Schulte Table

The numbers sit in fixed rows and columns. That sounds like a small detail. It is the whole difference between this and the trail making test, where the same 25 numbers are thrown across the page at random.

A grid gives your eye a route. You can sweep a row, drop to the next one, and know you have not missed anything. A scatter gives you nothing to hold on to, so you re-check places you have already looked. Searching a known layout and searching an unknown one are two different jobs, and they come out as two different times.

Anne Treisman's work on visual search is the reason the difference shows up. A number among other numbers shares every feature with its neighbours, so nothing about it pops out. You have to look at items more or less one at a time. Jeremy Wolfe's later work put figures on how much the layout changes that rate. A tidy arrangement is worth a lot.

Every other test here draws your score on a bar built from published results. This one does not, and that is deliberate.

Search for a typical Schulte time and you will be told 37.7 seconds. You will also be told that 8 seconds is a good adult result, and that beginners take 40 to 60. Those are the same task and the same grid. They disagree by a factor of five, and not one of them names a study.

We looked for the study. The nearest thing in the peer-reviewed literature is Lu and colleagues in 2022. They put 27 children aged 8 to 11 in front of a modified two-colour grid and measured brain responses. It reports time per number, not time per table, and 27 children is not a distribution of adult times.

So there is nothing honest to rank you against. Building a curve anyway would mean picking a shape out of the air and printing a caveat under it. That is how the numbers above got into circulation. Instead the page reports your time and your own history. For the argument in full, read what is a good reaction time. A single average is a weak thing to chase even when a real one exists.

A single table is one sample of a noisy process. Three things move your time by several seconds in either direction, and none of them are about you.

• Where the 1 happens to land.
• Whether the 17 sits next to the 18.
• How awake you were for those forty seconds.

The classic procedure runs five tables back to back, and this one defaults to the same. The score is the median of the set, so one unlucky table cannot drag it.

The drift is your last table minus your first. A negative drift means you sped up, which usually means the first table was really you learning where to look. A positive drift means you slowed down across two or three minutes of steady searching. That is a small window onto the thing the task is for.

Treat it gently. It is a difference between two single tables, so it carries all the noise of both. Age matters here as it does everywhere in this kind of measure. See average reaction time by age for how steeply speed moves across a lifetime.

The traditional instruction is to hold your gaze on the middle of the table. Find the numbers with the edge of your vision rather than by looking straight at them. The dashed centre cell is there so you have something to hold onto. Turn it off in the settings if you would rather scan freely.

This is where the claims start, so here they are plainly. Schulte tables are sold as a way to widen your field of view and to make you read faster. Those claims are marketing, not findings. Nobody has shown that tapping numbers in a grid changes how wide your vision is or how fast you get through a page of prose.

What is true is narrower and duller. Practise this task and you get faster at this task. That is worth having if you enjoy the chase, and it is not evidence about your attention anywhere else. A time that worries you is a conversation for a doctor, not for a grid of numbers.