Spatial visualization

Paper folding test that shows you the unfold

A square sheet is folded one, two or three times, a hole is punched through every layer at once, and you choose which of five unfolded sheets that produces. Sixteen items, 45 seconds each, and a corrected score that subtracts a quarter of a mark for every wrong answer because five options mean a fifth of a mark is available by luck. Each practice item ends with the sheet opening out fold by fold, so the operation you are being asked for is demonstrated rather than described.

100% free
No signup
16 items, five options
Up to three folds
Animated unfold on practice
Guessing correction shown

A square of paper is folded once, twice or three times, a single hole is punched through the whole stack, and the paper is opened out again. Your job is to pick the sheet that comes back. 16 items, five choices each, 45 seconds apiece.

A worked item

One punch through 3 thicknesses leaves 4 holes, and each of them sits where a crease reflects the one before it. The creases are drawn on the opened sheet here; in the test they are not.

What an item is made of

Sheet: a square with sixteen possible hole positions, four by four
Folds: one, two or three, each along the midline of what is left of the sheet
Punch: a single hole, through every thickness, so the answer has 2, 4 or 8 holes
Choices: five per item, lettered A to E, all with the same number of holes
Window: 45 seconds per item; an unanswered item costs nothing
Score: right minus wrong over 4, the standard correction for five choices

In the practice items the paper opens out in front of you after each answer, one crease at a time, with the new holes ringed. In the scored items it does not — watching the answer would teach the trick you are being measured on.

How to take the paper folding test

Read the folds left to right, find the punch, then open it back up in your head.

Read the strip of folds as a sequence, not a picture
Each item is a row of small diagrams: the flat sheet, then the sheet after each fold, then the last panel with a hole punched through the folded stack. The crease shown in one panel is the crease that has already happened by the next. Nothing is hidden and nothing is drawn to scale — the whole item is the fold order and the punch position, and everything else on the strip is there to make those two legible.
Unfold in reverse, mirroring the holes across each crease
Undoing a fold reflects every existing hole to the other side of that crease and keeps the originals. Work backwards through the folds doing exactly that and the answer falls out — two folds double twice, so one punch becomes four holes. This is worth doing deliberately on the three practice items, where the page then animates the same unfold and you can check your reconstruction against it stage by stage.
Answer with A to E, then read the replay afterwards
Five keys or five buttons, 45 seconds an item, and no feedback at all once the scored items begin — being shown the answer mid-run would teach you during the measurement, which is why the demonstration lives in practice instead. When the run ends, every scored item comes back as a grid with your choice, the correct sheet and the unfold available on each, so nothing is withheld in the end, only postponed.

Technical specifications

Items	16 scored, in a fixed difficulty progression from one fold to three, plus three practice items — one at each fold count — with the unfold animated afterwards
The sheet	A 4 by 4 lattice of possible hole positions, folded along a midline each time so the visible region halves. A one-fold item returns two holes, a two-fold item four, a three-fold item eight
Options	Five per item, every one carrying the same number of holes as the correct sheet. Counting the dots therefore eliminates nothing, which is the single most common way a folding item is accidentally made easy
The four wrong options	Each is the sheet somebody arrives at by one specific mistake: mirroring across a crease that was not there, turning the whole sheet, or getting the pattern right and its position wrong. The result names which mistake you made most often
Response	Keys A to E read by physical position, and five on-screen buttons. 45 seconds per item — long enough for three folds worked carefully, short enough that the run does not become an exercise in stamina
Scoring	Items right, and items right minus a quarter of the items wrong. 20% is what random choosing returns in the long run — One correct option in five, chosen without information — The reciprocal of the number of options, which is why the guessing correction divides wrong answers by four. — so the correction removes the marks luck supplies and can push a score below zero if most answers were guesses
Motion	The unfold animation steps every 750 ms and stops itself when the tab is hidden. A system-level reduced-motion preference replaces it with the same stages as still frames you step through yourself
Reference figure	None is printed. The published version of this task is a printed booklet whose norms are tied to its own items and its own two-part timing, and neither transfers to sixteen items generated in a browser, so this page reports counts and times and stops there

Frequently asked questions

What ability does paper folding actually measure?

Spatial visualization — holding a shape in mind while it changes, as opposed to holding it rigid while it turns. That distinction is the whole reason this task exists separately: folding is a sequence of transformations that each depend on the last, so it loads working memory in a way a single rotation does not. Factor studies routinely place visualization and rotation on different factors, and the practical version of that is that people who find these items easy are not automatically the people who are fast at block figures.

Why does the page not show the answer after each scored item?

Because feedback during a measurement changes what is being measured. The unfold animation is genuinely instructive, and that is exactly the problem: shown sixteen times, it would teach you the task while the task was being scored, and item 14 would be measuring something different from item 2. So the demonstration runs on every practice item, where teaching is the point, and the scored run stays silent until the end — at which point every item is available to replay, unfold and all.

Why subtract a quarter of a mark for a wrong answer?

Because five options make a fifth of the paper available without any folding at all, and the correction takes it back. Marking one wrong answer down by a quarter of a mark means that answering every item at random earns, on average, zero — which is the honest score for that strategy. The raw count is shown next to the corrected one, because on sixteen items the two can differ by a mark or two purely by luck and hiding one of them would overstate the precision of either.

Are all sixteen items the same difficulty?

No, and the order is deliberate rather than random. The run opens on one-fold items, moves through two folds, and spends most of its second half on three, because a three-fold item asks you to track eight holes that arrived in a particular order and that is a different task from tracking two. A run that shuffled difficulty would spend your first minute on the hardest item on the page, and the number that came out would be partly about where the hard items happened to land.

The animation makes me uncomfortable. Can I turn it off?

It is already off if your system asks for reduced motion — the page reads that preference and replaces the animated unfold with the same stages as still frames you advance yourself, at whatever pace you like. Nothing is lost by taking that route: the stages are identical, and stepping through them by hand is arguably the better way to check your own reconstruction against the page's. The scored items animate nothing in either case.

Can I use this to prepare for a spatial section of an assessment?

Yes, with one caveat about which part transfers. The item form is the standard one and working through it builds the habit that matters — unfolding in reverse rather than trying to picture the final sheet in one go. What does not transfer is the score: a supervised assessment times you differently, uses its own item set and scores against its own group, so a good run here is evidence you understand the operation and evidence of nothing else.

Where does the hole go when a fold lands on top of it?

It goes through everything underneath, which is the point of the punch. The hole is made after all the folding is finished, so it passes through every layer of the stack at that position at once, and unfolding therefore returns one hole per layer. That is why the hole count doubles with each fold rather than staying put, and why an option with the wrong number of holes would be trivially rejectable — which is precisely why no option here has one.

Folding as a paradigm, and what makes an item honest

The punched-hole item has been in use as a marker of spatial visualization for about seventy years, and its best-known form is the paper folding test filed as VZ-2 in the reference kit assembled for exactly this purpose — Ekstrom, French, Harman & Dermen (1976), Manual for Kit of Factor-Referenced Cognitive Tests, Educational Testing Service. That kit exists because psychologists needed items everyone agreed loaded the same factor, and the folding item earned its place by being hard to solve any other way: there is no verbal description of the answer, no feature to match, and nothing to eliminate unless the item was built carelessly. This page borrows the paradigm and not the booklet. The items here are folded in software on a 4 by 4 lattice, which means the answer is computed rather than drawn, and a fresh set arrives every run.

Building the item well is mostly about the four wrong answers. A folding item where the options differ in how many holes they have can be answered by counting, and a visitor who counts is being scored on arithmetic; every option here therefore carries the same hole count as the answer. Beyond that, each distractor is generated by making one real mistake — reflecting the holes across a crease that was never folded, rotating the finished sheet, or producing the right arrangement in the wrong quadrant — so a wrong answer is informative rather than merely wrong. The result panel counts which of the three you picked most, and that count is the closest thing this page offers to advice.

Where folding sits relative to its neighbors is worth knowing before reading much into one score. It is a visualization task, not a rotation task: the mental rotation test keeps a shape rigid and measures the rate at which you turn it, which is a cleanly different operation, and the block design test makes you build a pattern with your hands rather than recognize one with your eyes. For the composite view, the spatial reasoning test runs a folding sub-scale beside two others and says whether they differ. One practical check is worth doing before any of it: these items are small and low in contrast by nature, and if the dots are hard to resolve on your screen, the contrast sensitivity test will tell you whether that is the display, the lighting or your eyes before you conclude anything about visualization.

A reaction time here is the interval between the frame that painted the stimulus and the timestamp the browser attached to your key, both read from the same monotonic clock. What neither can see is the display pipeline behind it, so on a 60 Hz screen roughly 16 ms of every figure below is the machine rather than you. That is the timing floor: two numbers closer together than that are the same number, and this page reports no precision it cannot support.

Almost nothing on this page depends on that floor, because the times it reports are tens of seconds rather than fractions of one. It is stated anyway for the same reason the outlier rule is: a page that prints a duration owes you an account of what the duration includes, even when the answer is that a frame of it hardly matters.

This is a measurement exercise, not a clinical assessment. It reports what you did on this page against a stated reference and nothing more — it cannot establish a visual-processing difficulty or a spatial learning disability. Only a qualified professional, working with more than a browser, can make that judgment.

A low corrected score on sixteen items is compatible with a bad night, a small screen and a task you have never seen before, and there is no way to tell those apart from here.

Where the folding happens

Every number on this page is worked out by JavaScript running in the tab you are reading it in. Your answers, your reaction times and your score are never uploaded, logged or kept — which is also why the test carries on working after you disconnect from the network, and why nothing here can be held back behind an email address.

Every sheet, every crease and every option is constructed by JavaScript running in this tab, from a seed generated when you press start. No item is fetched, so a run works with the network disconnected, and your sixteen answers exist only as an array that the result panel reads and the tab discards when it closes.