Cognitive Tests
A test gives you a number in under a minute. Training changes that number over weeks. These are the tests — short, standardised, free, and no account required. Each one links to the Mind Dojo game that trains the same skill once you want the number to move.
One honest caveat up front: browser measurements include your display, input device and browser lag, so results are not comparable between websites or devices. Compare yourself to yourself, on the same hardware — that trend is real even when the absolute number is inflated.
Take a test now
Tests that are already games
For these constructs the Mind Dojo game itself is the instrument — it implements the classic paradigm directly, so there's no point shipping a thinner second copy. Each one measures you from the first run.
Mental Rotation Test
3D match-or-mirror judgment on polycube pairs.
Multiple Object Tracking Test
How many moving targets you can hold in attention.
Card Sorting Test
Set-shifting — adapting when a hidden rule changes.
Sequence Memory Test
How long an ordered sequence you can reproduce.
Divided Attention Test
Running two continuous tasks at once.
Classic lab tests you can play free
Each of these games is built on a task from cognitive psychology research. They are training games, not clinical instruments: each game page explains where it follows the original and where it departs from it.
| Classic test | What it measures | Play it as | Key paper |
|---|---|---|---|
| Shepard and Metzler mental rotation task | 3D match-or-mirror judgment on polycube pairs | Mirror Axis | Shepard, R. N., & Metzler, J. (1971) |
| multiple object tracking task | How many moving targets you can hold in attention | Constellation Drift | Pylyshyn, Z. W., & Storm, R. W. (1988) |
| Wisconsin Card Sorting Test | Set-shifting — adapting when a hidden rule changes | Kata Shift | Grant, D. A., & Berg, E. A. (1948) |
| Corsi block-tapping task | How long an ordered sequence you can reproduce | Echo Blitz | Kessels, R. P. C., et al. (2000) |
| dual-task paradigm | Running two continuous tasks at once | Gemini Protocol | Pashler, H. (1994) |
Why we don't show percentiles
Most test sites rank you against a distribution they never explain. We don't, because the number a browser measures is inflated by hardware we can't see — a percentile built on that would be invented precision dressed up as science. Instead, results are reported against figures from published research, with the papers cited, and the honest statement that the comparison that matters is you against your own past runs.