Pulse

What you see

A Pulse trial is one stream of 16 to 32 items. You press Start (or Space), a 3-2-1 countdown runs, and then the shapes arrive one at a time in the centre of the panel, under the prompt "Respond to the target and nothing else." An Item pill, such as Item 5 / 20, shows how far through the stream you are, and when the last item has passed the line under the board reads "All 20 shapes shown."

There are four things you can see:

• The target: a filled dot inside a circle. Press.
• The grey square: a grey rounded square. Hold back.
• The grey ring: a grey circle with a hole in the middle. It is round, like the target, which makes it harder to reject. It appears from level 3. Hold back.
• The stop signal: from level 6, some targets are followed a fraction of a second later by a red outline around them. The item starts out looking exactly like a normal target, so your press is already on its way when the stop signal arrives. Hold back.

You respond with the Space bar or the large Press button. Only your first press on each item is recorded, with its reaction time measured from the moment the shape appeared.

The timing of one item

Each item runs the same way:

1. A fixation cross for 300 ms. Presses here are ignored.
2. The shape for the exposure time: 900 ms at level 1, down to 400 ms at level 10.
3. A blank with a small dot until the deadline. Your press still counts here.
4. The next item's cross begins 150 ms after the deadline.

The first three items of every stream are always targets, so you settle into pressing before any no-go appears. After that, the order is random but never has more than four of the same kind in a row (a stop item counts as a target here, since it starts out looking like one). Stop items never fall in the first three, and any two of them have at least two other items between them.

Because every item in a stream takes the same time, the stream has a steady beat. That beat is part of the challenge: it builds a rhythm of pressing that the no-go shapes have to interrupt.

What changes as levels rise

Two shifts deserve attention. At level 6 the pace briefly relaxes because stop items arrive; the game gives you a little more time while you learn the new rule. And the stop signal arrives earlier as you climb, from 300 ms after the target at level 6 to 180 ms at level 10. An earlier stop signal is easier to obey, but the deadline also tightens, so there is less room to wait for it.

What it measures

Pulse measures response inhibition: stopping an action you are ready to make. In our attention map it sits on the Inhibition facet with Keypress, and it carries 22 per cent of the Attention dimension, the joint largest share with Arrows.

It combines two classic laboratory tasks.

The go/no-go task asks you to respond to one class of stimulus and withhold for another. The key design feature is that go items are common and no-go items are rare, which builds a habit of responding that the no-go item has to override. That rarity is deliberate. Jan Wessel's review of go/no-go designs found that inhibition-related brain activity was about 75 per cent weaker in slow versions with equal numbers of go and no-go items than in fast versions with rare no-gos. Young and colleagues found that people made the most false presses with go-to-no-go ratios between 2:1 and 4:1 and short gaps between items. Pulse's pure go/no-go levels run between roughly 4:1 (level 1) and 2.4:1 (level 5), inside that range.

The stop-signal task is harder. Every item starts as a go item, and on a minority of trials a stop signal appears after a short delay. You have already begun to respond, so you must cancel an action in flight, not merely decline to start it. The model behind it comes from Gordon Logan and W. B. Cowan: going and stopping are two independent processes that race each other, and whichever finishes first decides the outcome. A 2019 consensus guide written by the main stop-signal research groups recommends about 25 per cent stop trials and a delay that adjusts to each person, so that a stop-signal reaction time can be estimated.

How long does the stop process take? In a study of 165 healthy adults, the average stop-signal reaction time was between 146 and 172 ms, depending on the calculation method. So when the stop signal appears 220 ms after the target, the stop process typically finishes somewhere near 370 to 390 ms after the target. A press that lands before that has already escaped.

These are different kinds of restraint. A no-go shape tells you "this is not a target" from the moment it appears: it is action restraint. A stop signal tells you "the action you started must be cancelled": it is action cancellation. Pulse gives you both in one stream.

How it is scored here

Every item in the stream ends up in one of these buckets:

"In time" means at least 150 ms after the shape appeared and no later than the deadline. A press under 150 ms is faster than anyone can see and respond, so it is treated as an anticipation: it never counts as a hit. On a hold item, any press at all counts against you, including one in the short gap after the deadline.

The stream's score is built from two rates:

• Go accuracy: hits divided by targets.
• Restraint: correct holds and correct stops divided by all the items where you should have held back.

The score is go accuracy + restraint − 1, never less than zero. Researchers call this Youden's J. A perfect stream (every target hit, nothing pressed wrongly, no anticipations) scores full marks.

This formula has two consequences worth understanding.

Pressing on everything and pressing on nothing both score zero. If you press every item, go accuracy is 100 per cent and restraint is 0 per cent: the score is zero. If you never press, it is the other way round: also zero. Only telling the two apart earns credit.

Each mistake costs according to how rare its kind is. At level 1 there are 13 targets and 3 no-go squares. One missed target costs 1 ÷ 13, about 8 points. One false press costs 1 ÷ 3, about 33 points. That is why a wrong press costs more than a missed one at low levels. By level 10 the stream has 16 targets and 16 hold items, so the two mistakes cost the same.

Your reaction time on hits is recorded and reported (your median go time, and accuracy per second of reaction time), along with signal clarity and a note on your answering style. But the score only asks whether each press landed inside the window. Pressing at 300 ms rather than 500 ms earns nothing extra.

The difficulty of each stream is measured from its own settings: the deadline weighs most, then how rare the hold items are (rarer no-gos build a stronger urge to press), then how similar the distractors are to the target. Your game score is built from difficulty earned per stream, and 50 on our scale means you are performing at what the difficulty you work at predicts.

How the level moves

In adaptive practice, after each stream:

• A perfect stream moves you up one level.
• A stream with some discrimination but at least one mistake holds your level.
• A stream that scores zero moves you down one level.

Where you will meet it

Go/no-go and stop-signal tasks are among the most used tools in inhibition research, and versions of the idea appear in hiring assessments. None of these is our game, and we have no link with the vendors named; they are the same family of task.

• pymetrics (now Harver) lists attention and focus among its measured traits and says its games require "a series of fast clicks/taps on the screen". Candidates report a game called Stop in which shapes flash and you respond only to some of them, and a game called Keypress. Those game descriptions come from prep sites, not from pymetrics.
• Aon lists a two-minute test called scales e3+, "Ability to Concentrate", which it says "measures your aptitude to concentrate over time and act quickly when faced with certain signals". That is a sustained-attention cousin of the go/no-go stream.
• Mercer Mettl offers a Cognitive Speed Test of 60 questions in 18 minutes, which it says assesses "general intelligence, processing speed and attention".

The method

1. Learn the target as a shape, not a colour

   The target is the only shape with a solid dot in its centre. The ring distractor is round too, but its centre is empty. Train yourself to look for the dot, which is the one feature only the target has.

2. Hover, do not rest, on the key

   Keep one finger lightly over Space. A finger pressing down slightly between items is already halfway to a false press when a square arrives.

3. Press on seeing the dot, not on the beat

   The stream is evenly paced, so you will feel the rhythm. Use the rhythm to be ready, but let the dot trigger the press. If you ever find you pressed before you knew what the shape was, slow your next few presses slightly.

4. From level 6, keep a small buffer for the stop signal

   The stop signal arrives 180 to 300 ms after a target. You earn nothing for pressing very early, so do not aim for your fastest possible press. A calm press, well inside the deadline, leaves time to see the stop signal on the few items that have one.

5. Let misses go

   If you miss a target, there is nothing to recover. Do not press the next item harder or earlier to make up for it. The next item is a fresh decision.

Worked example 1: scoring a short stream

Here is a 12-item stretch at level 3, with what one candidate did underneath each item. Work out the go accuracy, the restraint and the score.

Show the solution

• Targets: items 1, 2, 3, 5, 6, 8, 9, 11 and 12, which is 9 targets. The candidate pressed on 8 of them and missed item 6. Go accuracy = 8 ÷ 9, about 0.89.
• Hold items: 4 (square), 7 (ring) and 10 (square). The candidate held on 4 and 10 but pressed on the ring at 7. Restraint = 2 ÷ 3, about 0.67.
• Score: 0.89 + 0.67 − 1 = 0.56, or 56 per cent.

Now compare the two mistakes. With only the miss at item 6, the score would be 0.89 + 1 − 1 = 0.89. With only the false press on item 7, it would be 1 + 0.67 − 1 = 0.67. The single false press cost three times as much as the single miss, because there were three times as many targets as hold items. Notice too which shape caught the candidate: the ring distractor, the one that looks round like the target.

Worked example 2: a stop item

At level 8 the stop signal appears 220 ms after a stop item's target, and the deadline is 950 ms. A candidate has been pressing at about 330 ms. A stop item arrives. What happens, and what would you change?

Show the solution

The target appears and the candidate's press is already on its way. The stop signal appears at 220 ms. For a typical adult, cancelling takes roughly another 150 to 170 ms, so the stop process would finish somewhere around 370 to 390 ms, after the press at about 330 ms. The press escapes: a stop failure.

The press also earned nothing by being fast. A hit at 330 ms scores exactly the same as a hit at 550 ms, and 550 ms is still far inside the 950 ms deadline. Settling your presses a little later, around the middle of the window, gives the stop signal time to register on stop items without costing you anything on normal targets. The caution: waiting too long risks the deadline, so the aim is a steady, unhurried press, not a deliberate pause.

Tips and traps

Levelling up

Levels 1 to 2 teach the basic habit: press the dot, hold for squares.

Levels 3 to 5 add the ring distractor and speed up the stream. The skill is fine discrimination under time pressure: dot versus hole.

Levels 6 to 10 add stop items. The skill is keeping a press controllable after it has started. As the deadline tightens from 1,100 ms to 850 ms, the room for a calm press shrinks, and by level 10 half the stream is hold items.

1. Week 1
   • Three sessions of 5 or 6 streams, adaptive on
   • Goal: zero false presses, even if you miss a target or two
   • After each session, check whether your false presses were on squares or rings
2. Week 2
   • Four sessions; now aim for zero misses as well
   • Practise hovering over Space rather than resting on it
   • Pair each Pulse session with a short Keypress session for a second kind of stopping
3. Week 3
   • Four sessions, pushing into the stop-item levels
   • On stop levels, keep your presses unhurried and watch your stop failures fall
   • Compare your answering style with week 1: you want it close to balanced

Check yourself

5 practice questions

1. At level 1 a stream has 13 targets and 3 squares. Which single mistake lowers your score more?
2. A candidate presses on every single item in a stream. What do they score?
3. You press on a target 120 ms after it appears. How is that press counted?

and 2 more

Create a free account to answerSign in

Sources

1. Logan, G. D., & Cowan, W. B. (1984). On the ability to inhibit thought and action: A theory of an act of control. Psychological Review, 91(3), 295–327. doi.org/10.1037/0033-295X.91.3.295
2. Verbruggen, F., et al. (2019). A consensus guide to capturing the ability to inhibit actions and impulsive behaviors in the stop-signal task. eLife, 8, e46323. doi.org/10.7554/eLife.46323
3. Congdon, E., Mumford, J. A., Cohen, J. R., Galvan, A., Canli, T., & Poldrack, R. A. (2012). Measurement and reliability of response inhibition. Frontiers in Psychology, 3, 37. doi.org/10.3389/fpsyg.2012.00037
4. Wessel, J. R. (2018). Prepotent motor activity and inhibitory control demands in different variants of the go/no-go paradigm. Psychophysiology, 55(3), e12871. doi.org/10.1111/psyp.12871
5. Young, M. E., Sutherland, S. C., & McCoy, A. W. (2018). Optimal go/no-go ratios to maximize false alarms. Behavior Research Methods, 50(3), 1020–1029. doi.org/10.3758/s13428-017-0923-5
6. Diamond, A. (2013). Executive functions. Annual Review of Psychology, 64, 135–168. doi.org/10.1146/annurev-psych-113011-143750
7. pymetrics support centre. Article on accommodations. pymetrics.zendesk.com
8. Aon. Prepare for your online assessment. aon.com
9. Mercer Mettl. Cognitive Speed Test. mettl.com/en/test/cognitive-speed-test
10. IBPS. Detailed notification, CRP PO/MT-XVI (2026). ibps.in
11. Psychometric Success. Pymetrics games (candidate-reported descriptions). psychometric-success.com

Put it into practice: PulseSign in to try it