Visual Memory Test: Find Your Spatial Pattern Span in a Free Timed Task

How the score is calculated

The task opens with a 4 by 4 grid. Three cells light up for a second and a half, then the grid goes dark and you tap or click the cells you remember. Order does not matter, only which cells were lit, and the screen counts your choices as you go so you always know how many are left. Once you have chosen as many cells as were lit, the task checks your answer immediately. Every cell is a button labeled with its row and column, so the whole task can be done from a keyboard rather than a mouse or a touch screen.

Recall a pattern correctly and the next one adds a cell. The grid widens from 4 by 4 to 5 by 5 once the pattern reaches 6 cells, which keeps the patterns from crowding together as they grow. Miss a pattern and you get one more try at the same length with a fresh arrangement of cells. Two misses in a row end the task, and so does a correct 9 cell pattern, which is the largest pattern this task will ever show. A correct recall resets the miss count, so a single unlucky pattern does not end your attempt.

Your span, the headline number, is the largest pattern you recalled completely and correctly. Alongside it we show how many patterns you got right and how many you were shown in total, so you can see whether your span came from a clean run or from a scrappier one with a miss along the way. If you leave the browser tab while a pattern is on screen, that pattern is cancelled rather than scored, because a pattern you did not see is not a memory failure.

Is photographic memory real?

A photographic memory, in the way films and novels use the phrase, would mean holding a page or a scene in mind as a perfect mental snapshot and reading details back off it later. That ability is not supported by the research record. Careful attempts to find adults who can do it have not produced convincing cases, and the memory literature instead describes recall as reconstructive: we keep the gist, some vivid details, and a fair amount of inference, rather than a stored image we can re-inspect.

Eidetic imagery is the closest documented relative, and it is a narrower thing. It has been reported mainly in children, and it is uncommon even among them. Reviewing two decades of work on it, Haber estimated that eidetic ability appears in roughly 2 to 15 percent of American children under twelve, and that it fades as children get older, with the ability essentially absent in adults. Even in children who show it, the image is imperfect, it drifts, and it decays within minutes. It is a lingering, unusually vivid visual trace, not a page you can go back and read.

The one adult case studied in real depth is the woman known as Elizabeth, described by Stromeyer and Psotka in Nature in 1970. She was reported to have viewed one half of a random dot stereogram, then fused it from memory with the other half shown later, which would require storing an enormous amount of visual detail with near perfect fidelity. The result drew methodological criticism at the time, Elizabeth was never retested, and no one else has come close to reproducing the performance in the decades since. A single unreplicated case is not evidence that the ability is impossible, but it is nowhere near enough to support the popular claim. So if you came here searching for a photographic memory test, the honest answer is that this task, and every other online quiz, measures span and accuracy rather than anything like a mental photograph.

Visual span and digit span are different measurements

The forward digit span at /tests/working-memory is a verbal task: digits arrive one at a time and you type them back in the same order, so both the items and their sequence have to be held. This task is spatial and simultaneous: the whole pattern appears at once, and only the set of locations has to be held, not the order they came in. Two people can land in different places on the two tasks, and that difference is interesting rather than contradictory.

Research has long treated verbal and visuospatial short term memory as partly separate systems, which is why neuropsychologists use a spatial span task alongside a digit span rather than treating one as a stand-in for the other. Luck and Vogel found that visual working memory holds roughly four integrated objects at once, a small number that seems to reflect objects rather than raw features. Spatial patterns on a grid can be grouped into shapes, lines, and corners, which is one reason most people manage more than four cells here: a five cell pattern that happens to look like a letter is much easier to hold than five scattered cells.

The best known spatial span task in clinical research is the Corsi block-tapping task, where the examiner taps a sequence of blocks and the person repeats it. Kessels and colleagues published standardized administration and normative data for it, with forward spans for healthy adults typically falling between five and seven blocks. Our task deliberately drops the sequence requirement, so it is easier in one respect and harder in another, and those published Corsi numbers do not transfer to it. Treat your span here as a number to compare against your own later attempts, not against a clinical norm.

What can move your number, and when to look further

Sleep, stress, distraction, and how much of the grid you can comfortably see all shift results on a task like this. A phone held close works, but a larger screen makes the 5 by 5 grid easier to take in at a glance. Strategy matters too: people who mentally join the lit cells into a shape, or who name rows and columns to themselves, usually score higher than people trying to hold each cell as a separate fact. That is a real skill, and it is also a reminder that a span score reflects how you approached the task as much as raw capacity.

Practice raises scores on span tasks, so a second or third attempt is usually better than a first. That improvement is mostly familiarity and strategy rather than a change in memory itself, which is why a single number is more useful as a starting point than as a verdict. If you retake it, keep the conditions similar, same device and a similarly quiet moment, so you are comparing like with like.

A short span on one browser attempt is not a sign of a memory disorder, and this task does not identify one. If what brought you here is a real worry about memory in daily life, forgetting recent conversations, repeating questions, losing track of familiar routes, or someone close to you raising it, the more useful page is /tests/memory-screen, which asks about those everyday changes rather than timing a laboratory style task. Persistent changes of that kind deserve a conversation with a doctor, who can look at sleep, mood, medication, thyroid function, and much else that this task cannot see.

Methodology & sources

This is an original timed task built for Psychology.com, not a reproduction of a published clinical instrument. It sits in the family of spatial span measures that runs from Corsi's block-tapping task through the visual pattern tasks used in working memory research, and it borrows their core logic: show a spatial arrangement briefly, take it away, ask for it back, and grow the load until recall breaks down. The adaptive structure, a start at 3 cells, one cell added after each success, a second attempt allowed at a failed length, and a stop after two consecutive misses, follows the standard staircase design used in span testing.

Where it departs from the Corsi task matters, and we would rather say so than imply more precision than the task has. The Corsi task is sequential, so it measures memory for an order of locations; ours is simultaneous, so it measures memory for a set of locations. That makes the published Corsi normative data, such as the standardization by Kessels and colleagues that puts adult forward spans in the range of five to seven blocks, useful background but not a scoring key for this page. The 1.5 second exposure, the grid sizes, and the 9 cell ceiling are our design choices, made to keep the task short and readable on a phone, not values inherited from a validated protocol.

The result bands on this page are educational, not clinical cutoffs. There are no published norms for this exact task, so the ranges were set to reflect the spread we would expect given what visual working memory research describes, including Luck and Vogel's finding that visual working memory holds about four integrated objects at a time, and the well documented fact that grid patterns can be chunked into shapes, which lifts performance above a raw object count. On eidetic and so-called photographic memory we have stayed with what the literature supports: Haber's review of the evidence in children, and the single, never replicated adult case reported by Stromeyer and Psotka. Nothing on this page should be read as detecting, ruling out, or measuring a photographic memory.

Reviewed by Michael Callans, MSW
Reviewed to keep the spatial pattern span task framed as one narrow visual memory measure, and to keep the photographic memory section faithful to what the research actually shows. Last reviewed September 23, 2026.

References

1. Kessels, R. P. C., van Zandvoort, M. J. E., Postma, A., Kappelle, L. J., and de Haan, E. H. F. (2000). The Corsi Block-Tapping Task: Standardization and normative data. Applied Neuropsychology, 7(4), 252 to 258. doi.org/10.1207/S15324826AN0704_8
2. Berch, D. B., Krikorian, R., and Huha, E. M. (1998). The Corsi block-tapping task: Methodological and theoretical considerations. Brain and Cognition, 38(3), 317 to 338. doi.org/10.1006/brcg.1998.1039
3. Luck, S. J., and Vogel, E. K. (1997). The capacity of visual working memory for features and conjunctions. Nature, 390, 279 to 281. doi.org/10.1038/36846
4. Stromeyer, C. F., and Psotka, J. (1970). The detailed texture of eidetic images. Nature, 225, 346 to 349. doi.org/10.1038/225346a0
5. Haber, R. N. (1979). Twenty years of haunting eidetic imagery: Where's the ghost? Behavioral and Brain Sciences, 2(4), 583 to 594. doi.org/10.1017/S0140525X00064542

Visual Memory Test FAQ