Working Memory Test: How the 2-Back Task Works

Working memory — the ability to hold and actively update a small amount of information over a few seconds — is one of the most studied components of cognitive ability, because it underlies tasks as different as mental arithmetic, following multi-step instructions, and reading comprehension. The CognitiveIndex Working Memory Test uses a 2-back task, a well-known research paradigm, to estimate this ability with 30 timed trials.

What Is the N-Back Task?

The N-back task was introduced by psychologist Wayne Kirchner in 1958 and has since become one of the most widely used working memory paradigms in cognitive research, including later, more demanding versions studied by researchers such as Susanne Jaeggi. In an N-back task, a sequence of stimuli — letters, in CognitiveIndex's version — is presented one at a time. On each trial, you decide whether the current item matches the one shown N steps earlier. A "2-back" task means comparing the current letter to the letter shown two positions before it, which requires holding two items in mind and continuously updating that comparison as new letters appear.

This is a generic, decades-old academic task design that is not owned by any test publisher. CognitiveIndex's letter set, trial sequence, and scoring are original to this site.

How the CognitiveIndex Test Works

The test shows 30 letters, one at a time. For the first two letters, there is no earlier letter to compare against, so no response is expected. From the third letter onward, you click Match whenever the current letter is the same as the letter shown two steps earlier, and you do nothing when it isn't. Each letter stays on screen briefly before the next one appears automatically.

How Scoring Works

The report shows two separate numbers:

• Accuracy — the percentage of eligible letters correctly judged as either a match or a non-match.
• Speed score — the percentage of eligible letters that were judged correctly and responded to within a target time window.

Both correctly clicking "Match" on a true match (a hit) and correctly doing nothing on a non-match (a correct rejection) count toward accuracy. Clicking "Match" on a non-match (a false alarm) or missing a true match both count as errors. This scoring approach, common in N-back research, avoids rewarding a strategy of clicking "Match" on every trial just to catch the true matches.

Why Working Memory Is Tested Separately From Reasoning and Speed

Working memory, reasoning, and processing speed are related but distinguishable cognitive factors. A person can have strong reasoning ability, as measured by a test like CognitiveIndex's Numerical & Verbal Reasoning Test, while finding a continuously updating memory task like the 2-back more demanding, or vice versa. Testing working memory with its own dedicated task, rather than folding it into a general reasoning score, gives a clearer picture of this specific ability. For a related, though different, task type, see CognitiveIndex's Processing Speed Test.

How to Prepare

• Focus on the letter two positions back specifically — it is easy to accidentally compare against the immediately preceding letter (a 1-back error) under time pressure.
• Respond as soon as you are confident rather than waiting; the task is timed per letter, and hesitating too long is scored the same as not responding.
• Take the test in a quiet setting without distractions, since the 2-back task is sensitive to momentary lapses in attention.
• Do not expect a perfect score; even in research settings, 2-back accuracy well below 100% is common and does not indicate a memory problem.

Limitations

This is an original practice task for education and self-review. It has not been clinically validated and should not be used to diagnose memory, attention, or neurological conditions. A substantial body of research has also found that N-back training does not reliably transfer into broad improvements in general fluid intelligence, so a single practice session should be treated as a snapshot of task performance, not a fixed measure of memory capacity.

Ready to test your working memory? Take the Working Memory Test.

Frequently Asked Questions

What does "2-back" mean?

It means you compare the current letter to the letter shown two positions earlier in the sequence, rather than the immediately preceding letter (which would be a "1-back" task).

How many letters are in the test?

30 letters total. The first two cannot be judged as a match or non-match since there is no earlier letter to compare against, so scoring is based on the remaining 28 eligible trials.

What counts as a correct response?

Clicking "Match" when the current letter matches the letter two steps back, or doing nothing when it does not. Both are scored as correct; only false alarms (clicking on a non-match) and missed matches count as errors.

Does a lower score mean I have a memory problem?

No. N-back accuracy varies widely across the general population and is affected by attention, fatigue, and familiarity with the task format. A single online result is not a diagnostic measure.

Sources

• Kirchner, W. K. (1958). Age differences in short-term retention of rapidly changing information. Journal of Experimental Psychology.
• N-back, background on the task paradigm and its use in cognitive research.
• Overview of Psychological Testing, National Academies via NCBI Bookshelf.