The Gridrush wiki
Schulte table: the rules, the method and what it really trains
A Schulte table is a grid of shuffled numbers that you find in order as fast as you can, in principle with your gaze resting on the centre. The classic version has 5 × 5 squares holding the numbers 1 to 25; it was first used to study attention in psychology.
On Neurostim, Gridrush turns it into a race: three sizes, a daily grid shared by everyone and a clock. Below you will find the complete rules, measurements on 2,000 of the game's grids, search methods compared with a reading model, and what studies can actually claim about speed reading and peripheral vision.
In short
- Genre: visual search against the clock, solo or in a race.
- Goal: touch every number in ascending order, starting from 1.
- Sizes: 4 × 4 (1 to 16), 5 × 5 (1 to 25), the classic size and the one offered by default, and 6 × 6 (1 to 36).
- Daily grid: one per size, identical for everyone, renewed at midnight, Paris time. Your ranked time is the one from your first try.
- Free run: a new grid every game, with your best time of the day on the leaderboard.
- Penalty: one second per number touched out of order.
- Limits: per size and per day, ten games on the daily grid and 200 in the free run.
- Account: not necessary. Starting a game is enough, and your time goes on the leaderboard under a guest name.
- Devices: touch, mouse, or keyboard with the arrow keys.
The rules of the Schulte table
The grid
A Schulte table holds each number once, from 1 up to the number of squares, placed at random. Here is a 5 × 5 example. The marked square is the centre, where the classic method asks you to rest your gaze. The 1 is just to its right, the 2 two rows above it, the 3 down on the left, the 4 diagonally up and to the left: no order shows.
The goal and the clock
Touch the 1, then the 2, and so on to the last. The clock starts when the grid appears and stops on the last number.
Mistakes
Any number touched out of order counts as a mistake and adds one second to your score. That includes a number you have already found, touched a second time. A mistake blocks nothing: the expected number stays the same, and you carry on. The final score is your playing time plus those seconds; on a tie, the player with fewer mistakes ranks higher.
What stays on the screen
Numbers you have found neither disappear nor stay marked: a brief flash confirms each right touch, then the square goes back to normal. So the grid never thins out during a game, and you look for the 25 among as many squares as for the 2. That is what sets a real Schulte table apart from a game where squares clear away.
Finding the numbers, from beginner to advanced
Reading row by row
This is the beginner's instinct: scan the grid from the top left corner until you land on the right number. It works, and its cost is easy to work out: on 5 × 5, reading from the corner comes to 1 + 2 + ... + 25 squares, or 325 reads, whatever the grid. Every method below tries to bring that figure down.
Resting your gaze on the centre
The classic Schulte method is to fix the centre and search out of the corner of your eye, without moving your eyes. From the centre of a 5 × 5, a square is on average 1.87 squares from your gaze, 2.83 at the furthest. Start with a 4 × 4, where the average falls to 1.50 squares, then widen out.
Stay realistic: what you see in the periphery is blurred, as the summary of Rayner and colleagues' review on reading points out. In practice, a centred gaze mainly helps you locate an area quickly; on a 6 × 6 on a large screen, small eye movements are still normal, and the clock does not ask you to eliminate them.
Keeping what you see along the way
While looking for the 7, you may have spotted the 8 and the 9: if you remember them, you no longer need to search for them. In the model set out below, which measures no player, keeping the last five squares read on 5 × 5 removes a third of the reading. Practise noting the next two or three numbers while you look for the current one.
Thinking in areas
Keeping an exact position is hard, keeping a corner less so. Split the grid into four in your head: on this 4 × 4, knowing the 2 is in the bottom left narrows the search to four squares instead of sixteen.
Preparing the next move
Try to send your eyes to the next number before your hand has finished touching the previous one. Do not wait for the confirmation flash to start searching again: it tells you nothing new if you aimed correctly.
Getting faster
- Count what a mistake costs. One second lost on 5 × 5 equals 40 ms more on each of the 25 numbers. Touching on a hunch almost always costs more than checking with a glance.
- Watch out for numbers that look alike. 12 and 21, or 16 and 19, are easily confused in the periphery. Once you have spotted a number, look at it for as long as it takes to touch it.
- Move up a size once your time settles. Searching does not grow with the number of squares: in our model, going from 4 × 4 to 6 × 6 multiplies the numbers by 2.25, but the reading by nearly 5.
- Measure your set. HandWiki describes an efficiency, the mean time over several tables, and a warm-up, the first time divided by that mean; a value of 1 or less indicates a good warm-up. Five free runs in a row are enough to try it, for fun and not as a diagnosis.
- Keep the keyboard for comfort. On the keyboard, each number takes several arrow presses: it makes the game accessible, but a finger or a mouse will be faster.
Benchmarks: what 2,000 grids say
Distances and the path of your hand
We drew 2,000 grids of each size with Gridrush's generator, in November 2026, and measured the path from each number to the next.
| Size | Average distance between two numbers | The next one touches the last | Same row or column | Total path, mid-series |
|---|---|---|---|---|
| 4 × 4 | 2.15 squares | 34.8%, or 5.2 times per grid | 39.6% | 32 squares |
| 5 × 5 | 2.65 squares | 24.1%, or 5.8 times per grid | 33.5% | 64 squares |
| 6 × 6 | 3.17 squares | 17.5%, or 6.1 times per grid | 28.6% | 111 squares |
The middle column says something useful: these shares are those of pure chance, give or take sampling noise. On 5 × 5, a square has on average 5.76 neighbours out of the 24 other squares, which is 24%. So the shuffle hides no pattern to learn: the next number is no likelier to be beside the last than anywhere else. Looking around the last number first is still handy, because that is where your eyes already are, but it is no statistical shortcut.
Since the grid keeps roughly the same width on screen, your hand's journey comes to about 8 times the grid's width on 4 × 4, 13 times on 5 × 5 and 18.5 times on 6 × 6.
How many squares you read
This table comes from a model, not from real players. On 500 grids of each size, we simulated a simple reader: it starts again from the square of the last number found, reads row after row, and remembers without error a certain number of the last squares it read. A remembered number is touched without reading. Since it rereads its starting square on each search, it counts 339 reads on 5 × 5, not the 325 worked out from the corner.
| Size | No memory | 5 squares kept | 10 squares kept | Whole grid kept |
|---|---|---|---|---|
| 4 × 4 | 143 reads | 75 | 34 | 16 |
| 5 × 5 | 339 reads | 222 | 135 | 25 |
| 6 × 6 | 683 reads | 519 | 374 | 36 |
On 5 × 5, keeping 5 squares removes 35% of the reading, and keeping 10 removes 60%. With 10 squares kept, nearly 9 numbers per grid get touched without searching. A real player sees several squares at once, forgets and does not read in this order: these figures say nothing about seconds saved. They only show that what you keep in mind can take a lot of searching away.
What about time?
We give no average time for a whole grid: neither the reference pages on the Schulte table consulted here nor the study by Lu and colleagues provide one. That study gives a partial point of comparison: among 27 pupils aged 8 to 11, finding 1, 2 then 3 in a black table took 4.7 seconds on average, and 5.2 seconds when red numbers were mixed in. To see where you stand on a complete grid, the simplest way is still the leaderboard for your size, on the same grid as everyone else.
The history and effects of the Schulte table
Its origin
Russian Wikipedia and HandWiki attribute the table to the German psychiatrist Walter Schulte (1910-1972), professor in Tübingen from 1960 and later director of the university's neurological clinic. Take that attribution with care: his German Wikipedia entry does not mention these tables, the study by Lu and colleagues does not name him, and no source consulted here gives a creation date.
Its use in psychology
The table was first a psychodiagnostic task for studying attention. The method described on Russian Wikipedia records the time per table and the numbers missed, over five tables in a row, to assess work capacity. There are variants, such as the Schulte-Platonov table, with Arabic and Roman numerals, and the two-colour Schulte-Gorbov table. Another close test, the Trail Making Test, was in use by 1944: its Part A asks you to connect the numbers 1 to 25 in order.
What studies say
- Speed reading: the review by Rayner and colleagues concluded in 2016 that promises of very fast reading with good comprehension are overstated, and that only modest gains come from ordinary practice. Its summary points out that information seen in the periphery stays blurred.
- Attention: Lu and colleagues note in 2022 that the grid is widely used for attention training, but their own study measures visual search and tests no training.
- Transfer: the review by Simons and colleagues found in 2016 that brain training mainly improves the trained task, with very little evidence of an effect on everyday life.
The research done for this page found no controlled trial showing that a Schulte table alone makes people read faster. You will probably get faster at the table itself; beyond that, nothing is proven. If you enjoy this kind of effort, try Counter too, which reshuffles a grid of 100 numbers every round, Flashtap for pure reaction and Hueclash for selective attention.
Frequently asked questions
What is a Schulte table?
A Schulte table is a grid of randomly placed numbers that you find in ascending order as fast as possible. The classic size is 5 × 5, with the numbers 1 to 25. It is attributed to the German psychiatrist Walter Schulte, and it serves as an exercise and a task for visual attention.
How do you play a Schulte table?
Rest your gaze on the centre of the grid, spot the 1, touch it, then look for the 2 and carry on to the last number, without saying them aloud. On Gridrush, the clock starts when the grid appears and each number touched out of order adds one second to your time.
What is a Schulte table good for?
It was originally used to study attention in psychology, and today it is mostly used as a visual search and concentration exercise, for adults and children alike. Playing makes you faster at the table itself, but benefits beyond the exercise have not been demonstrated.
Do Schulte tables work?
They work as a way to get better at Schulte tables, and that is about all that is established. The review by Simons and colleagues finds that brain training tends to improve the trained task, with little evidence for other tasks. The research for this page found no controlled trial of Schulte tables on reading or on daily life.
Does a Schulte table increase IQ?
Nothing supports that, and Neurostim does not promise it. The research done for this page found no study showing it. More broadly, the review by Simons and colleagues finds very little evidence that brain training improves tasks distant from the one you practised.
Does a Schulte table help with speed reading?
As far as we know, no solid study shows that it does, even though the exercise is often sold for that. The review by Rayner and colleagues calls the promises of speed reading overstated: reading faster usually comes at a cost in comprehension, and only modest gains come from ordinary reading practice.
Is a Schulte table good for ADHD?
The research done for this page found no solid evidence that it helps, and Gridrush is a game, not a treatment. It may be a pleasant concentration exercise, but anyone looking for help with attention difficulties should talk to a health professional rather than rely on a timed grid.
What is a good Schulte table time?
The sources consulted give no reliable norm for a whole grid, including the often requested 5 × 5 average, and your time also depends on your screen size and how you touch. Compare yourself with yourself, size by size, and with other players on the daily grid's leaderboard, since everyone plays the same one.
What is the Schulte table world record?
As far as we know, no recognised organisation keeps a Schulte table world record, and the times claimed on various websites cannot be compared: the size, the device and the grid all change. On Gridrush, the daily grid is the same for everyone, which makes the times on its leaderboard comparable.
How do you solve a Schulte table faster?
Keep track of the next numbers you come across while you look for the current one: in our model, that is what cuts the reading. Think in areas rather than exact squares, avoid the mistakes that cost a second, and move up a size once your time has settled.
Is Gridrush free, with no account?
Yes. Gridrush is free and starts with no sign-up: one click on "Start the clock" has you playing as a guest, with a time that goes on the leaderboard. You can log in if you already have an account, or create one later to keep your results.
Sources
- Keith Rayner et al., So much to read, so little time: How do we read, and can speed reading help?, Psychological Science in the Public Interest 17 (1), 2016.
- Aitao Lu et al., Attention mechanisms underlying dual-color digital visual search based on Schulte grid, Brain and Behavior 12, 2022.
- Daniel J. Simons et al., Do "Brain-Training" Programs Work?, Psychological Science in the Public Interest 17 (3), 2016.
- Wikipedia (Russian), Таблицы Шульте, and HandWiki, Schulte table, checked on 7 November 2026.
- Wikipedia (German), Walter Schulte, checked on 7 November 2026.
- Wikipedia, Trail Making Test, checked on 7 November 2026.
Your turn
Today's Gridrush challenge is the same for everyone until midnight, Paris time. No account needed: you get a guest name with your first game.
Games of the same kind
This guide also exists as plain text, for AI assistants: markdown version. All the guides of the wiki.