Memory Test

Chimp Test

Numbers flash briefly on screen. Remember their positions, then click them in order. Based on the research that revealed chimpanzees outperform humans at this task.

Numbers 4
Streak 0
Ready

Chimp Test

Numbers appear briefly. Memorize their positions, then click them in order 1 → N after they vanish.

What the Chimp Test measures

The Chimp Test directly measures visuospatial working memory - Your brain's ability to hold and recall the precise spatial positions of objects you have seen briefly. Unlike digit span (which uses verbal rehearsal), this task relies on the visuospatial sketchpad component of working memory - The same system probed by visual memory and the ordered-recall sequence memory test. For context on capacity limits, see how many items working memory can hold.

Visuospatial memory

Encoding and retaining the precise grid positions of each number during the flash window - The spatial-span ability also tested by Corsi Block

Sequential processing

Maintaining a strict ordered recall (1→N) while spatial locations are no longer visible

Processing speed

Rapid encoding of multiple spatial positions within the brief display window

Baddeley's Working Memory Model - Which component this tests

Central executive
Attention control - Coordinates the other components
Visuospatial sketchpad
Primary component tested here Holds visual and spatial information
Phonological loop
Verbal rehearsal - Used in digit span tests, NOT this one
Episodic buffer
Integrates information across memory systems

Chimp Test Score Distribution

The distribution of maximum numbers recalled before first failure. Most users fail at 4–6. Reaching 8 puts you in the top 5%. Chimpanzees trained by Matsuzawa reliably reach 9+.

Max numbers before failure - Distribution

35% 26% 17% 8% 0% Chimp avg Human avg 2 3 4 5 ★ 6 7 8 9 10 11+ human avg
Max numbers Percentile Classification % of users
2–3 Bottom 5% Very low 5%
4 25th Below average 20%
5 - Human avg 50th Average 28%
6 72nd Above average 22%
7 87th Strong 13%
8 95th Elite 7%
9+ - Chimp avg Top 2% Chimpanzee territory <2%

"Max numbers" = highest level reached before first incorrect click in a single session.

Humans vs. chimpanzees: the Matsuzawa studies

In 2007, Tetsuro Matsuzawa and colleagues at the Primate Research Institute of Kyoto University published landmark research showing that young chimpanzees outperform adult humans on this exact task. The study was a seismic moment in comparative cognition research.

The Matsuzawa study (2007)

Ayumu, a five-year-old chimpanzee, and two other young chimps competed against adult humans using an identical touchscreen task. Numbers 1–9 appeared briefly (0.2 seconds) on screen, then were masked. The chimps recalled positions with accuracy levels that humans could not match. The research suggested that young chimpanzees possess a form of photographic working memory that humans lose during language development. Published in Current Biology, Vol. 17, Issue 12.

Metric Chimpanzee (Ayumu) Average human Top human (trained)
Max numbers recalled 9+ 4–5 8–9
Accuracy at 0.2s flash ~80% ~40% ~65%
Performance improvement over training Rapid, steep Moderate Slow after plateau
Encoding strategy Photographic / iconic memory Spatial mapping + verbal labels Hybrid strategies
Effect of language on performance Not impacted Verbal interference degrades performance Varies by strategy

The evolutionary trade-off hypothesis

Matsuzawa proposed that humans traded photographic short-term memory for language. As the brain evolved language capabilities, the phonological loop (verbal working memory) expanded and took precedence over the visuospatial sketchpad. Chimpanzees, lacking language, retained the full visuospatial encoding capacity.

This remains debated. Alternative hypotheses suggest the chimp advantage is due to superior attention focus, greater motivation from intensive training protocols (3,000+ sessions), or differences in iconic memory duration. Regardless, the performance gap is real and consistent across multiple replications.

Age effects on visuospatial memory

Visuospatial working memory peaks in the early 20s and declines more steeply with age than verbal working memory. Children under 12 show particularly strong performance on rapid visuospatial encoding tasks.

Average max numbers by age group

8 7 6 5 4 8–11 12–17 18–24 25–34 35–49 50–64 65+
Age group Avg max numbers Top 10% Notable characteristic
8–11 6.2 9+ High iconic memory capacity, fewer verbal interference effects
12–17 6.8 9+ Strong performance, gaming experience often helps
18–24 ← Peak 7.1 10+ Best combination of speed + capacity
25–34 6.5 9 Minor decline, strategy use increases
35–49 5.8 8 Noticeable decline in flash encoding speed
50–64 5.1 7 Visuospatial WM declines faster than verbal WM at this age
65+ 4.3 6 Significant encoding speed reduction

Age trends segmented by self-reported age; visuospatial working memory norms from published developmental studies.

How to improve your score

Unlike verbal memory, visuospatial memory is difficult to dramatically improve through short-term strategies. Before investing in commercial apps, read our look at the brain training myth and transfer effects. However, several approaches yield consistent gains:

1

Don't sub-vocalize - Go spatial

Most people naturally try to remember numbers verbally ("three is top-left, one is center"). This degrades performance by routing through the slower phonological loop. Instead, mentally "paint" the entire pattern as a single spatial snapshot - Think of the grid as a picture, not a list.

2

Scan in a fixed order during flash

Develop a consistent scan pattern during the flash window. Many top performers scan left-to-right top-to-bottom, building a mental snapshot in a predictable sequence. Consistency matters more than the specific pattern you choose.

3

Practice action video games

Action games that require rapid tracking of multiple objects (RTS games, MOBAs) have been shown to improve visuospatial working memory more than any direct training approach. The transfer effect is stronger for this type of memory than for verbal memory.

4

Sleep before testing

Visuospatial working memory is particularly sensitive to sleep deprivation. Research shows a larger impairment from sleep loss for spatial tasks than for verbal tasks. Aim to test on ≥7 hours of sleep for your personal best.

Track your chimp test progress

Create a free account to save your scores and see how your visuospatial memory changes over time.

Create free account All tests

Frequently Asked Questions - Chimp Test

What exactly did Ayumu the chimpanzee do in the original study?
In Inoue & Matsuzawa (2007, Current Biology), Ayumu and two other chimps were trained to tap numerals 1–9 in ascending order on a touchscreen. After the numbers were masked, Ayumu's accuracy at 210ms exposure exceeded that of adult humans. This was not about recognising numbers - It was about spatial mapping under extreme time pressure, the exact skill tested here.
Is this test harder than Sequence Memory?
They require different skills. The Chimp Test demands rapid spatial mapping of multiple simultaneous targets under time pressure. Sequence Memory requires encoding and reproducing an ordered sequence without the extreme time pressure. Most people find the Chimp Test harder at higher levels because the flash duration shrinks.
Does the Chimp Test measure the same thing as IQ?
No. The Chimp Test measures a specific visuospatial short-term memory skill under time pressure - Not general intelligence. It correlates weakly with Raven's Matrices (r≈0.25–0.35) but is a poor proxy for IQ. It is more closely related to Visual Memory and Corsi Block.
Can I train myself to Ayumu's level?
With practice you can improve substantially, but matching Ayumu's 9-number accuracy at 210ms is beyond what most adult humans achieve in controlled studies. Ayumu was also trained from infancy - Likely a critical period for this specific spatial encoding system. See the Chimp Test leaderboard for where top human performers land.