Squares light up one after another on a grid. Repeat them in the same order. Each time you succeed the sequence gets one step longer, and the test ends when you miss twice.
It measures visual-spatial working memory: how much position-and-order information you can hold for a few seconds while nothing else is on screen to help you.
The task belongs to a family used in research since the 1970s, where positions are shown one by one and then reproduced in sequence. Our version keeps the idea — a grid, a growing sequence, order required — but it is a browser adaptation, not the laboratory instrument.
If any order were accepted, the test would measure recognition of places and little else. Requiring the sequence brings in the ordering part of working memory, which is the part that matters when you hold a phone number, a route, or the steps of something you are about to do.
Each square lights for about half a second, with a short gap between them. The timing is deliberate and sits between two failure modes:
A growing sequence always ends in failure; that is how a span test works. One slip should not end it, so you get two attempts before it stops. That filters a moment of inattention without dragging the test out.
Your score is the longest sequence you reproduced correctly. It is a count of positions, nothing more — not a grade, not a percentile, and certainly not an intelligence score.
Sleep, noise, how long you have been staring at a screen, and whether you are being interrupted. Spans of this kind are small numbers, so one step up or down between two sittings is ordinary variation rather than progress or decline.
Your span on this task — partly through genuine memory, partly through strategy, since most people start grouping positions into shapes after a while. Whether a longer span here means a better memory in daily life is exactly the claim that brain-training research has struggled to support, so we do not make it.
It is free and runs in your mobile or desktop browser with nothing to install and no sign-up. Signing in is optional, and it only saves your results so you can follow them from any device.