Go no-go test online: can you hold back?
See how fast you react, and how well you can stop a habit mid-press: the control that keeps you from interrupting or tapping out of reflex. Press for every M, hold back on every X.
40 scored letters, up to 1.5 seconds each · Tutorial and practice included · Nothing to install
Press for M. Wait for X.
MMM
X
PressLearn and practice
A short tutorial and 6 practice rounds that count for nothing.
The scored letters
No feedback as you go: the next letter simply appears.
Your results
Speed, accuracy and false alarms, read against reference ranges.
What is a go/no-go test?
A go/no-go test shows you one letter at a time: you press for every M and hold back on every X. Because most letters are M, pressing turns into a habit, and the task measures how well you inhibit that prepotent response, along with sustained attention, motor impulsivity and error monitoring. Press for M, Wait for X is CogniFit's digital version of the Go/No-Go paradigm, following the published procedure.1
Press for the M. Leave the X alone.
That's the whole rule. The catch is the rhythm: most letters are M, so by the time an X shows up, your finger is already on its way.
Press for M, wait for X
Go trial: press as fast as you can
- The letter
- An M, in the middle of the screen. This is a go trial.
- Your window
- Up to 1.5 seconds to answer. Faster counts.
- The one control
- Press it. A missed M counts as an omission.
Press for M, wait for X
No-go trial: keep your finger still
- The letter
- An X. This is a no-go trial: the hard part.
- Your window
- Let it run out. The next letter follows on its own.
- The one control
- Leave it alone. A press here counts as a false alarm.
One possible run of 40 scored letters
- Mostly go
- Go trials make up 75% of the task and no-go trials 25%, so three in four letters are M.1
- Mixed, not in blocks
- The Xs are spread through the task, so the gap between go and no-go isn't tangled up with tiredness.
- Fresh every time
- The letters and their order are drawn anew for each session, so a retake is never the same sequence.
One letter at a time, no verdict in between
A letter appears
Always in the middle of the screen, on a plain background.
You press, or you wait
M: press fast. X: do nothing. You have up to 1.5 seconds.
The next one follows
You aren't told whether you were right, so each letter starts clean.
Knowing the rule is easy. Following it at speed is the test.
What the go/no-go task measures
Its main target is response inhibition. Three related abilities come along with it, each one familiar from ordinary days.
- Response inhibition
- Stopping an action that is already primed and on its way: the prepotent response.
- Holding back a reply while someone is still finishing their sentence.
- Sustained attention
- Keeping your focus steady across a long run of repetitive events.
- Staying alert on a familiar drive where little changes.
- Motor impulsivity
- The pull to act before you have checked whether you should.
- Tapping "Accept" on a pop-up before reading it.
- Error monitoring
- Noticing your own slip as it happens, and adjusting for the next one.
- Sensing a typo before you look back at the screen.
Curious how your brakes hold up when the pace picks up?
How to read your go/no-go result
Every letter ends in one of four outcomes. Your results are built from how often each one happened, and how quickly you pressed.
| You press | You wait | |
|---|---|---|
| An M appears | HitA correct press, in time | OmissionA press you missed |
| An X appears | False alarmA press you should have held back | Correct waitYou held back |
- Speed
- Your median reaction time, in thousandths of a second: half your presses were faster, half slower. One very slow answer doesn't move it. Lower is faster.
- Steadiness
- How much your speed varied from one press to the next. A small number means a steady rhythm; a large one, bursts of fast and slow. Lower is steadier.
- Accuracy
- Out of every hundred answers you gave, how many were right. Higher is better.
- False alarms
- How many times you pressed on an X: the count that speaks most directly to holding back. Lower is better.
Sensitivity (d′): the key measure
It sets two numbers side by side: how often you pressed for an M, and how often you pressed for an X. The further apart they are, the higher your d′.
Pressing for everything, or for nothing, brings it down, so speed alone can't buy a good score.
- 0About 0: no better than guessing.
- 3+Around 3 or more: almost always right.
When you finish, the task shows your own numbers read against CogniFit's reference ranges, so you can see where your session sits. That's context, not a clinical assessment: one session describes that sitting, not who you are.
The matrix fills in only once you've played.
The science behind go/no-go
In a go/no-go task, one kind of stimulus asks for a response (go) and another asks you to withhold it (no-go). The classic description of this arrangement goes back to Donders, whose 1868 work on the speed of mental processes included a reaction to one stimulus and not to another (Donders, 1868/1969).2 When go trials are the majority, pressing becomes the default, so every no-go trial checks whether you can stop a response that's already primed; how strongly that response is primed shapes how much inhibitory control a given variant of the task demands (Wessel, 2018).3 This version follows the published Go/No-Go procedure.1
That is what this version is built to show: how well you stop a response that's already on its way, while keeping pace on the letters you should press. Research uses go/no-go performance to study response inhibition across groups of people, as in a meta-analysis of go/no-go task performance and psychopathology (Wright et al., 2014).4 The paradigm also has limits: the go/no-go and stop-signal tasks rely on different mechanisms, so one task doesn't capture every way of stopping an action (Raud et al., 2020).5
More detail: how this version is built
- What you see, for how long, what counts as an answer and what is measured are all set from the published procedure, not invented for this version.
- Go and no-go trials are mixed rather than run in blocks, so the difference between them isn't measured together with tiredness.
- Steadiness is computed on correct answers only, because the timing of errors comes from a different process.
- d′ is calculated as z(hit rate) − z(false alarm rate), with each rate held between 0.01 and 0.99.
What this version can tell you
- How fast and how steadily you pressed in this session.
- How often you pressed when you should have held back.
- Where your numbers sit against the reference ranges on your results screen.
What it can't tell you yet
- Anything about your health: it doesn't diagnose and isn't a clinical assessment.
- What a given score means for you personally: that needs population data for this version that hasn't been collected yet.
- Study findings describe groups, not any one individual, and that includes research comparing go/no-go performance across groups (Wright et al., 2014).4
You've read the paradigm. Now feel it from the inside.
Questions about the go/no-go test
How long does the go no-go test take?
The scored part is 40 letters, each allowing up to 1.5 seconds to answer. Before it come a short tutorial and 6 practice rounds that don't count. You can repeat the tutorial as often as you like, and that doesn't affect any result.
Do I need an account to take it?
No. You can take the task and see your results at the end. Signing up is only for keeping your results and tracking them over time.
Can I take it on my phone?
Yes. It works on a computer, a tablet or a phone, and you answer with a single control.
Why doesn't it tell me when I'm wrong?
During the scored part the next letter simply appears, with no feedback. That keeps each answer about the letter in front of you, not the one before.
Can I take the test again?
Yes. The letters and their order are drawn fresh every time, so you won't meet the same sequence twice.
Is this a test for ADHD or any other condition?
No. It measures how you performed in one session and places it against reference ranges. It doesn't diagnose anything. If you have concerns about your attention or health, talk to a qualified professional.
Now you know what it measures. Can you hold back on the X?
Learn the rule in the tutorial, play the 40 scored letters, and see your speed, accuracy and false alarms at the end.
Sources
- CogniFit. Press for M, Wait for X (Go/No-Go): task description, conditions and reported measures.
- Donders, F. C. (1969). On the speed of mental processes (translation of the 1868 original). Acta Psychologica, 30, 412-431. https://doi.org/10.1016/0001-6918(69)90065-1
- Wessel, J. R. (2018). Prepotent motor activity and inhibitory control demands in different variants of the go/no-go paradigm. Psychophysiology, 55(3), e12871. https://doi.org/10.1111/psyp.12871
- Wright, L., Lipszyc, J., Dupuis, A., Thayapararajah, S. W., Schachar, R. (2014). Response inhibition and psychopathology: A meta-analysis of go/no-go task performance. Journal of Abnormal Psychology, 123(2), 429-439. https://doi.org/10.1037/a0036295
- Raud, L., Westerhausen, R., Dooley, N., Huster, R. J. (2020). Differences in unity: The go/no-go and stop signal tasks rely on different mechanisms. NeuroImage, 210, 116582. https://doi.org/10.1016/j.neuroimage.2020.116582
This page and the task are for information and self-exploration only. They are not medical advice, do not diagnose any condition and do not replace care from a qualified health professional.
United States