How the stop game works
Pymetrics’ 2017 US patent names go/no-go as the task behind the game it lists as Stop1. Go/no-go means pressing a button when one stimulus appears (go) and not pressing it when a different one appears (no-go). In the patent’s version you see either a red circle or a green circle, and you press the space bar for red and nothing for green.
Pymetrics does not publish the number of trials, the time you have to respond or the share of go and no-go trials in the live game.
What it measures
The patent describes go/no-go as a test of attention span and response control. Performance, especially on the inhibition trials where you must not respond, relates to the executive functions measured by the Wisconsin Card Sorting Task, the Stroop Color-Word Test and the Trail Making Test. Pymetrics found test-retest reliability of .82 on a key measure of its version.
How to approach it
There are two ways to get a trial wrong: pressing on a no-go trial, and missing a go trial. Fixing one by causing the other does not help, so respond to what is on screen each time rather than settling into a rhythm. Play on the web or on Pymetrics’ phone app, not a tablet: Pymetrics says the games are not available for tablets and warns that results may be affected if you use one.
How our practice version works
Our Stop practice runs 120 trials with 800 milliseconds each, about 70% of them go trials. A red circle means press the space bar, or tap the circle on a phone; an empty outline means do nothing. The trial count, time window and ratio are ours, and our no-go signal is an empty outline rather than the patent’s green circle. You are not told during the run whether a trial went your way, and the end screen reports your error rates and reaction times.