Go / No-Go test online: press, then hold back

Letters flash one at a time and you press for each one, except X. Most letters are go letters, so pressing turns into a habit, and the test counts how often that habit wins when X shows up.

Start the Go / No-Go test

160 letters · about 5 min · space bar or tap

When the last letter has gone you see your numbers, and a free account keeps them.

K M X T
K, M: press. X: hold back1 letter in 4 is an X

How it works

A 12-letter practice with feedback, then 160 scored letters without it. Each letter takes 1.5 seconds from its start to the next one, so the scored part runs about 4 minutes and the whole task about 5.

  1. 1 · Practice

    12 letters with feedback

    8 go letters and 4 X's in random order. A tick follows each right answer; a miss or a press on X brings a short reminder of the rule.

  2. 2 · Go letters

    Press for every letter

    Any letter from A to Z apart from X, drawn at random. Press the space bar, or tap anywhere on a touch screen. 120 of the 160 scored letters are go letters.

  3. 3 · No-go letter

    Hold back on X

    When X appears, do nothing and let it pass. 40 of the 160 letters are X, mixed in at random.

  4. 4 · Each letter

    Half a second on screen

    A letter shows for 0.5 seconds, and a press counts until 1 second after it appeared. Then the screen stays blank for 0.5 seconds before the next one.

The letter appears large in the middle of the screen. Pressing does not make it vanish early, and a press during the blank gap is not recorded. On touch screens a small line at the bottom reads "Tap anywhere to respond". After the practice comes a 3, 2, 1 countdown and a thin bar at the top that fills as you go. During the scored letters nothing tells you whether you were right.

Clinical versions of the task can be more elaborate. The Parametric Go/No-Go Test, for example, has 3 levels of rising difficulty, and its scores separate processing speed, sustained attention and inhibition (Votruba and Langenecker, 2013). This version keeps to one level with a single no-go letter, which makes the rule easy to hold and the habit to press strong.

What it measures

The two kinds of letter ask for opposite things. Go letters ask you to act fast; X asks you to stop an action you have been repeating.

Go letters

  • Response speed: how quickly you press once a letter appears.
  • Sustained attention: catching go letters for about 4 minutes without drifting off.
  • Omissions: go letters you let pass without a press.

The X

  • Response inhibition: stopping a press that 3 letters in 4 have trained you to make.
  • Commission errors: presses on X, the task's main count.
  • Your strategy: rushing tends to bring more presses on X, while a slower, careful pace can bring fewer.

Omissions and commissions answer different questions. In 1,151 children aged 9 and 10, presses on no-go letters went with caregiver and teacher ratings of hyperactivity and impulsivity, and missed go letters went with ratings of inattention, though each kind also overlapped with the other (Bezdjian et al., 2009). In that study the two error types were almost unrelated to each other.

Go / No-Go is one of the 4 tasks in the ADHD check, under response control. What the task captures is still debated: in a 2020 brain-imaging study of 143 people aged 8 to 21, the activity usually read as the brain stopping a press was unrelated to how well people did (Weigard et al., 2020). Tasks and questionnaires also often disagree; across 20 studies of executive function, only about 1 in 4 correlations between task scores and rating scales reached significance (Toplak et al., 2013).

A test, not a diagnosis.

One run of this task describes how you pressed and held back on that day. It cannot confirm or rule out any condition.

What you get

Your numbers appear as soon as the 160th letter has passed.

On the result screen

  • A headline set by your share of presses on X, in 3 fixed bands: under 20%, 20% to under 35%, and 35% or more. The bands do not change with age.
  • Three figures: that share, labelled false alarms, your average go time in milliseconds, and d′, which combines hits and false alarms.
  • Short lines on presses on X, on missed go letters and on your speed, with a few sentences that read them together.
  • A details panel with each error count, go and no-go accuracy, and a CSV of every scored letter.

Once it is saved

  • With a free account the result joins your other tests and is there on any device you use.
  • Filed under attention in your results, next to Flanker and the other attention tasks.
  • A next test is suggested straight away, so you can add another area while you are warmed up.

How this version is scored

Each scored letter is logged with its type, whether you pressed and how long the press took. The result screen is built from that log.

Hit
A press on a go letter within 1 second of it appearing. Go time runs from the moment the letter appears to your press.
Omission
A go letter with no press, shown as a count out of 120 and as a share.
Commission error
A press on X, shown as a count out of 40 and as a share. The headline is based on this share.
d′ (d-prime)
d′ = z(hit rate) minus z(false alarm rate), where z turns a proportion into a standard score. Both rates are held between 1% and 99% first, so a perfect run still gives a finite number. Higher means go letters and X were told apart more cleanly.
Enough presses to read
Holding back on X only means something when there is a pressing habit to stop, so the result needs at least 60 of the 120 go letters answered. Below that the screen says the attempt cannot be scored, shows your counts and leaves d′ out.

Why the letter stays put. A press does not end a letter early, so every letter takes the same 1.5 seconds and the pace never changes. A fixed rhythm with 3 go letters for every X is how the task builds the habit that X then has to interrupt.

Random letters. Go letters are drawn at random from the 25 letters other than X on every attempt, and all 160 are shuffled, so no two runs are the same.

No percentile. The result screen keeps to your own counts and does not rank you against other people. The published Go / No-Go tables this site holds fit its results poorly, and versions differ in letters, timing and how often the no-go letter comes up.

Before you start

Four minutes of letters asks for steady attention, so set yourself up before the countdown.

For a cleaner attempt

  • On a keyboard, keep one thumb resting on the space bar.
  • On a touch screen, keep a finger just above the glass: a tap anywhere counts.
  • Close other tabs and silence your phone; the letters keep coming whether you look or not.
  • Aim for speed and accuracy together, as the instructions ask, rather than waiting to be sure.

Interpret cautiously when

  • You were tired, in pain, rushed or interrupted.
  • Assistive input or a motor-access need adds delay to each press.
  • You used a phone, where touch timing is less exact than a key.
  • Many go letters were missed: few presses on X then say little about holding back.

Questions

What does the Go / No-Go test measure?

Mainly response inhibition, the ability to stop an action you are about to make. Presses on X count how often the pressing habit won. Missed go letters and your go time add information about sustained attention and speed.

What is a good score on a Go / No-Go test?

Few presses on X together with few missed go letters is the stronger pattern, and d′ sums up both. There is no single standard, because versions differ in letters, timing and how often the no-go letter appears. In one large online study, 521 adults averaged a d′ of about 3.6 on that study's own version (Eisenberg et al., 2019). Your result screen uses fixed bands for presses on X, not age norms.

Can I do the Go / No-Go test on a phone?

Yes. On a touch screen a tap anywhere counts as a press, and the instructions change to say tap. Phone timing is less exact than a keyboard, so compare a phone run with other phone runs.

What is the difference between omission and commission errors?

An omission is a go letter you let pass without pressing. A commission is a press on X, the letter you were meant to skip. In a study of 1,151 children the two kinds of error were nearly unrelated, which is why the result screen reports them separately.

Sources