Chimp Test Statistics

A distribution with its left end cut off

The test begins at four numbers, so no score below four exists and the chart starts there. That is a deliberate truncation rather than an absence of data: the levels below four are cleared by essentially everyone and running them would have added rounds without adding information.

The consequence is that the bottom bar of this chart is not comparable to the bottom bar on the memory pages. It contains everyone who failed at the starting level, which mixes genuine difficulty with people who clicked out of order while working out what the test wanted.

Two failed rounds are allowed rather than one, and failures retry at the same count. That keeps a single mis-click from ending a run, and it stops a run from oscillating around one level indefinitely the way a drop-back rule would.

Why scores here are lower than people expect

The task divides attention in a way pure memory tests do not. You have to hold the layout and simultaneously track which number you are on, and each click is a small interruption to the thing you are trying to keep. Most runs end from clicking two numbers out of order rather than from forgetting where one was.

That makes the distribution narrower than the memory ones. There is less room for strategy to separate people, because the binding constraint is not how much you can hold but how well you can hold it while acting.

The upper tail belongs mostly to attempts that plan the route while the numbers are still visible instead of memorising positions and working out the order afterwards. It is the single largest difference between a middling score and a good one.

About the comparison the test is named for

The original laboratory task involved trained chimpanzees and largely untrained human students, so the widely repeated claim that chimps beat humans at this is true of that experiment and weaker as a general statement. Practice on the task was not equal between the two groups.

Nothing in the distribution on this page speaks to that comparison either way. It is a browser sample of self-selected human attempts, most of them first or second tries, which is not the population any published animal comparison used.

It is worth saying plainly that this is a game rather than an instrument. The score is interesting, the chart makes it more interesting, and neither is an assessment of anybody's cognition or health.

This is a measurement toy, not a medical or psychological test. No score here says anything about your health, attention or cognition, and it is not a screening tool of any kind.

Questions

What is the average chimp test score?

Most attempts cluster in the low teens on this site, with the chart showing the current spread and sample size. Scores well into the high teens are uncommon.

Why does the chart start at four?

Because the test starts at four numbers. Lower counts are cleared by nearly everyone, so running them would only lengthen the test.

Does a low score mean anything is wrong?

No. This is a game with an unusual attention demand, most people are playing it for the first time, and it is not a screening or diagnostic tool of any kind.