Reward and Decision Making
Reward Perseveration Task
Also called: door-opening task, response perseveration task, declining-reward task
Measure how long people keep seeking points as gains become less frequent. Participants reveal a capsule or stop and keep their total; scores describe persistence and the points gained or lost.
2-7 minUntimed
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
- 2
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures how long people continue seeking points as gains become less frequent. Participants reveal one capsule at a time or stop and keep their current total. The main score is the number revealed before stopping, up to 90.
Constructs
- Reward perseveration
- Response to declining rewards
- Voluntary stopping
- Learning from outcomes
Research uses
- Research on reward sensitivity and impulsivity
- Studies of stopping decisions after gains and losses
- Within-study comparisons of reward persistence
How the paradigm works
The Reward Perseveration Task measures when people stop seeking rewards as the returns become less favorable. It belongs to the door-opening family of response-perseveration tasks introduced by Daugherty and Quay (1991). The main outcome is how many opportunities the participant takes before stopping. This captures persistence in pursuit of points; it is different from response inhibition on a stop-signal task.
More about the paradigm
Participants begin with 100 points and reveal one capsule at a time. Each reveal adds or removes 10 points. They can stop and keep their total at any choice, including before the first reveal. Stopping ends the whole task. There are at most 90 reveals in nine blocks of ten. The blocks contain nine gains, then eight, down to one, with their order shuffled within every block from a saved seed. Each outcome remains visible for 3,000 ms. Two practice reveals, one gain and one loss, use separate points; a question checks that stopping ends the task. The point unit is a setting from 1 to 100, with a starting balance ten times that unit. Choices have no response deadline and work by click or tap.
The primary score is capsules revealed. Additional scores describe persistence after the first 50 opportunities, gains and losses, final and highest points, points below the peak, and whether a voluntary stop followed a loss. The scorer reconstructs outcomes from the saved seed rather than trusting stored point totals. A short completed run is a valid stopping decision. Missing endings and unreadable rows are flagged for review.
Scores
- Capsules Revealedcount
- Reveals before the voluntary stop, or 90 when every capsule was opened. Zero is a valid stopping choice.
- Higher means continuing for longer as gains become less frequent
- Reveals After the Even Blockcount
- Reveals after opportunity 50, when each block contains more losses than gains.
- Higher means more persistence during unfavorable blocks
- Final Pointspoints
- Starting balance plus gains minus losses, reconstructed from the saved seed and choices.
- Informational
- Gainscount
- Reveals that added points.
- Informational
- Lossescount
- Reveals that removed points.
- Informational
- Highest Point Totalpoints
- Highest balance reached, including the starting balance.
- Informational
- Points Below Peakpoints
- Highest balance minus the final balance.
- Informational
- Stopped After a Loss0/1
- 1 if the voluntary stop immediately followed a loss; 0 after a gain; null without a preceding reveal or at the automatic limit.
- Informational
- Median Choice Timems
- Median reveal/stop choice time at or above 100 ms.
- Informational
- Choices Under 100 mscount
- Accepted choices within 100 ms of the next screen; participants can prepare a choice during feedback.
- Quality check; not excluded from the preference score
- Stopping Check Correct0/1
- 1 for understanding that stopping ends the whole task; 0 for the other answer.
- Quality check
- Complete Sequence0/1
- 1 for an intact sequence ending in a stop choice or all 90 reveals; otherwise 0.
- Quality check
- Recorded Decisionscount
- Readable decisions, including the final stop when chosen.
- Informational
- Unreadable Decisionscount
- Decisions that could not be reconstructed in sequence from the saved settings.
- Quality check
Use and evidence
2 papers from 2021 to 2025 name this task in their title or abstract. That ranks it 485 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 1 |
| 2024 | 0 |
| 2023 | 0 |
| 2022 | 0 |
| 2021 | 1 |
| 2020 | 0 |
| 2019 | 2 |
| 2018 | 1 |
| 2017 | 0 |
| 2016 | 1 |
| 2015 | 1 |
| 2014 | 6 |
| 2013 | 5 |
| 2012 | 1 |
| 2011 | 1 |
| 2010 | 0 |
| 2009 | 2 |
| 2008 | 1 |
| 2007 | 4 |
| 2006 | 3 |
| 2005 | 3 |
| 2004 | 2 |
| 2003 | 1 |
| 2002 | 2 |
| 2001 | 1 |
| 2000 | 1 |
Use grade E
Fewer than 30 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 1
- Papers on reliability or validity
- 1
- Registered clinical studies using it as an outcome
- 0
Note on the counts: Original capsule version; not published norms..
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 this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.22.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Sequence
- Up to 90 reveals in nine blocks of ten. Gain counts are 9, 8, 7, 6, 5, 4, 3, 2 and 1 per block; every block is shuffled from the saved seed.
- Points
- 10 points gained or lost per reveal by default; a setting from 1 to 100. Starting balance is ten times this amount: 100 points by default.
- Stopping
- Stop and keep points ends the entire sequence. Stopping before the first reveal is a complete zero-open result.
- Practice
- Two practice reveals, one gain and one loss, followed by a stopping-rule question with feedback. Practice points do not count.
- Timing
- No response deadline. Outcomes remain visible for 3,000 ms before the next choice. About 2 to 7 minutes, depending on when the participant stops.
- Scores
- Capsules revealed is primary. Also reports reveals after opportunity 50, gains, losses, final and highest points, points below the peak and stopping after a loss. Preference scores have no universal better direction; percentiles are off.
- Devices
- Click or tap on desktop, tablet or phone, with a minimum touch screen of 360 by 640.
An original capsule version of the door-opening method. Unlike Munsch and colleagues, every block is shuffled; there is one sequence for points, without food/nonfood rewards. The task uses no source graphics or software.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scores reveal/stop choices from the latest attempt. Stopping at any point, including before the first reveal, is a complete result. No choice is marked wrong and repeated revealing is accepted. A missing ending, unreadable rows or a wrong practice check is flagged for review.
Why these rules
The outcome is how long a person keeps seeking points as gains become less frequent. Early stopping and repeated revealing are the behavior being measured. Participants can prepare their next choice during feedback, so choice speed does not invalidate a preference.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Points gained or lost | 10 | 1 to 100 |
| Payment line | Hypothetical points (standard) | Hypothetical points (standard), The study pays for points |
Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.
Need a setting this test does not have?
Devices and languages
- Untimed test: runs on desktops, laptops, tablets and phones. 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 Reward Perseveration Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Daugherty & Quay, 1991; Munsch et al., 2019), with the standard settings (points gained or lost: 10; payment line: Hypothetical points). The main score was capsules revealed (reveals before the voluntary stop, or 90 when every capsule was opened).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Daugherty, T. K., & Quay, H. C. (1991). Response perseveration and delayed responding in childhood behavior disorders. Journal of Child Psychology and Psychiatry, 32(3), 453–461. https://doi.org/10.1111/j.1469-7610.1991.tb00323.xHistorical response-perseveration method; abstract consulted.
- Munsch, S., Dremmel, D., Wilhelm, P., Baierlé, S., Fischer, S., & Hilbert, A. (2019). To eat or not to eat: Reward delay impulsivity in children with loss of control eating, attention deficit / hyperactivity disorder, a double diagnosis, and healthy children. PLOS ONE, 14(9), e0221814. https://doi.org/10.1371/journal.pone.0221814Full task methods: declining gain frequency and voluntary stopping.
Source papers
- Munsch S, Dremmel D, Wilhelm P, Baierlé S, Fischer S, Hilbert A (2019). PLOS ONE, 14(9):e0221814.N 100, 8–13 years
Questions
How long does the Reward Perseveration Task take?
About 2 to 7 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops, tablets and phones. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Capsules Revealed, Reveals After the Even Block, Final Points, Gains, Losses, Highest Point Total, Points Below Peak, Stopped After a Loss, Median Choice Time, Choices Under 100 ms, Stopping Check Correct, Complete Sequence, Recorded Decisions and Unreadable Decisions. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Reward Perseveration Task used in research?
2 papers from 2021 to 2025 name it in their title or abstract. That ranks it 485 of 495 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.22.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Reward Perseveration Task have?
In the published research, 1 paper reports its test-retest reliability and 0 registered clinical studies use it as an outcome (evidence grade D in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Points gained or lost (1 to 100) and payment line (Hypothetical points (standard), The study pays for points). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
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.