Go / No-Go: The Science of Knowing When Not to React

Being fast is only half of reacting well. The other half is notreacting when you shouldn't: not biting on a feint, not swinging at a ball outside the zone, not jumping the gun. Go / No-Gotrains both at once. It is one of the oldest tasks in experimental psychology, and here's what the research says about it and what score to expect.

How the game works

After the countdown and a random wait, the screen shows either GO or STOP.

• On GO, tap as fast as you can. Your time counts towards your average.
• On STOP, keep your hands still until the cue has gone (1.2 seconds). Holding successfully doesn't count as a rep, so you get another cue.

About 25 % of cues are STOP on Easy, 30 % on Medium and 40 % on Hard, and you never get more than two STOPs in a row. Reacting on STOP is an error. On Easy you just repeat the rep, but on Medium and Hard it resets the set. Tapping before any cue appears is a false start, as in every other game.

A 150-year-old experiment

In 1868 the Dutch physiologist F. C. Donders tried to time thought itself. He compared three kinds of reaction:

• A-reaction: one signal, one response (a simple reaction, like our Lights Out).
• B-reaction: several signals, each with its own response (a choice reaction, like Left / Right).
• C-reaction: several signals, but you respond to only one of them and ignore the rest. This is go/no-go.

Donders found the C-reaction slower than the simple one and faster than the full choice. He reasoned that subtracting one time from another isolates a mental step. The C-reaction minus the simple reaction gave him about 36 ms for telling the signals apart. His “subtraction method” became a foundation of cognitive psychology. A 2018 replication by Roelofs confirmed the basic ordering, though it found the difference between the C- and B-reactions much smaller than Donders reported.

What the research says

Typical Go times

Hasegawa and colleagues (2021) tested 213 university students on a go/no-go task with 30 % no-go trials, close to our Medium. Average Go reaction times were 306 to 317 ms, depending on how fast the trials came. That is noticeably slower than the roughly 213–231 ms found for simple reactions in large lab studies. Having to check that the cue is the right one costs time.

The rarer STOP is, the harder it is to hold

When most cues are GO, tapping becomes a habit, and habits are hard to stop. In a study by van Campen and colleagues (2013), people who saw 20 % no-go trials averaged 333 ms on Go but wrongly responded on about 21 %of no-go trials. When 80 % of trials were no-go, Go responses slowed to 431 ms, and false responses almost disappeared (0.3 %). The mistaken responses were fast, quicker than the correct ones, which is what jumping before you've checked looks like. The groups were small, but the pattern matches the wider literature. Young and colleagues (2018) found that fast-paced tasks with two to four Go trials for every no-go produce the most false alarms.

Error rates vary enormously between versions of the task, from about 3 % to about 30 % in student samples. Fast, repetitive tasks with rare no-go trials sit at the top of that range. Ours sits lower: STOP is fairly common (25–40 %), and each cue follows a 1.5–5 second wait, so you aren't on autopilot.

Speed and accuracy trade off

You can always buy accuracy with time, by waiting a little longer to be sure. The research consistently shows that the conditions that make Go faster also make STOP errors more likely. The skill this game measures is keeping your Go time down without letting errors creep up.

Does it matter for sport?

Athletes do seem to be better at it, though studies comparing groups can't show training is the cause:

• Nakamoto and Mori (2008) found baseball and basketball players faster than non-athletes on both a simple task and a baseball-style go/no-go task.
• Muraskin, Sherwin and Sajda (2015) found EEG signs of stronger inhibition in college baseball players during a pitch-recognition go/no-go task. Their title says it all: “knowing when not to swing”.
• A 2023 meta-analysis by Albaladejo-García and colleagues pooled 11 studies and found athletes stopped responses faster than non-athletes, a small-to-moderate difference.

Whether practising the task makes you better at anything else is less clear. A three-week training study by Enge and colleagues (2014) found people improved on the trained tasks, but no more than an active control group, and with no transfer to other tests. Treat Go / No-Go as a way to measure and sharpen your restraint in the drill, alongside sport-specific practice, not as a replacement for it.

What score to expect

Your average only counts Go reps, so the grade doesn't show errors. Watch your resets too. A fast average built on frequent STOP mistakes means you're guessing, not reacting.

How to get better

• Look, then go.Read the cue as one quick check rather than reacting to “something appeared”. It costs a few milliseconds and saves resets.
• Start on Easy and keep errors near zero, then move up. Hard's 40 % STOP rate breaks the habit of always tapping.
• Hold all the way through.A STOP isn't over until the cue clears. Relaxing early leads to late twitches.

For broader training ideas, see how to improve your reaction time. For the fight-specific side of not biting on feints, see reaction time in boxing.

Sources

1. Donders, F. C. (1868/1969). On the speed of mental processes. Translated by W. G. Koster, Acta Psychologica, 30, 412–431.
2. Roelofs, A. (2018). One hundred fifty years after Donders: Insights from unpublished data, a replication, and modeling of his reaction times. Acta Psychologica, 191, 228–233.
3. Hasegawa, K. et al. (2021). Do shorter inter-stimulus intervals in the go/no-go task enable better assessment of response inhibition? Scandinavian Journal of Psychology, 62(2).
4. van Campen, A. D. et al. (2013). Paired-pulse transcranial magnetic stimulation reveals probability-dependent changes in functional connectivity between right inferior frontal cortex and primary motor cortex during go/no-go performance. Frontiers in Human Neuroscience, 7, 736.
5. Young, M. E., Sutherland, S. C. & McCoy, A. W. (2018). Optimal go/no-go ratios to maximize false alarms. Behavior Research Methods, 50, 1020–1029.
6. Nakamoto, H. & Mori, S. (2008). Sport-specific decision-making in a go/nogo reaction task. Perceptual and Motor Skills, 106(1), 163–170.
7. Muraskin, J., Sherwin, J. & Sajda, P. (2015). Knowing when not to swing: EEG evidence that enhanced perception–action coupling underlies baseball batter expertise. NeuroImage, 123, 1–10.
8. Albaladejo-García, C. et al. (2023). The role of inhibitory control in sport performance: Systematic review and meta-analysis in stop-signal paradigm. Neuroscience & Biobehavioral Reviews, 147, 105108.
9. Enge, S. et al. (2014). No evidence for true training and transfer effects after inhibitory control training in young healthy adults. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40(4), 987–1001.