Sequence Memory Test
Watch the squares flash, then repeat the sequence. Each success adds one step — how long a sequence can you hold?
What this test measures
Visuospatial short-term memory span: how many sequential spatial positions you can hold and reproduce in order. The grid flashes a sequence starting at 3 squares; every correct reproduction adds one square. You get two attempts at each length — the same two-trial rule as the clinical Corsi task — and the test ends after two misses at the same length. Your score is the longest sequence you completed. This is distinct from verbal memory (remembering words or digits) — it engages the visuospatial sketchpad in Baddeley's working memory model.
The science behind it
This paradigm is a computerized variant of the Corsi block-tapping task, introduced by Philip Corsi in 1972 and now one of neuropsychology's standard memory measures. The standardization study by Kessels et al. 2000 (Applied Neuropsychology; 70 healthy adults) puts the healthy adult mean span at 6.0 blocks — strikingly close to Miller's famous 'seven plus or minus two' for verbal span. Their supervised sample's spread was tighter (SD ≈ 1.3); we score against a broader SD of 2.0 to absorb the extra variance of unsupervised web testing, a conservative choice that pulls extreme percentiles toward the middle. Span declines gradually with age and is sensitive to attention load. Gaming research finds modest working-memory advantages for players: Thompson et al. 2013's meta-analysis (Perspectives on Psychological Science) reports positive associations between video game experience and working memory measures, and Kowal et al. 2018 found esports players outperform casuals on cognitive batteries that include memory tasks.
How to improve your memory span
Raw span is fairly stable, but strategy makes a measurable difference. Chunking — grouping positions into shapes or paths ('L-shape, then diagonal') — is the best-documented technique and is how memory athletes exceed normal spans. Rehearse the sequence during playback rather than after, trace the path with your eyes, and stay under low working-memory load (no music with lyrics, no split attention). Training effects on span tasks are real but narrow: practice improves the trained task substantially while far transfer to general intelligence remains debated in the literature.
Frequently Asked Questions
What is a good sequence memory score?
The healthy adult mean is a span of 6 (Kessels et al. 2000). Reaching 8 puts you around the top 16%, and 10+ is roughly the top 2-3%. A span of 4 lands around the bottom 16% — worth re-testing when rested if you were distracted.
Is this the same as the Corsi block test?
It is a computerized variant of the same paradigm: sequential spatial positions reproduced in order, with span as the score. We follow the clinical administration closely — two trials per length with the length credited if either passes, and roughly one second per item — but clinical Corsi uses 9 physical blocks under supervision, so treat clinical norms as a reference rather than a diagnosis.
Why do I fail at the same length every time?
That plateau is your span. Working memory holds only about 4±1 chunks (Cowan 2001) — but a chunk can contain several squares, which is why typical spans on this test run 5-7 individual positions. To break through, improve your chunking: group positions into shapes or paths so several squares count as one item.
Does sequence memory decline with age?
Yes, gradually. Corsi span declines roughly half a block per decade after middle age in normative studies (Kessels et al. 2000). Staying cognitively and physically active is associated with slower decline.
Verbal memory vs. visual memory — what's the difference?
They use partly separate systems: verbal span (digits, words) engages the phonological loop, while this test engages the visuospatial sketchpad (Baddeley's model). People can be strong in one and average in the other, which is why test batteries measure both.
More Cognitive Tests
Watch cells light up on a 3×3 grid and press Match whenever the position repeats from exactly 2 steps back. 30 fixed-pace trials (28 scored, exactly 9 true matches) measure your working-memory balanced accuracy.
See an original shape, then pick which of 4 options is that shape rotated — the rest are mirror images. 12 scored rounds of increasingly complex polygons, about one minute.
Click when the screen turns green. 12 scored rounds measure your average simple visual reaction time in milliseconds — then a 6-round distractor block tests your impulse control separately.