Attention Span Test — Measuring What, Exactly?
You have probably seen the claim that the human attention span has fallen to eight seconds, less than a goldfish. It gets cited constantly and it has no source. Nobody has ever produced the study, the goldfish figure has no origin either, and attention researchers do not use a number like that because attention is not a duration you can quote.
What can be measured is more specific and more useful: how well you hold a task when nothing is prompting you, how well you resist an irrelevant signal, and how much you can keep track of at once. Those are separate, and each has a real paradigm behind it.
What you can actually measure here
Each of these targets one component rather than 'attention' in general. That specificity is the whole reason they are worth doing.
Stroop Test
Selective attention under interference. Reading is automatic, so naming the ink colour means actively suppressing the word — the classic demonstration that attention is effortful.
~2 minPlay →Simon Effect Test
The same conflict from a different direction: the stimulus is unambiguous, but where it appears fights the hand you need to move. Your Simon cost is the size of that leak.
~1 minPlay →Multiple Object Tracking
Divided attention. Several identical dots move at once and you have to hold a subset in mind without being able to tell them apart by sight.
~1 minPlay →Change Blindness Test
Where attention isn't. A large, obvious change is invisible if your attention is elsewhere — usually far longer than people expect.
~2 minPlay →
Why 'attention span' resists being a single number
Attention is not a tank that empties. The same person can lose the thread of a meeting in four minutes and stay locked into something absorbing for three hours. That is not a contradiction — it means the interesting variable is the interaction between the task and the person, not a fixed personal capacity.
This is also why the honest tests do not report a duration. They report how accurately you performed under a specific demand, which is a claim you can actually defend.
The parts that can be measured separately
Selective attention is holding onto the relevant feature while an irrelevant one competes for the response — the Stroop and Simon paradigms both isolate this, in different ways.
Divided attention is tracking more than one thing at once, which is what multiple object tracking targets. Most people manage about four items, and the point of the test is finding where your own limit sits rather than passing it.
Visual scanning is covering a field systematically without losing track of where you have been. It sounds mechanical, but it is where a lot of everyday 'I looked right at it' failures come from.
Attention, or interest?
When people say their attention span is gone, the evidence they give is usually about things they did not want to do: a document that would not go in, a meeting that slid past. Meanwhile the same person can lose an evening to something absorbing without noticing. That pattern points at motivation and task design at least as much as at capacity.
This is not a scolding — it is a useful distinction, because the two have completely different remedies. If the limit is capacity, practice and rest are the levers. If the limit is that a task offers no feedback, no visible progress and no clear stopping point, then no amount of attention training will fix it, and restructuring the task will. Most people trying to fix their attention are actually facing the second problem.
It also explains why a test can feel easy while your workday feels impossible. These tests are short, unambiguous and give you a number at the end — they are engineered to be attention-friendly. That makes them decent instruments and poor simulations.
If your attention genuinely feels worse
Two things move same-day attention far more than anything on this site: sleep and how many streams you are running at once. Neither is a satisfying answer, and both have far stronger evidence behind them than any brain-training claim.
A test here can tell you how you performed today against how you performed last week. It cannot tell you whether something is wrong, and it is not built to. If your attention has changed in a way that worries you, that is a conversation for a professional, not a score on a website.
Frequently asked questions
Is the human attention span really 8 seconds?+
No. The claim traces back to a widely-repeated statistic with no identifiable source study, and the goldfish comparison has no source either. Attention researchers don't describe attention as a duration, because how long you stay with something depends heavily on what the something is.
What does an attention span test actually measure, then?+
A well-formed one measures performance under a specific, controlled demand — accuracy while an irrelevant signal competes, how many moving items you can track, how long it takes you to notice a change. Those are defensible measurements. A duration in seconds is not.
Can this tell me if I have ADHD?+
No, and no online test can. The tests here are practice exercises, not screening instruments — they are not validated, not normed against a population, and not administered under controlled conditions. A diagnosis involves clinical history and trained assessment. If you're concerned, that's the route.
Does attention span shorten with age or screen time?+
Both are studied and neither has the clean answer that headlines suggest. What is consistently observed is that heavy task-switching degrades performance while you're doing it — which is a statement about the situation you're in rather than about permanent damage.