Social Cognition
Hidden Payoffs Task
Also called: moral wiggle room, willful ignorance, strategic ignorance, information avoidance, hidden information dictator game
Measure moral wiggle room: participants choose between two splits of points with another person whose points are either shown or hidden behind a free click.
1-3 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
- 44
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures moral wiggle room: participants choose between two splits of points with another person, whose points are either shown or hidden behind a free click. Scores are how often people look and how often they choose the equal split in each version.
Constructs
- Moral wiggle room
- Willful ignorance
- Information avoidance
- Social preferences
Research uses
- Behavioral economics of fairness
- Moral psychology research
- Between-group comparison of known and hidden versions
How the paradigm works
The Hidden Payoffs Task measures moral wiggle room, also called willful ignorance: people who choose fairly when they can see how a choice affects another person often choose selfishly when they can avoid knowing. In Dana, Weber and Kuang (2007), dictators chose between $6 for themselves and $1 for a receiver, or $5 each. When both payoffs were shown, 14 of 19 (74%) chose the equal split. When the receiver's payoffs were hidden but could be revealed with a free click, 18 of 32 (56%) revealed them, and only 15 of 32 (47%) revealed them and chose the option better for the receiver. Across 56 treatment effects with 6,531 participants, 40% of people avoided the information, and choices that helped the other person at a cost to oneself fell by 15.6 percentage points (95% CI 10 to 21) when people could avoid knowing (Vu, Soraperra, Leib, van der Weele and Shalvi, 2023).
More about the paradigm
Here the participant chooses between options A and B, shown as a table of the points each gives them and another person. Option A always gives the participant more. In the known version the table shows the conflict: option A gives the other person fewer points, and option B splits the points equally. In the hidden version the other person's cells show a question mark: a free button shows them, and the instructions say that for each decision it was chosen at random, with the same chance either way, whether the options conflict or option A is better for both, and that the other person will not learn whether the participant looked. One decision is standard, as in the original studies; a study can ask up to eight, each with its own points, written for this task. The version is drawn from the session's seed, so half of the sessions get each, unless the study fixes it. A table question with feedback comes first. The other person is simulated, and the last screen says so, unless the study chooses another participant kept anonymous. Points only.
The main score is the share of decisions in which the options conflicted and the participant chose the equal split. Each session plays one version, so the share is also reported in that version's column: across a study, the known mean minus the hidden mean is the drop in fair choices, and the share who look is the mean of the reveal share. Scores describe choices in a task played for points; they are research measures for comparing groups, and they do not describe a person's character or diagnose.
Scores
- Fair Choice Share%
- Over the decisions in which the options conflicted, the share in which the participant chose the equal split; null when no decision conflicted.
- Compare the known and hidden versions across a study
- Fair Choices, Known Version%
- The fair choice share when the session played the known version; null in the hidden version.
- The comparison for the hidden version
- Fair Choices, Hidden Version%
- The fair choice share when the session played the hidden version, whether or not the participant looked; null in the known version.
- A study mean below the known version's is moral wiggle room
- Points Shown%
- In the hidden version, the share of decisions in which the participant showed the other person's points before choosing; null in the known version.
- Across a study, the share of people who look
- Fair Choices After Looking%
- In the hidden version, the fair choice share over the conflicting decisions in which the participant looked; null without one.
- Compare with the known version
- Choices Worse for Bothcount
- Choices of the option worse for both people while both people's points were in view.
- 0 expected; flagged for review
- Table Checkflag
- 1 for a correct answer to the table question, 0 for a wrong one.
- A wrong answer is flagged for review
- Median Decision Timems
- Median time from a decision screen appearing to the choice, over choices at or after 100 ms.
- Informational
- Decisions Under 100 mscount
- Choices made less than 100 ms after their screen appeared.
- 0 expected
- Decisionscount
- Unique decisions in the selected run.
- Compare with planned decisions
- Planned Decisionscount
- The decisions the session planned, 1 by default and up to 8.
- Informational
- Last Screen Reachedflag
- 1 when the participant continued from the last screen, which says who the other person was.
- Zero is flagged for review
Use and evidence
44 papers from 2021 to 2025 name this task in their title or abstract, up 52% compared with 2016 to 2020. That ranks it 490 of 711 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 13 |
| 2024 | 7 |
| 2023 | 7 |
| 2022 | 9 |
| 2021 | 8 |
| 2020 | 13 |
| 2019 | 7 |
| 2018 | 6 |
| 2017 | 1 |
| 2016 | 2 |
| 2015 | 3 |
| 2014 | 2 |
| 2013 | 2 |
| 2012 | 0 |
| 2011 | 0 |
| 2010 | 2 |
| 2009 | 1 |
| 2008 | 0 |
| 2007 | 0 |
| 2006 | 1 |
| 2005 | 1 |
| 2004 | 0 |
| 2003 | 0 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 0 |
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
- 4
- 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 12 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Choice
- Two options, A and B, in a table of the points each gives the participant and another person. Option A always gives the participant more. When their interests conflict, option A gives the other person fewer points and option B splits equally.
- Versions
- Known: the other person's points are shown. Hidden: they show as question marks until the participant presses a free button; for each decision, chance decided whether the options conflict or option A is better for both, and the other person never learns whether the participant looked. The version is drawn from the session's seed, so half of the sessions get each, or fixed by the study.
- Decisions
- 1 decision by default, as in the original studies, and up to 8, each with a different person and its own points: 8 pairs written for this task, with gains of 5 to 30 points for choosing option A. One table question with feedback comes first.
- Other person
- Simulated, and the last screen says so, or another participant kept anonymous when the study gives the points to one. Points only: no money is paid.
- Scores
- The share of conflicting decisions in which the participant chose the equal split, also reported in the column of the session's version; the share of hidden decisions in which the participant looked; and fair choices after looking. Across a study, the known mean minus the hidden mean is the drop in fair choices.
Across 56 treatment effects with 6,531 participants, 40% of people avoided easily obtainable information about how their choice affects another person, and choices that helped the other person fell by 15.6 percentage points when they could avoid knowing (Vu, Soraperra, Leib, van der Weele and Shalvi, 2023). The original studies paid money; this task uses points, so compare the versions within a study. Scores describe choices for points, not a person's character, 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 selected run's unique decisions. No choice is correct, so accuracy is not judged against chance, and the same choice every time is accepted. A run short of its planned decisions, a wrong answer to the table question, a choice worse for both people after seeing both, and a last screen not reached are flagged for review. Decisions under 200 ms on at least half of four or more decisions invalidate; fewer are flagged.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. Choosing for oneself, looking or not looking are preferences, so repeated choices are not checked. A participant who misreads the table may choose the opposite of what they intend, so a wrong check answer, or choosing the option worse for both once both are in view, is flagged and never invalidates. The last screen says who the other person was, so a session that stops before it is flagged. 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 |
|---|---|---|
| Version | Drawn for each session (standard) | Drawn for each session (standard), Known: the other person's points shown, Hidden: shown with one free click |
| Number of decisions | 1 | 1 to 8 |
| Who the other person is | Simulated, said on the last screen (standard) | Simulated, said on the last screen (standard), Another participant, kept anonymous |
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 Hidden Payoffs Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Dana J, Weber RA, Kuang JX, 2007; Vu L, Soraperra I, Leib M, van der Weele J, Shalvi S, 2023), with the standard settings (version: Drawn for each session; number of decisions: 1; who the other person is: Simulated, said on the last screen). The main score was fair choice share (over the decisions in which the options conflicted, the share in which the participant chose the equal split; null when no decision conflicted).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Dana, J., Weber, R. A., & Kuang, J. X. (2007). Exploiting moral wiggle room: Experiments demonstrating an illusory preference for fairness. Economic Theory, 33(1), 67-80. https://doi.org/10.1007/s00199-006-0153-zThe known and hidden versions of a binary dictator choice, and the free button that shows the other person's payoffs.
- Grossman, Z., & van der Weele, J. J. (2017). Self-image and willful ignorance in social decisions. Journal of the European Economic Association, 15(1), 173-217. https://doi.org/10.1093/jeea/jvw001Why avoiding information can protect a person's view of themselves.
- Vu, L., Soraperra, I., Leib, M., van der Weele, J., & Shalvi, S. (2023). Ignorance by choice: A meta-analytic review of the underlying motives of willful ignorance and its consequences. Psychological Bulletin, 149(9-10), 611-635. https://doi.org/10.1037/bul0000398How many people avoid the information, and how much it lowers fair choices.
Source papers
- Vu, Soraperra, Leib, van der Weele and Shalvi (2023): across 56 treatment effects with 6,531 participants, 40% avoided information, and choices that helped the other person fell by 15.6 percentage points when information could be avoided.Paradigm source, Adults
Questions
How long does the Hidden Payoffs Task take?
About 1 to 3 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?
Fair Choice Share, Fair Choices, Known Version, Fair Choices, Hidden Version, Points Shown, Fair Choices After Looking, Choices Worse for Both, Table Check, Median Decision Time, Decisions Under 100 ms, Decisions, Planned Decisions and Last Screen Reached. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Hidden Payoffs Task used in research?
44 papers from 2021 to 2025 name it in their title or abstract, up 52% compared with 2016 to 2020. That ranks it 490 of 711 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.58.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Hidden Payoffs 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?
Version (Drawn for each session (standard), Known: the other person's points shown, Hidden: shown with one free click), number of decisions (1 to 8) and who the other person is (Simulated, said on the last screen (standard), Another participant, kept anonymous). 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.