Chimp Test: Can You Beat a Chimpanzee’s Memory?

What is the Chimp Test?

The Chimp Test is a short working-memory game. Numbers appear on a grid, you note where they are, and the moment you tap the first one the rest turn into blank tiles. Then you tap the blanks in order from memory. Each round you clear adds one more number, so the test finds the longest sequence of positions you can hold in your head at once.

Where the Chimp Test comes from

The game copies a task from Kyoto University’s Primate Research Institute. Starting in 2004, Sana Inoue and Tetsuro Matsuzawa taught six chimpanzees, three mothers and their young, to touch the numerals 1 to 9 in order on a touchscreen. When the young chimps were about five, the researchers added a “masking” rule: touch the first numeral and all the others turn into white squares. That is exactly the rule you play by above.

Their December 2007 paper in Current Biology added one more twist. Five numerals flashed on screen for 650, 430 or just 210 milliseconds, too brief to scan the screen with your eyes, before being masked. Nine university students did worse the shorter the flash; a young chimp named Ayumu held steady and beat all of them on both speed and accuracy. The result made the task famous, and “can you beat a chimp?” became a web game.

The headline did not go unchallenged. Ayumu had a lot of practice on the task before he was tested; the students had none. When Silberberg and Kearns (2009) gave two people practice, their accuracy matched Ayumu’s, and in Cook and Wilson (2010) two practiced students clearly outperformed him. That is good news if you play it for training: the score moves with practice.

The idea is older than Ayumu, too. Since a 1972 doctoral thesis by P. M. Corsi, psychologists have used the Corsi block-tapping test: an examiner taps a sequence on nine blocks fixed to a board and you repeat it. A 1998 review called it arguably the single most important nonverbal task in neuropsychological research. The browser version most people know is Human Benchmark’s: start at four numbers, add one per level, three strikes and you are out.

How to play

1. Look at the grid. Numbers 1 to 4 sit on random tiles. There is no timer while you memorize.
2. Tap 1. All the other numbers flip to blank tiles.
3. Tap the blanks in order, 2, 3, 4 and so on.
4. Clear the round and the next one has one more number, up to 25.
5. A wrong tile costs one of your three lives and replays the same length. Lose all three and the game ends.

Your score is the longest sequence you cleared. In Memo, the same rules apply; the app adds a hint that costs a life, saves your best each day, and breaks ties on time.

What the Chimp Test measures

It is a test of visuospatial working memory: holding several locations, and their order, in mind for a few seconds. That is the same skill the Corsi block-tapping test measures; the Chimp Test is a self-paced cousin of that task.

It is not an IQ test and not a diagnosis of anything. Your score swings with sleep, attention and how many times you have played, which is exactly why it is useful to track over weeks rather than read once.

Does it improve your brain? What the research says

You will get better at the Chimp Test if you play it. Whether that makes the rest of your memory better is a separate, much weaker claim. Here is what the studies show.

• Practice closes the chimp gap. Ayumu was right about 79% of the time on five numerals flashed for 210 milliseconds, and none of the nine untrained students came close. Once people were given practice, they matched him in one follow-up and beat him in another. The skill is trainable, in chimps and in people.
• Trained games improve; untrained tasks barely move. In a six-week online study run with the BBC, 11,430 adults practiced memory, attention and reasoning games for at least 10 minutes a day, three times a week. Scores rose on every game they trained, with large effects, but gains on untrained tests were small at best, and a control group that just looked up answers to obscure general-knowledge questions improved about as much.
• Near transfer, not far transfer. A 2016 meta-analysis pooled 87 publications with 145 comparisons of working memory training. Right after training, people did reliably better on other verbal and visuospatial working memory tests. On reasoning, reading or arithmetic there was no convincing benefit compared with an active control group.
• The big review agrees. A 2016 review of brain-training studies found extensive evidence that training improves the trained tasks, less for closely related tasks, and little for distantly related tasks or everyday cognitive performance.

So treat your Chimp Test score as what it is: a measure of this skill, today. Watching it climb is satisfying and real; just don’t read it as a measure of your whole memory.

How to get a higher score

• Trace a path, not a list. Before tapping 1, run your eyes from 1 to the last number as one continuous route. A shape is easier to remember than nine separate addresses.
• Chunk in threes. Remember 1–2–3 as one zigzag, 4–5–6 as the next. Long sequences fall apart when every step is its own item.
• Use the free memorizing time. Nothing happens until you tap 1. Rushing the first second is the most common way to lose a life.
• Watch the edges. Numbers on the border and in corners are easier anchors than ones in the middle; hang the sequence on them.
• Play it daily, briefly. Five minutes a day beats an hour once a week. Most of the early gain is learning the strategy, so expect a quick climb and then a plateau.