No. 08
How Digit Span Is Measured — This Test's Rules and Limits
How the standard digit span task is given, what this site does the same and differently, and an honest look at how your device and environment affect results.
A digit span test looks simple, but scores change with presentation speed, mode and the stopping rule. Here is how this site measures span, and the limits to keep in mind when interpreting results.
How the standard task is given
In the digit span tasks used in intelligence and neuropsychological testing, an examiner usually reads the digits aloud at one digit per second. Each length has two trials, and testing stops when both trials at the same length are failed. The digit strings, scoring method and age norms are all standardized together.
This site's rules
- You start at the starting length (3 digits by default).
- A correct answer adds one digit.
- After a miss, you get one more new string at the same length.
- Two misses at the same length end the test, and your score is the longest length you got right at least once.
- The test continues up to 24 digits and ends there if you get everything right.
The difference from the standard task is that once you pass the first trial at a length, you move straight to the next length without a second trial. This saves time but changes how the score is calculated, so it is not on the same scale as the raw score of a standard test.
How the strings are generated
- Digits 0–9 are drawn at random, but the same digit never appears twice in a row.
- Up to 10 digits, no digit repeats within a string.
- Runs of three that go up or down by one, like 1-2-3 or 8-7-6, are never generated, to prevent overly easy chunks.
Presentation modes
| Mode | Presentation | Character |
|---|---|---|
| --- | --- | --- |
| All at once | The whole string shown for number of digits × your time setting | Easy to scan and group; the most forgiving condition |
| One by one | One digit per interval (visible for 75% of the interval) | Serial recall close to the standard task |
| Listening | Speech synthesis reads one digit per interval | Auditory, like the standard task; affected by voice quality |
Input is blocked while the digits are presented and becomes available once they disappear. If you switch to another tab during a presentation, that round is voided and restarts with a new string. There is no time limit on input, so your typing speed and screen refresh rate don't affect your score.
What affects accuracy
- Speech synthesis: it uses your browser's and operating system's voices, so pronunciation speed and clarity vary by device. On some devices the first word comes out late, so with short intervals (0.5 seconds) the rhythm may wobble.
- Screen: in all-at-once mode, long strings shrink to fit the screen width. On small screens, landscape mode or a larger screen helps.
- Environment and condition: noise, tiredness and nerves lower scores. See what affects working memory.
- Practice effects: repeating the test back to back tends to raise scores. When comparing results, use the same settings and a similar environment.
What these results can't tell you
This site doesn't show comparisons like percentiles, age-adjusted scores or "IQ equivalents". Such comparisons require norms built with standardized procedures and samples, and this tool's conditions are different. Results are for fun and self-checking, and they are not a substitute for an IQ test or a medical diagnosis.
References: Wechsler, D. (2008) WAIS-IV Administration and Scoring Manual, Pearson · Jacobs (1887) Mind 12:75–79