Processing speed
How quickly you take in information and act on it. It is one of the seven domains we measure, and it is the one where how you measure it changes the answer most.
First, the disambiguation, because the phrase is overloaded. "Processing speed" in this section always means the cognitive sense — a person taking something in and responding. It has nothing to do with a computer's clock speed, a CPU benchmark, or how fast a phone feels. Search results for the phrase are dominated by the computing sense, and that is not what any page here is about.
What it is
Processing speed is the rate at which you carry out simple, well-understood operations — recognising something, matching it to a rule, and acting. The operations themselves are deliberately easy. What is being measured is not whether you can do them, but how quickly you keep doing them.
That is why speed tasks tend to look trivial. If a task is hard, the time it takes tells you about the difficulty. If a task is easy and you still vary in how fast you complete it, the time tells you about something else.
It is a rate, not a capacity. Working memory asks how much you can hold at once. Pattern recognition asks what you notice. Processing speed asks how fast you get through material you already understand. They are separable, and a person can be unremarkable on one and unusual on another.
What genuinely moves it
More than the number suggests, and much of it has nothing to do with you.
- The device. A phone screen and a physical keyboard are not the same instrument. On a task scored by completed responses, how fast you can physically press is part of the number.
- Familiarity with the task. Nearly every speed measure rises on a second attempt, because the rule stops needing to be looked up. That is learning the task, not getting faster.
- Sleep, illness, medication, caffeine, time of day. Ordinary daily variation is large relative to the differences people are usually trying to detect.
- Whether you were interrupted. A single distraction inside a timed window moves the result more than most stable characteristics do.
This is the honest frame for any speed number, including ours: it describes a rate under a specific set of conditions on a specific device. It is a real measurement. It is narrow.
The part almost nobody publishes
The most consequential thing about a speed measurement is not the person. It is the equipment.
When an online test tells you your reaction time was 240 milliseconds, that figure includes the time your screen took to show the stimulus, the time your input device took to register the press, and the time the operating system took to hand both to the page. Those delays are not small, and they are not the same on two different devices.
We take this seriously enough that it decided which task we built. Our Processing Speed instrument is a symbol-matching task scored over ninety seconds, not a reaction-time test — because a rate measured over ninety seconds barely notices a few tens of milliseconds of equipment lag, and a single reaction-time reading is dominated by it.
We have not measured anyone. Our reference sample is zero people. We publish that figure as a zero rather than omitting it.
So nothing on this site will tell you where your speed sits relative to other people. The methodology page sets out exactly where we are and what would have to change.
Take the symbol match test → — ninety seconds, no account, nothing stored.
Read next
Are online reaction time tests accurate?
The question has a real answer, and it has been measured directly — with the human taken out of the experiment, so that what remains is the equipment. The numbers are larger than almost anyone assumes, and they are the reason this site does not publish a reaction time.
Are online reaction time tests accurate?
Check our working
- The instrument register — every task, its norming rung, its sample size, its version
- How BrainScore works — and where we are on the norming ladder
- Corrections — every change we have made to a published claim, dated