Social Cognition
Bribe or Refuse Task
Also called: bribery game, corruption game, bribery experiment, Abbink bribery game, Barr and Serra bribery game
Measure how the chance of a check changes hidden exchanges: participants send points to a simulated player, or accept or refuse the points it sends, when accepting costs a third player, scored by the share of rounds with points sent.
5-7 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
- 40
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures how often people send and accept a private transfer of points that pays the two of them and costs a third player, when each round states the chance of a check. The participant sends 0 to 20 points in 4 rounds, one at each chance from 0% to 50%, and accepts or refuses the other player's transfers in 8. Both other players are simulated, and the end screen says so.
Constructs
- Trading a private gain against a cost to others
- Deterrence by the chance of a check
- Reciprocity in a hidden exchange
- Moral decision making
Research uses
- Behavioral economics research on corruption and its deterrence
- Moral psychology research
- Studies of social preferences and incentives
- Group comparison within a study
How the paradigm works
The Bribe or Refuse Task measures how often people send and accept a private transfer that pays the two of them and costs someone else, and how that changes with the chance of a check. In the bribery game of Abbink, Irlenbusch and Renner (2002), a first player may send a transfer to a second, who accepts or rejects it and then chooses between two options; one favors the first player and, in one treatment, costs every other participant points, and a small chance of disqualification after each accepted transfer stands for the risk of discovery. Barr and Serra (2010) played a one-shot form in which a private citizen may offer a public official a bribe and other members of society lose points for each bribe accepted. This version folds the favor into the acceptance: an accepted transfer gives its sender 20 points more and costs a simulated third player 20 points. As in Abbink's neutral presentation, no screen names a bribe, an official or honesty; the players are "you", "the other player" and "the third player", with no country, party, institution or group.
More about the paradigm
Each player starts every round with 30 points. In part 1 the participant chooses how many points, 0 to 20, to send to the other player in 4 rounds, one at each stated chance of a check: 0%, 10%, 25% and 50%. In part 2 the other player sends 3 or 8 points in 8 rounds, both amounts at each chance, and the participant accepts or refuses. When a transfer is accepted, the sender has 50 points minus the transfer, the receiver 30 plus it and the third player 10; a refused transfer goes back. A check after an accepted transfer leaves the sender and the receiver 0 points. Each round's check is drawn from the session's seed at its stated chance, so the chance changes what a transfer is worth, the question Schulze and Frank (2003) asked of monitoring. The other player is a computer program that accepts n points with probability n/10, at most 0.9, and the standard instructions say that both other players are computer programs; a neutral wording leaves that line out, and the end screen says it in both. No outcome is shown until the end, when one round drawn from the seed sets the points. One check question with feedback comes first, and there is no time limit. The study decides what the points are worth, and the payment line says whether they are exchanged for money.
The main score is the share of part 1's rounds in which the participant sent points. The points sent at each chance and their drop from 0% to 50%, and the share of transfers accepted overall, at each chance and by size, with its drop from 0% to 50%, are also reported. Each chance has one sending round and two transfers to answer, so a score at one chance is read across a group, as the sources report it: in Barr and Serra's one-shot game with no risk of punishment, 79% of participants in the briber's role offered a bribe and 83% in the official's role accepted one (Tanner, Linder and Sohn, 2022). A transfer to a disclosed computer program may weigh less than one to a person, so compare people who saw the same instructions and payment line. The scores describe choices in a game played for points, never a person's honesty, and they do not diagnose.
Scores
- Offer Rate%
- Share of part 1's 4 rounds in which the participant sent at least 1 point.
- Read across a group; no better direction
- Mean Points Sentpoints
- Mean points sent over part 1's 4 rounds, from 0 to 20.
- Informational
- Points Sent at 0%points
- Points sent in the round with no chance of a check.
- Read across a group
- Points Sent at 10%points
- Points sent in the round with a 10% chance of a check.
- Read across a group
- Points Sent at 25%points
- Points sent in the round with a 25% chance of a check.
- Read across a group
- Points Sent at 50%points
- Points sent in the round with a 50% chance of a check.
- Read across a group
- Drop in Points Sentpoints
- Points sent at 0% minus points sent at 50%; above 0 when a likely check lowers the transfer.
- Read across a group
- Accept Rate%
- Share of part 2's 8 transfers that the participant accepted.
- Read across a group; no better direction
- Accept Rate at 0%%
- Share of the two transfers, of 3 and 8 points, accepted with no chance of a check.
- Read across a group
- Accept Rate at 10%%
- Share of the two transfers accepted with a 10% chance of a check.
- Read across a group
- Accept Rate at 25%%
- Share of the two transfers accepted with a 25% chance of a check.
- Read across a group
- Accept Rate at 50%%
- Share of the two transfers accepted with a 50% chance of a check.
- Read across a group
- Drop in Acceptance%
- Accept rate at 0% minus accept rate at 50%.
- Read across a group
- Accept Rate, 3-Point Transfers%
- Share of the 4 transfers of 3 points accepted, one at each chance.
- Informational
- Accept Rate, 8-Point Transfers%
- Share of the 4 transfers of 8 points accepted, one at each chance.
- Informational
- Points Wonpoints
- The participant's points in the round drawn to count.
- Informational
- Comprehension Question Correctflag
- 1 when the question on what a check does to an accepted transfer was answered correctly, 0 when not.
- 1 supports interpretation
- Median Decision Timems
- Median time from a round's screen to the choice, over choices at or after 100 ms.
- Informational
- Choices Under 100 mscount
- Choices made less than 100 ms after the round's screen appeared.
- 0 expected
- Roundscount
- Rounds played.
- Informational
Use and evidence
40 papers from 2021 to 2025 name this task in their title or abstract, down 34% compared with 2016 to 2020. That ranks it 506 of 711 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 8 |
| 2024 | 7 |
| 2023 | 7 |
| 2022 | 9 |
| 2021 | 9 |
| 2020 | 9 |
| 2019 | 18 |
| 2018 | 9 |
| 2017 | 12 |
| 2016 | 13 |
| 2015 | 13 |
| 2014 | 10 |
| 2013 | 5 |
| 2012 | 12 |
| 2011 | 5 |
| 2010 | 9 |
| 2009 | 6 |
| 2008 | 3 |
| 2007 | 3 |
| 2006 | 2 |
| 2005 | 1 |
| 2004 | 2 |
| 2003 | 0 |
| 2002 | 1 |
| 2001 | 0 |
| 2000 | 2 |
Use grade D
30 to 99 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 17
- Registered clinical studies using it as an outcome
- 0
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 this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.58.0, counted 16 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Players
- The participant, the other player and the third player, each starting every round with 30 points. The other player and the third player are computer programs that follow set rules. The standard instructions say so, a neutral wording leaves that line out, and the end screen says it in both.
- Part 1
- 4 rounds, one at each stated chance of a check (0%, 10%, 25% and 50%), in an order drawn from the session's seed. The participant sends 0 to 20 points on a bar that starts with no amount chosen. The other player accepts n points with probability n/10, at most 0.9.
- Part 2
- 8 rounds in an order drawn from the seed: at each chance the other player sends 3 points once and 8 points once, and the participant accepts or refuses. A table shows each player's points for both answers.
- Points
- An accepted transfer leaves its sender 50 points minus the transfer, the receiver 30 plus it and the third player 10; a refused one goes back. A check after an accepted transfer leaves the sender and the receiver 0 points. Each round's check is drawn from the seed at its stated chance.
- Counted round
- No outcome shows until the end, when one of the 12 rounds, drawn from the seed, sets the points. Points are hypothetical by default; a setting changes the instruction line to say the study pays.
- Check
- One question with feedback before part 1: the participant sends 5 points, the other player accepts, and there is a check.
- Scores
- The share of part 1's rounds with points sent (the main score), the points sent at each chance and their drop from 0% to 50%, and the share of transfers accepted overall, at each chance and by size.
No screen names a bribe, an official, corruption or honesty, and no country, party, institution or group appears, as in the neutral presentation of Abbink, Irlenbusch and Renner (2002). A transfer to a disclosed computer program may weigh less than one to a person, so compare people who saw the same instructions. The scores describe choices in a game played for points, never a person's honesty, and do not diagnose.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: the scored run's choices, the points sent in each of part 1's rounds and the answer to each of part 2's transfers. No choice is right or wrong, so accuracy is not judged against chance, and the same choice throughout is accepted. A run short of its planned rounds and a wrong check answer are flagged for review. Choices under 200 ms on at least half the rounds invalidate.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. Sending, accepting and refusing are choices in a game played for points, so the rule judges only how the choices were made, never what they say about the participant. An answer under 200 ms is faster than a simple visual reaction, so it cannot reflect the item. Following response-time effort screening for untimed tests (Wise and Kong 2005), the session is invalid only when at least half the answers are that fast.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Other players' wording | Computer programs (standard) | Computer programs (standard), Neutral: the other player and the third player |
| 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 Bribe or Refuse Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Abbink K, Irlenbusch B, Renner E, 2002; Barr A, Serra D, 2010), with the standard settings (other players' wording: Computer programs; payment line: Hypothetical points). The main score was offer rate (share of part 1's 4 rounds in which the participant sent at least 1 point).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Abbink, K., Irlenbusch, B., & Renner, E. (2002). An experimental bribery game. Journal of Law, Economics, and Organization, 18(2), 428-454. https://doi.org/10.1093/jleo/18.2.428The bribery game: a transfer accepted or rejected, a favor that can cost other participants, and a small chance that an accepted transfer is discovered.
- Barr, A., & Serra, D. (2010). Corruption and culture: An experimental analysis. Journal of Public Economics, 94(11-12), 862-869. https://doi.org/10.1016/j.jpubeco.2010.07.006A one-shot game in which each accepted bribe costs other members of society.
- Schulze, G. G., & Frank, B. (2003). Deterrence versus intrinsic motivation: Experimental evidence on the determinants of corruptibility. Economics of Governance, 4(2), 143-160. https://doi.org/10.1007/s101010200059A game with and without a chance of detection: monitoring deters, but it also weakens the intrinsic motivation not to take part.
- Tanner, C., Linder, S., & Sohn, M. (2022). Does moral commitment predict resistance to corruption? Experimental evidence from a bribery game. PLOS ONE, 17(1), e0262201. https://doi.org/10.1371/journal.pone.0262201Barr and Serra's game with offers of 1 to 20 points: 79% of participants in the briber's role offered a bribe and 83% in the official's role accepted one.
Source papers
- Tanner, Linder and Sohn (2022), Barr and Serra's one-shot game with no risk of punishment: 60 of 76 participants in the briber's role (79%) offered a bribe of 1 to 20 points, and 63 of 76 in the official's role (83%) accepted at least one amount.Paradigm source, Adults
Questions
How long does the Bribe or Refuse Task take?
About 5 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?
Offer Rate, Mean Points Sent, Points Sent at 0%, Points Sent at 10%, Points Sent at 25%, Points Sent at 50%, Drop in Points Sent, Accept Rate, Accept Rate at 0%, Accept Rate at 10%, Accept Rate at 25%, Accept Rate at 50%, Drop in Acceptance, Accept Rate, 3-Point Transfers, Accept Rate, 8-Point Transfers, Points Won, Comprehension Question Correct, Median Decision Time, Choices Under 100 ms and Rounds. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Bribe or Refuse Task used in research?
40 papers from 2021 to 2025 name it in their title or abstract, down 34% compared with 2016 to 2020. That ranks it 506 of 711 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.58.0, counted 16 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Bribe or Refuse Task have?
In the published research, 0 papers report 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?
Other players' wording (Computer programs (standard), Neutral: the other player and the third player) 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.