Social Cognition
Act or Omit Task
Also called: omission bias task, omission bias, omission-commission asymmetry, action-omission asymmetry, act-omission distinction, action-inaction asymmetry, commission and omission judgments
Measure the omission bias: participants choose whether to act when acting and doing nothing both carry a risk, then rate the same harm caused by acting and by doing nothing.
6-8 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
- 81
- Evidence grade
- D
- Registered clinical studies
- 0
Doing nothing: 10 in 100 catch the fever. The treatment: 7 in 100 get it from the treatment.
Take the treatmentDo not take itHow bad was what Ana did?
What it measures
Measures the omission bias, the preference for harm by doing nothing over equal or smaller harm by acting. Participants choose what they would do in 8 situations where acting and not acting both carry a risk, then rate how bad people's acts are in 6 stories, each told once with harm by acting and once with the same harm by doing nothing.
Constructs
- Omission bias
- Moral judgment
- Decisions under risk
- Responsibility for outcomes
Research uses
- Moral psychology research
- Judgment and decision-making research
- Risk perception and risk communication research
- Within-study group comparison
How the paradigm works
The Act or Omit Task measures the omission bias: the tendency to prefer harm that comes from doing nothing over equal or smaller harm that comes from acting. In part 1 the participant reads 8 short situations. In each, doing nothing carries a risk of a harm, such as a fever from an invented illness or a parcel arriving after a birthday, and acting removes that risk but carries its own risk of the same harm, such as a treatment that can itself cause the fever. The risk of acting is 40%, 70%, 100% or 150% of the risk of doing nothing, and the participant chooses what they would do. In part 2 the participant reads 6 stories, each told twice in different halves of the part: once a person causes a harm by acting, and once the same person sees the same harm coming and does nothing. Each version is rated from 1, not bad at all, to 7, extremely bad. There is no time limit.
More about the paradigm
The choices follow Ritov and Baron (1990), whose participants imagined vaccinating a child against a flu that would kill 10 in 10,000 children with a vaccine that could itself kill: in their second experiment the highest vaccine risk accepted averaged 5.5 in 10,000 for their own child, about half the risk of not vaccinating. The ratings follow Spranca, Minsk and Baron (1991), who held knowledge and motive constant across an act and an omission and found that many people judged the act as more immoral, while many others rated the two the same. A preregistered replication found the same asymmetry between participants (Jamison, Yay and Feldman 2020), and a meta-analysis found it stronger for judgments of morality and blame than for decisions (Yeung, Yay and Feldman 2022). The situations and stories are written for this task in each language, with invented illnesses and no real medicine.
Scores
- Doing Nothing at Equal or Lower Risk%
- Percent of choices to do nothing when acting carried an equal or smaller risk of the same harm.
- Informational; higher means a stronger preference for doing nothing
- Doing Nothing at Lower Risk%
- Percent of choices to do nothing when acting carried 40% or 70% of the risk of doing nothing.
- Informational
- Doing Nothing at Equal Risk%
- Percent of choices to do nothing when acting carried the same risk as doing nothing.
- Informational
- Acting at Higher Risk%
- Percent of choices to act when acting carried 150% of the risk of doing nothing.
- Informational
- Rating Differencepoints
- For each story rated in both versions, the rating of harm by acting minus the rating of the same harm by doing nothing, averaged over stories (1 to 7 scale).
- Informational; positive when harm by acting is rated worse
- Rating of Harm by Actingpoints
- Mean rating of the stories in which the person causes the harm by acting, from 1 to 7.
- Informational
- Rating of Harm by Doing Nothingpoints
- Mean rating of the stories in which the person lets the harm happen, from 1 to 7.
- Informational
- Stories with Acting Rated Worse%
- Percent of stories rated in both versions where harm by acting got the higher rating.
- Informational
- Stories Rated the Same%
- Percent of stories rated in both versions where both versions got the same rating.
- Informational
- Median Choice Timems
- Median time from a situation appearing to the choice. Choices under 100 ms are left out.
- Informational
- Median Rating Timems
- Median time from the rating question appearing to the rating. Ratings under 100 ms are left out.
- Informational
- Median Reading Timems
- Median time on the story screens.
- Informational
- Answers Under 100 mscount
- Choices and ratings given less than 100 ms after their screen appeared.
- Lower is better
- Choices Answeredcount
- Situations answered.
- Informational
- Ratings Givencount
- Story versions rated.
- Informational
- Choices Plannedcount
- Situations in the session's plan: 8, or 0 when the study runs the ratings only.
- Informational
- Ratings Plannedcount
- Story versions in the session's plan: 12, or 0 when the study runs the choices only.
- Informational
Use and evidence
81 papers from 2021 to 2025 name this task in their title or abstract, up 35% compared with 2016 to 2020. That ranks it 325 of 711 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 24 |
| 2024 | 13 |
| 2023 | 16 |
| 2022 | 5 |
| 2021 | 23 |
| 2020 | 15 |
| 2019 | 8 |
| 2018 | 14 |
| 2017 | 16 |
| 2016 | 7 |
| 2015 | 19 |
| 2014 | 16 |
| 2013 | 8 |
| 2012 | 10 |
| 2011 | 8 |
| 2010 | 15 |
| 2009 | 10 |
| 2008 | 6 |
| 2007 | 7 |
| 2006 | 5 |
| 2005 | 8 |
| 2004 | 6 |
| 2003 | 3 |
| 2002 | 2 |
| 2001 | 3 |
| 2000 | 4 |
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
- 13
- 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.
Note on the counts: includes surveys of vaccination and treatment decisions.
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 13 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Choices
- 8 situations written for this test, the same for everyone: 3 with an invented illness and a treatment, cream or spray that can itself cause it, and 5 everyday ones such as a delivery route or a hotel booking. Doing nothing carries a risk of 10 or 20 in 100.
- Risks
- Acting removes that risk but carries its own risk of the same harm, at 40%, 70%, 100% or 150% of the risk of doing nothing, two situations at each, set by the session's seed.
- Answers
- Two buttons under each situation, acting and doing nothing, each first on half of the situations. The situation stays on screen, and there is no time limit.
- Stories
- 6 stories, each told twice in different halves: once a person causes a harm by acting, once the same person sees the same harm coming and does nothing. Each person is a drawn face with a name.
- Ratings
- How bad was what the person did, from 1, not bad at all, to 7, extremely bad, after reading each story at one's own pace.
- Practice
- 1 choice with no risk before the choices and 1 story with no harm before the ratings, not scored.
- Settings
- Choices and ratings (standard), choices only or ratings only.
The situations and stories are written for this test in each language, with invented illnesses and no real medicine, brand or person. No choice or rating is right: the scores describe how a person weighs harm by acting against harm by doing nothing. The scores support research 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: act_or_omit_choice and act_or_omit_rating from the most complete run. No choice or rating is correct, so accuracy and repeated answers are not checked. Fewer than 75% of the planned choices and ratings answered invalidates; other incomplete runs are flagged. Answers under 200 ms on at least half of 4 or more answers invalidate. A hidden tab is flagged for review.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. Always doing nothing, or always acting, is the preference measured, and rating every story alike can be a considered view that acting and not acting are equally bad, which many of Spranca, Minsk and Baron's (1991) participants held, so repeated answers are not checked. The 75% completion threshold is a ConductScience choice. 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 |
|---|---|---|
| Parts | Choices and ratings (standard) | Choices and ratings (standard), Choices only, Ratings only |
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 Act or Omit Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Ritov I, Baron J, 1990; Spranca M, Minsk E, Baron J, 1991), with the standard settings (parts: Choices and ratings). The main score was doing nothing at equal or lower risk (percent of choices to do nothing when acting carried an equal or smaller risk of the same harm).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Ritov, I., & Baron, J. (1990). Reluctance to vaccinate: Omission bias and ambiguity. Journal of Behavioral Decision Making, 3(4), 263-277. https://doi.org/10.1002/bdm.3960030404A treatment that can itself cause the harm it prevents: many people accepted only a smaller risk from acting than from doing nothing.
- Spranca, M., Minsk, E., & Baron, J. (1991). Omission and commission in judgment and choice. Journal of Experimental Social Psychology, 27(1), 76-105. https://doi.org/10.1016/0022-1031(91)90011-TStories in which an actor causes the same harm by acting or by doing nothing, with the same knowledge and motive, rated for how immoral the actor is.
- Jamison, J., Yay, T., & Feldman, G. (2020). Action-inaction asymmetries in moral scenarios: Replication of the omission bias examining morality and blame with extensions linking to causality, intent, and regret. Journal of Experimental Social Psychology, 89, 103977. https://doi.org/10.1016/j.jesp.2020.103977A preregistered replication: harm by acting was rated more immoral and more blameworthy than the same harm by doing nothing.
- Yeung, S. K., Yay, T., & Feldman, G. (2022). Action and inaction in moral judgments and decisions: Meta-analysis of omission bias omission-commission asymmetries. Personality and Social Psychology Bulletin, 48(10), 1499-1515. https://doi.org/10.1177/01461672211042315Meta-analysis of 21 samples: the asymmetry is stronger for judgments of morality and blame than for decisions.
Source papers
- Ritov I, Baron J (1990). Reluctance to vaccinate: Omission bias and ambiguity. Journal of Behavioral Decision Making, 3(4):263-277.N 20, Students
- Jamison J, Yay T, Feldman G (2020). Action-inaction asymmetries in moral scenarios. Journal of Experimental Social Psychology, 89:103977.N 313, Adults
Questions
How long does the Act or Omit Task take?
About 6 to 8 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?
Doing Nothing at Equal or Lower Risk, Doing Nothing at Lower Risk, Doing Nothing at Equal Risk, Acting at Higher Risk, Rating Difference, Rating of Harm by Acting, Rating of Harm by Doing Nothing, Stories with Acting Rated Worse, Stories Rated the Same, Median Choice Time, Median Rating Time, Median Reading Time, Answers Under 100 ms, Choices Answered, Ratings Given, Choices Planned and Ratings Planned. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Act or Omit Task used in research?
81 papers from 2021 to 2025 name it in their title or abstract, up 35% compared with 2016 to 2020. That ranks it 325 of 711 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.58.0, counted 13 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Act or Omit 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?
Parts (Choices and ratings (standard), Choices only, Ratings only). 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, with its words in the same language. 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.