What is dual n-back?
N-back is a memory-updating task. A square lights up in a grid, and you may also hear a spoken letter or note. Each new square or sound is one step. For each thing you follow, decide whether it is the same as N steps earlier. Dual means you follow two things at once: the square and the sound. The 2 in 2-back means you compare with 2 steps back. Those are different choices.
How to play
- Try the short 1-back practice. Each new square is one step. Tap Position when the same square lights up again 1 step later. If nothing is the same, tap nothing.
- In dual mode, you follow the square and the sound separately. Tap Position when the square is the same as N steps back, and Sound when the sound is. One, both or neither may match; if nothing is the same, tap nothing. At 2-back, compare step 3 with step 1, step 4 with step 2, and so on.
- On a keyboard, use A for position, L for audio, S for colour and K for shape. Space pauses. On a phone, use the buttons below the grid.
- In triple mode, add colour. In quad mode, add shape too. Colour always has a numbered badge so you can distinguish it without relying on hue.
A familiar schedule, with room for your own pace
The standard preset follows the timing of the Jaeggi protocol: 20 scored steps plus N warm-up steps, one every 3 seconds, and an optional daily set of 20 blocks. Take a break between blocks. Adaptive mode goes up if everything you follow has at most 2 errors, down if any one of them has at least 6, and stays at the same level otherwise. Errors are missed matches plus false alarms. Levels stay between 1 and 9. [1] [2]
By default, Sound uses spoken letters; Notes is an optional adaptation. Colour, Shape, the brief tutorial and the mobile controls are also our adaptations; this trainer is not a replication of a research intervention. Manual mode lets you choose your pace and length.
The research schedule is an option, not a target you have to meet. Stop when you want a break.
Fair sequences, useful comparisons
Each thing you follow matches on 30% of scored steps. Every pair of them matches together on 10%, including steps with three or four matches in triple and quad modes. There are single matches too. The sequence number lets you repeat a run. Your graphs compare your own sessions; there is no player ranking. Higher levels and faster steps are different challenges, so keep settings in mind.
The average of the matches you caught and the non-matches you correctly left alone. Doing nothing scores 50%. Warm-up steps do not count.
What does the research say?
People get better at the task they practise. Evidence that n-back training carries over to other abilities is mixed and debated. Jaeggi and colleagues reported reasoning gains in an early study. Later reviews, including Soveri and colleagues and Melby-Lervåg and colleagues, found much stronger evidence for improvement on trained or closely related tasks than for broad transfer. These graphs show your n-back practice, not intelligence or everyday ability. This is not medical advice. [1] [3] [4]
A few common questions
Is the trainer completely free?
Yes. Every mode, graph, export and account progress feature is free. You can start without signing up, and there are no ads or paywalls.
Where are my earlier n-back test results?
Your earlier website test results are still in your history. New practice sessions have their own progress graphs. Training here does not change your Helibrain app scores, levels or Brain Score.
Will this make me smarter or raise my IQ?
We do not promise that. Practice improves performance on n-back. Whether that improvement transfers to other abilities is mixed and debated, and broad gains are not established by your training graph.
How much should I practise?
Start with one short session and find a comfortable routine. The optional daily preset follows the block count used in early research, but it is not a proven ideal dose or a requirement. You can stop after any block.
Can I keep my progress on another device?
Yes. Choose Save my progress and connect a free account. After signing in on another device, open Your progress here. JSON export and import also let you move your practice history.
Can I pause or close the page?
Finished sessions are saved immediately. An unfinished session pauses when you leave the page and can resume after a reload. The current step replays when you return; only finished sessions count in your graphs.
How this test works
This is a free n-back practice tool. N-back trains working memory, the memory you use to hold and update information. Squares light up on a grid, and in dual mode you also hear letters. For each stream, press when the current cue matches the one N steps back. A standard block has 20 scored cues, one every 3 seconds. Your score is balanced accuracy, so pressing nothing gives 50%. Psychologists first described the n-back task in 1958.
Version nback-training-1 · Cite this test:
Helibrain (2026). N-back test, vnback-training-1. https://helibrain.com/n-back-test/methods
References
- Jaeggi SM, Buschkuehl M, Jonides J, Perrig WJ (2008). Improving fluid intelligence with training on working memory. PNAS 105, 6829–6833. DOI: 10.1073/pnas.0801268105
- Jaeggi SM, Buschkuehl M, Shah P, Jonides J (2014). The role of individual differences in cognitive training and transfer. Memory & Cognition 42, 464–480. DOI: 10.3758/s13421-013-0364-z
- Soveri A, Antfolk J, Karlsson L, Salo B, Laine M (2017). Working memory training revisited: A multi-level meta-analysis of n-back training studies. Psychonomic Bulletin & Review 24, 1077–1096. DOI: 10.3758/s13423-016-1217-0
- Melby-Lervåg M, Redick TS, Hulme C (2016). Working memory training does not improve performance on measures of intelligence or other measures of “far transfer”. Perspectives on Psychological Science 11, 512–534. DOI: 10.1177/1745691616635612