap and rotating is expensive.

What slows progress

Common mistakes, and the fix for each

• Treating a mirror image as a valid rotation. Fix: check handedness on every item, before you rotate anything. A mirror image matches the original feature for feature and differs only in the order those features run, which is why reflections are the standard distractor and why they survive a quick visual comparison.
• Rotating the whole object at once. Fix: pick one anchor feature and follow it. Holding an entire figure stable in your head while turning it is a memory task, and it degrades exactly when the items get hard.
• Ignoring face adjacency on cube questions. Fix: list which faces touch before you consider any option. Rotation changes orientation and never changes adjacency, so the list is valid for every view of that cube.
• Reading a net as a flat picture. Fix: apply the two-apart rule rather than trusting how the squares look side by side. Neighbouring squares on the page become adjacent faces, not opposite ones, and that inversion is the whole point of the format.
• Practising the wrong dimension. Fix: match your practice to the assessment. Flat rotation drills do not prepare you for folding, and Pearson's Space Relations is explicitly about visualising two-dimensional shapes in three dimensions.

In practice

Rotation or reflection — what settles it?

The order of the features settles it, and nothing else does reliably. Turning a figure within the plane keeps the clockwise order of its parts; mirroring it reverses that order. No amount of rotation will turn a shape into its mirror image, so one check — read three features clockwise on the original, then on the option — separates the two transformations every time.

On a shape with two distinguishable arms the check collapses to a single swing: stand at the corner, point along the long arm, and note whether the sweep round to the short arm runs clockwise or anticlockwise. Sample questions 2 and 3 above both turn entirely on that swing, and it survives the situations that defeat eyesight — a tile lying at an unfamiliar angle still swings the same way, and a mirrored tile never does.

Reflections make the standard distractor precisely because they pass a visual comparison: a mirror image matches the original feature for feature and differs only in the order those features run. Corner position is the other false signal — two cards can share a corner position with only one of them a true rotation — so decide the check before you look at the options, and apply it to every card rather than to the one that looks wrong.

In practice

How do you fold a cube net in your head?

Mostly, you do not. Two facts replace the folding: on any straight run of squares, faces two apart end up opposite each other, and squares that share an edge on the net can never be opposite on the cube. A third follows from the shape of a cube itself — there are exactly three pairs of opposite faces, so once two pairs are fixed, the two remaining squares must form the last pair, wherever they sit on the page.

That elimination step is what settles staggered nets. In sample question 4, the two arm squares hang off different ends of the row and never line up on the page, yet they must fold to opposite faces because every other square is already paired by the straight-run rule. Page layout is the trap in this format: squares that look related on paper are often unrelated on the cube, and adjacency on the page is exactly what folding destroys.

A cube has eleven genuinely different nets, so the format cannot keep surprising you. Fold one real paper net, marked up, and watch which faces meet — the two-apart rule stops being something taken on trust. After that, reserve actual mental folding for the one join the rules leave undecided: it is the most expensive move available, so it should be the last one made.

In practice

How do you compare two views of the same cube?

Start from the one property rotation cannot touch: adjacency. However the cube is turned, faces that touch keep touching and opposite faces stay opposite. Each corner view proves three mutual adjacencies at once, so two views build a short list of established facts before any visualisation happens — and any offered view that contradicts the list is impossible, no rotation required.

When the two views share a face, you can usually prove an opposite pair outright. In sample question 5, both views keep the same face on top, so the cube has only been turned about its vertical axis, and the face arriving on the right must have come round from the back — which pins down the back of the first view, and with it a full opposite pair. One proven opposite pair then eliminates every option that shows those two faces together, and that is often most of the list.

Mental rotation comes last, if it comes at all. Eliminating by rule is cheap and rotating is expensive, so work the adjacency evidence until at most one contender is left, then make the single rotation needed to confirm it. The order is the method: candidates who rotate first are doing the hardest task in the family at the moment of maximum time pressure.

A clear route

How to prepare in the days you have

Spatial preparation puts physical rehearsal early and screen rehearsal late. This order fits a short run-up.

1. 1

   First, learn the four rules. Handedness, adjacency, the two-apart rule on nets, and the fact that a cube has eleven distinct nets. Each one turns a question you would otherwise visualise into a check you can apply, and the list is short enough to write out in full.

2. 2

   Next, fold something real. Draw a net on paper, mark the faces, fold it, and confirm which faces end up opposite. Folding a paper net makes adjacency and opposite-face relationships tangible, so the two-apart rule becomes something you have seen happen rather than something you are taking on trust.

3. 3

   Then drill rotation with an anchor. Work untimed and say which feature you are tracking before each answer. The aim is to make anchoring automatic rather than a thing you remember to do.

4. 4

   Practise elimination order. Force yourself to reject options by rule before visualising anything. Rejecting by rule is a deliberate order of operations, not something that arrives by itself once the clock starts.

5. 5

   Add the provider's format. A standalone Saville spatial test, a spatial section inside a 10-minute technical battery and interleaved CCAT items at roughly 18 seconds a question are three different jobs. Rehearse the one your invitation names.

Cognivy uses your assessment date to choose the route rather than asking you to predict a study schedule. When you sit down to practise, you choose the session length that fits that day.

Pace

Getting faster without losing accuracy

Speed comes from doing less visualisation, not faster visualisation. Four habits remove the most work.

• Reject before you rotate. Handedness and adjacency each eliminate options without any rotation being performed at all. Two eliminations usually leave one comparison, which is the only place mental rotation is needed.
• Count along one axis. On block-counting and view-matching items, count layers or columns one direction at a time and write the totals. Judging a whole arrangement at a glance is where miscounts come from.
• Learn the net families. A cube has eleven genuinely different nets. Recognising the family in front of you means you are applying a known result rather than solving from scratch.
• Keep the same orientation convention. Always read features clockwise from the same starting point. A consistent convention is what lets you compare two figures without re-deriving your reference each time.

Keep exploring

Related test types and how they differ

Spatial reasoning borders three test types that use similar images for different purposes.

Direct answers

Spatial reasoning test FAQs