How the test works

A 3 by 3 grid lights up one square at a time, for half a second each with a short pause. When it stops, tap the squares in the same order. You start at level 3, three squares, and each level you pass adds one. A mistake costs one of your two lives and replays the level with a new sequence.

Your score is the last level you completed. In the app, a morning game stops after 30 seconds to fit the routine. On this page there is no time limit, so your score can be compared with the public norm, which has none either.

What is a good level?

Across 35,868 Human Benchmark games, the median is level 9 (Human Benchmark, 2026). The 10th percentile is level 5, the 75th is level 12 and the 90th is level 15. The public version keeps the same sequence and adds one square per level, while this one draws a new sequence each time, which is slightly harder to chunk.

Human Benchmark, 35,868 games, score distribution read on 13 October 2026.
Percentile	Score	Reading
P10	5	at least 10% of games end at this score or lower
P25	7	at least 25% of games end at this score or lower
P50	9	at least 50% of games end at this score or lower
P75	12	at least 75% of games end at this score or lower
P90	15	at least 90% of games end at this score or lower
P99	25	at least 99% of games end at this score or lower

The scientific parent, the Corsi block task

Sequence memory is a digital version of the Corsi block tapping task, where an examiner taps wooden blocks in an order you must repeat. It measures visuospatial short-term memory, and it has published standard procedures and norms (Kessels et al., 2000).

Computerized, it is reliable. In young adults retested 20 minutes apart, the Corsi block test reached a reliability of 0.85 (Morawietz et al., 2024). That is a good sign for this game, but the app still treats its own version as provisional until it measures it.

Why you get better fast at first

Memory span tests are sensitive to practice. Across 68 studies and 234 effect sizes, scores on working memory capacity tests rose between the first and the second session and kept rising up to the fourth (Scharfen et al., 2018b). Part of the gain is learning the test, finding a rhythm, grouping squares into shapes.

So do not judge your memory on your first game. Play three or four, then look at where you settle.

Questions

What is the average on the sequence memory test?

Level 9 is the median across 35,868 Human Benchmark games (Human Benchmark, 2026). Beyond level 12, you are past the 75th percentile.

Is sequence memory the same as working memory?

It is one part of it, the spatial part, holding an order of positions. Our guide on working memory explains the other parts and how they are measured.

What this test does not tell you

It measures how long a sequence of positions you can repeat on this screen, today. It does not measure intelligence, and it cannot detect a memory disorder or dementia.

Fatigue, noise and practice change the score a lot. If you are worried about your memory, talk to a doctor rather than a game.

Sources

Human Benchmark, score distributions embedded in humanbenchmark.com, read on 13 October 2026. Reaction time n 70 940, sequence memory n 35 868, chimp test n 27 233, number memory n 24 084. humanbenchmark.com
Kessels R.P.C., van Zandvoort M.J.E., Postma A. et al. (2000). The Corsi Block-Tapping Task, standardization and normative data. Applied Neuropsychology 7(4), 252 to 258. doi.org/10.1207/s15324826an0704_8
Morawietz C. et al. (2024). Consistency of spatial ability performance in children, adolescents, and young adults. Frontiers in Psychology 15, 1365941. PMC10938598. doi.org/10.3389/fpsyg.2024.1365941
Scharfen J., Jansen K., Holling H. (2018). Retest effects in working memory capacity tests, a meta-analysis. Psychonomic Bulletin and Review 25, 2175 to 2199. doi.org/10.3758/s13423-018-1461-6

Written by the Headgame team

Updated 16 October 2026. Scientific reviewer not yet appointed. Every figure links to its source.

Sequence memory test