Response Inhibition
Two Lamps Game
Also called: response inhibition game
Measure how well children aged 8 to 12 and adults hold back an action that has already started. A lamp lights on one side, the player presses its button, and sometimes the light changes.
7-9 minTimed: desktop or laptop with a mouse or trackpad
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 897
- Evidence grade
- B
- Registered clinical studies
- 321
Figures for the Stop-Signal Task, the published test this game is compared with.
How these are countedWhat it measures
A game for research on stopping an action that has already started, in children aged 8 to 12 and in adults. A lamp lights on the left or the right, and the player presses the button on its side. On some trials the light changes, and then the player tries not to press. The main score is the time the player needs to stop. It is a raw research measure with no percentile.
Constructs
- Response inhibition
- Stopping an action
- Action control
Research uses
- Research on stopping an action already started in school-age children
How the paradigm works
The Two Lamps Game is a game for research on stopping an action that has already started, in children aged 8 to 12 and in adults. A lamp stands on each side of the screen. Both lamps are dark for a moment, then one lights up, and the player presses the button on its side as soon as it lights, or presses F or J on a computer. On some trials the light changes to a dark square with a white cross shortly after it came on, and then the player tries not to press. The change comes after a stop delay that starts at 250 ms and moves in steps of 50 ms, 50 ms later after a stop and 50 ms sooner after a failed one, between 100 and 900 ms. The player has 1,500 ms after the light comes on to answer. This is the stop-signal paradigm, and the design follows the consensus guide (Verbruggen and colleagues 2019, eLife): at least 50 stop trials, a staircase of 50 ms steps, a fixed window for answers and instructions not to wait for the light to change. The lamps, the visual change, the side pads, the 100 ms floor on the delay (the go light is on screen at least 99 ms), the three dark periods, the four blocks, the forms and the words are our own.
More about the paradigm
The paper counts and grades on this page are those of the Stop-Signal Task, the published test the game is compared with, not the game's own. A story stop-signal game matched the plain task's estimate in adults (Friehs and colleagues 2020), and a game version for ages 8 to 12 has been described (Gallagher and colleagues 2023); this game has no data of its own yet. Before each lamp lights, both lamps stay dark for 500, 750 or 1,000 ms, from the form's table, so the onset cannot be timed. A practice of 8 go trials comes first, then one of 8 trials with 2 stops, each with the right button glowing; the second runs once more when fewer than 6 of the 8 are right. The game then runs 4 blocks of 50 trials, 50 of the 200 with a stop signal, with the same short reminder between blocks every time: press as soon as a lamp lights, and do not wait to see if it changes. The game shows no speed figure and no score. Each session takes one of 2 forms, the second the first with every side swapped. Each participant gets the forms in an order drawn once for them, and each session takes the next form. A form's scores are pooled with the other's only after first sessions show the two are equally hard.
Scores
- Stop-Signal Reaction Timems
- The integration method with go omissions replaced: the go time at the point given by the share of stop trials answered, less the mean planned stop delay. Reported whenever it can be computed, and valid only when validSsrt is 1 (the guide's checks hold and at least 40 stop trials are kept). Empty when the stop delays do not follow the staircase, when no go answer or no stop trial is kept, or when every kept stop trial was answered or none was.
- Lower means a faster stop; use only the sessions with validSsrt 1
- Mean Stop Delayms
- The mean planned stop delay of the kept stop trials.
- Informational: it shows where the staircase settled
- Share of Stop Trials Answered%
- The share of kept stop trials with an answer, one answered before the light changed included.
- Near .5 when the staircase has settled; the guide advises no estimate outside .25 to .75, so validSsrt is 0 there
- Median Go Timems
- Median time from the go light to a right answer on a go trial.
- Informational
- Go Omission Rate%
- Share of kept go trials with no answer within 1,500 ms.
- Secondary: a high rate means the lamps were not answered
- Go Side Error Rate%
- Share of kept go trials answered on the wrong side.
- Secondary
- Failed Stop Median Timems
- Median time of the answers on stop trials, which came before the stop finished.
- Informational: under the race model it is faster than the go time
- Go Time Changems
- Median time of the right go answers in the fourth block less that in the first.
- Informational: a large rise means the player may have started to wait for the light to change
- Valid Stop-Signal Estimatecount
- 1 when the estimate meets the guide's checks and at least 40 stop trials are kept, 0 otherwise. Empty for a run that holds two forms or no rows.
- Quality check: use sessions with 1 only
- Stop Trials Scoredcount
- Stop trials kept in the stop scores, out of 50.
- Quality check
- Valid Trialscount
- Test trials kept, go and stop, out of 200.
- Quality check
- Share of Key Answers%
- Share of the kept answers given with the F or J key.
- Informational: the touch-versus-keyboard comparison reads it
- Share of Touch Answers%
- Share of the kept answers given by a touch of a pad.
- Informational: the touch-versus-keyboard comparison reads it
Use and evidence
897 papers from 2021 to 2025 name this task in their title or abstract, about the same as from 2016 to 2020. That ranks it 30 of 632 test families in our census. These counts belong to the Stop-Signal Task, the published test this game is compared with. This game is a new measure, so its own reliability and norms come from use. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 200 |
| 2024 | 181 |
| 2023 | 170 |
| 2022 | 159 |
| 2021 | 187 |
| 2020 | 202 |
| 2019 | 174 |
| 2018 | 192 |
| 2017 | 166 |
| 2016 | 166 |
| 2015 | 176 |
| 2014 | 152 |
| 2013 | 140 |
| 2012 | 127 |
| 2011 | 128 |
| 2010 | 100 |
| 2009 | 60 |
| 2008 | 70 |
| 2007 | 56 |
| 2006 | 39 |
| 2005 | 31 |
| 2004 | 28 |
| 2003 | 23 |
| 2002 | 15 |
| 2001 | 13 |
| 2000 | 11 |
Use grade B
300 to 999 papers from 2021 to 2025.
Evidence grade B
At least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers.
- Papers on test-retest reliability
- 23
- Papers on reliability or validity
- 153
- Registered clinical studies using it as an outcome
- 321
Measured on ConductCognition
Pooled across the studies that feed our platform norms, by device group. A figure needs 50 or more people.
No pooled figure yet.
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on the Stop-Signal Task, the published test this game is compared with, not the game's own.
Source: ConductCognition test census 0.44.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Lamps
- One lamp stands on each side of the screen, drawn in code in three states: dark, the go light (a round yellow glow) and the stop light (a dark square with a white cross). The go and stop lights differ in outline and in lightness, and every pair of the bulb, the two lights and the tile behind them is at least CIEDE2000 20 apart in normal and in simulated color-deficient vision. The game look adds a soft background and a soft shadow under the lamps and the pads; the plain look shows them on white. Both looks draw every lamp the same way.
- Pads
- The player answers on one of two pads in the bottom corners of the screen, each at least 120 by 96 px, or presses F for the left or J for the right on a computer. A lamp carries no answer target, so a tap on the lit lamp answers nothing. The layout is worked out from the window before each trial, and a turned screen refits before the next one.
- Trial
- Both lamps stay dark for 500, 750 or 1,000 ms from the form's table, then one lights up and answers open. On a stop trial the light changes to the stop light on the first frame at or after the stop delay, less 1 ms. Answers stay open until 1,500 ms after the light came on; the lamp goes dark at the answer or then, and a 500 ms gap follows.
- Stop delay
- The stop delay starts at 250 ms and moves in steps of 50 ms between 100 and 900 ms: 50 ms later after a stop and 50 ms sooner after a failed one. Practice and test each keep their own. A stop trial answered before its delay never shows the stop light and counts as a failed stop at that delay.
- Blocks
- A practice of 8 go trials, then one of 8 trials with 2 stops, both with the right button glowing; the second runs once more when fewer than 6 of the 8 are right. Then 4 blocks of 50 trials, 50 of the 200 with a stop signal, with the same short reminder between blocks every time: press as soon as a lamp lights, and do not wait to see if it changes. The game shows no speed figure and no score.
- Forms
- 2 alternate forms, the second the first with every side swapped, with the same trial kinds, dark periods and stop trials. Each participant gets the forms in an order drawn once for them, and each session takes the next form.
- Scores
- The stop-signal reaction time by the integration method with go omissions replaced (the main score), the mean stop delay, the share of stop trials answered, the median go time, the share of go trials missed and answered on the wrong side, the median time of failed stops, the change in median go time from the first block to the fourth, whether the estimate meets the guide's checks, the stop trials scored and the shares of answers given by key and by touch. A trial is left out when a second finger or a palm touched the screen, the tab was hidden or the window changed size or turned. Each stop delay is checked against the staircase the game ran.
- Devices
- Desktops and laptops at release. Tablets open once two filmed tablets pass the device timing check, and phones do not run the game. Use one device class at every visit.
The design follows the consensus guide of Verbruggen and colleagues (2019); the lamps, the visual change, the side pads, the dark periods, the forms, the words and the looks are our own. A story stop-signal game matched the plain task's estimate in adults (Friehs and colleagues 2020), and a game version for ages 8 to 12 has been described (Gallagher and colleagues 2023); neither is this game, and the figures published in those studies are not its own. This is a new measure: its reliability and norms come from use, and it computes no percentiles. A practice visit before the baseline is advised, and for a retest 7 to 28 days later, add a second visit to the same study.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: go trials of two_lamps_game, two sides (chance 50%), with the side and the dark period taken from the form's table. A trial with a second contact, a hidden tab or a resized window is left out. Below chance invalidates and not above chance is flagged. One side on 90% of go answers invalidates when accuracy is not above chance. Missed go responses and responses under 100 ms follow the timed-test rule. A stop delay that does not follow the staircase invalidates. Stop data that fail the race-model checks, or fewer than 40 kept stop trials, are flagged and the percentile is withheld. A median go time in the last block more than 150 ms above the first block's is flagged.
Why these rules
Accuracy significantly below chance (one-sided exact test, p < .05) cannot come from guessing; it means the response mapping was reversed or the task was not followed (Slick, Sherman and Iverson 1999 treat below-chance forced-choice performance as definite response bias). Chance-level accuracy can be a real result in impaired participants, so it is only flagged. One option on at least 90% of answered trials, when the correct options are balanced and accuracy is not above chance, means the answers did not depend on the items (the task counterpart of long-string responding in surveys; Curran 2016). Missed responses and responses under 100 ms keep the platform's timed-test rule: more than max(3, 25%) missed responses or more than max(2, 20%) responses under 100 ms invalidate, and any is flagged. A response under 100 ms cannot be a reaction to the stimulus. The consensus guide (Verbruggen et al. 2019, recommendations 7 and 9) advises against estimating SSRT when p(respond|signal) is outside 0.25 to 0.75 or when failed-stop RT is longer than go RT, and asks for at least 50 stop trials; the scorer's validSsrt checks these and a minimum of 40 kept stop trials, a ConductScience choice. The guide also advises feedback on go speed after each block, which the games program bars (rule 17), so the game shows one fixed reminder not to wait and flags a rise of more than 150 ms in the median go time, a ConductScience choice. The staircase is replayed from the rows, because a stop delay that does not follow it cannot come from the game.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Look | Game | Game, Plain |
Every plan can change the look, which never changes how a result is scored.
Need a setting this test does not have?
Devices and languages
- Timed test: participants need a desktop or laptop with a mouse or trackpad. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- No percentile. Scores and quality flags only. How norms work
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Two Lamps Game in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Verbruggen and colleagues, 2019), with the standard settings (look: Game). The main score was stop-signal reaction time (the integration method with go omissions replaced: the go time at the point given by the share of stop trials answered, less the mean planned stop delay).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Verbruggen, F., Aron, A. R., Band, G. P., Beste, C., Bissett, P. G., Brockett, A. T., et al. (2019). A consensus guide to capturing the ability to inhibit actions and impulsive behaviors in the stop-signal task. eLife, 8, e46323. https://doi.org/10.7554/eLife.46323The guide the design follows: at least 50 stop trials, a staircase of 50 ms steps, the integration method with go omissions replaced, and instructions not to wait for the light to change (full text read).
- Friehs, M. A., Dechant, M., Vedress, S., Frings, C., & Mandryk, R. L. (2020). Effective gamification of the stop-signal task: Two controlled laboratory experiments. JMIR Serious Games, 8(3), e17810. https://doi.org/10.2196/17810A story stop-signal game gave an estimate that matched the plain task's in adults (full text read).
- Gallagher, R., Kessler, K., Bramham, J., Dechant, M., & Friehs, M. A. (2023). A proof-of-concept study exploring the effects of impulsivity on a gamified version of the stop-signal task in children. Frontiers in Psychology, 14, 1068229. https://doi.org/10.3389/fpsyg.2023.1068229A game version of the stop-signal task for children aged 8 to 12; no reliability was reported (passages read).
Source papers
Questions
How long does the Two Lamps Game take?
About 7 to 9 minutes, as listed in the test library.
What device does a participant need?
Timed test: participants need a desktop or laptop with a mouse or trackpad. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Stop-Signal Reaction Time, Mean Stop Delay, Share of Stop Trials Answered, Median Go Time, Go Omission Rate, Go Side Error Rate, Failed Stop Median Time, Go Time Change, Valid Stop-Signal Estimate, Stop Trials Scored, Valid Trials, Share of Key Answers and Share of Touch Answers. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Stop-Signal Task used in research?
These counts belong to the Stop-Signal Task, the published test this game is compared with. This game is a new measure, so its own reliability and norms come from use. 897 papers from 2021 to 2025 name the task in their title or abstract, about the same as from 2016 to 2020. That ranks it 30 of 632 test families in the ConductCognition test census (use grade B). Source: ConductCognition test census 0.44.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Stop-Signal Task have?
These counts belong to the Stop-Signal Task, the published test this game is compared with. This game is a new measure, so its own reliability and norms come from use. In the published research, 23 papers report the task's test-retest reliability and 321 registered clinical studies use it as an outcome (evidence grade B in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Look (Game, Plain). Every plan can change the look, which never changes how a result is scored.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.
Related tests