Subitizing Test — Count Dots at a Glance, Free Online

The Subitizing test looks at your ability to tell how many dots there are at a glance, without counting them one by one — a skill known as subitizing. There are 30 trials in total: every count from 1 to 9 appears 3 times, and 3, 4 and 5 dots each appear once more. The order is shuffled every time, so you can't predict how many dots are coming next.

Each trial runs like this. A cross appears in the center for 0.45 seconds — keep your eyes on it. Then white dots flash on a dark background for 0.18 seconds and disappear, and a noise screen of random dots follows for 0.25 seconds to wipe out any afterimage. After that, you tap the number you saw on a 1–9 pad. The dots are placed in an invisible 5×5 grid so they never overlap, and your response time is measured from the moment the pad appears.

The first result screen shows your overall accuracy (%) and how many of the 30 trials you got right; the comparison with others is still a provisional reference estimate. In the detailed report, your 'subitizing range' is the largest count, climbing from 1 without a break, that you answered at least 90% correctly. Each count appears only 3–4 times, so a single miss stops the range right there. A range of 4 or more reads as above average, 3 is the typical range most adults show, and anything narrower simply means it came out narrow this time.

The report also shows your mean response time on correct answers for 1–4 dots versus 5–9 dots, plus your accuracy for each count. The point where you stop being fast and accurate and start slowing down or making errors marks the boundary between numbers you simply 'see' and numbers you have to count or estimate. Because the dots pass so quickly, fatigue, drowsiness, distraction, a dim or small screen, or simply mistapping the pad can easily shift the result. Treat today's result as a snapshot of that day.

The word 'subitizing' was coined by Kaufman, Lord, Reese and Volkmann (1949) in their study of how people discriminate visual number. Their observation — that small numbers are grasped at a glance while larger ones must be counted or estimated — became a foundation for later research on number perception. Soundary's version is its own web implementation of that paradigm; the result shows a perceptual tendency, not a diagnosis.

Your results are saved, so when you retake the test you can compare with your earlier record. Because each count appears only a few times, one slip can move your range up or down by one step, so looking at several runs together is more reliable than a single value. For a fair comparison, use the same device, similar screen brightness and a similar time of day. For classes or research, a group link lets several people take part anonymously, without signing in.