N-Back Test - Online Working Memory Test (1, 2, 3-Back)
N-back test
n = 2: press Match (or A) whenever the square lights up in the same place as n steps earlier. Press Start when you are ready.
Each block has 20 + n squares; 6 of the 20 scored squares are matches (30%). A square is shown for 0.5 s, followed by 2.5 s of empty grid. Only the first press per square counts. Personal best: the highest n with at least 80% accuracy. This is a practice task, not a psychological diagnosis.
About the n-back test
The n-back task is one of the most widely used working-memory tests in psychology. You do not just remember one position – you keep updating a short list in your head, comparing each new square with the one n steps back and then letting go of the oldest one.
How it works
- 1 Choose n = 1, 2 or 3. With n = 2 you compare each square with the one two steps earlier.
- 2 Press Start. Every 3 seconds a square lights up for half a second. Press Match (or the A key) when it is in the same place as n squares ago; otherwise do nothing.
- 3 After the block you see hits, misses, false alarms, accuracy and d′. In adaptive mode the next block gets harder at 80% or more and easier at 50% or less.
Where the n-back comes from
The task goes back to the psychologist Wayne Kirchner, who in 1958 used it to study age differences in the short-term retention of rapidly changing information. Since then it has become a standard tool in working-memory research and brain-imaging studies, often in a spatial version like this one or with letters.
Hits, false alarms and d′
Accuracy alone can mislead: someone who never presses still gets 70% here, because 14 of the 20 scored squares are not matches. That is why the test also reports d′ from signal detection theory. It combines the hit rate and the false-alarm rate, d′ = z(H) − z(F): guessing gives values near 0, and the better you tell matches from non-matches, the higher it gets.
Does n-back training raise IQ?
In 2008 a widely cited study reported that training on a dual n-back task improved fluid intelligence. Later replications and several meta-analyses found the effect to be small or absent: people clearly get better at the n-back itself, but transfer to other tasks and to intelligence remains disputed. Treat this test as a measurement and a challenge, not as proven brain training.
What affects your result
The load rises quickly with n: many people find 1-back easy and 3-back hard. Tiredness, distraction and stress lower performance. A common source of false alarms are lures – a square in the same place as n − 1 or n + 1 steps ago, which feels familiar but is at the wrong distance. The result is not a medical or psychological diagnosis.