Executive Function
Plea Decision Task
Also called: plea bargaining task, plea-bargain vignettes, guilty plea decisions, plea offer acceptance
Measure when people accept a plea deal across eight invented cases, with guilt, strength of evidence and the size of the offer varied within each person.
4-6 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
- 70
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Participants imagine being accused in eight invented minor cases, guilty in some and innocent in others, and decide whether to accept a plea deal or go to trial. Measures acceptance by guilt, strength of evidence and size of the offer.
Constructs
- Legal decision making
- Risky choice
- Fairness
- Expected penalty
Research uses
- Legal psychology research
- Plea bargaining research
- Behavioral economics of legal choices
How the paradigm works
The Plea Decision Task asks participants to imagine being accused in eight invented cases of minor, non-violent offenses, such as taking a bicycle or reusing a train ticket. Each case says whether the participant did it, describes the evidence with the lawyer's estimate of convictions in 10 at trial, and states the penalties: 10 penalty points if found guilty at trial, none if found not guilty, or the deal's penalty with no trial. The participant accepts the deal or goes to trial, and no outcome follows a choice.
More about the paradigm
The design crosses guilt (guilty or innocent), evidence (8 or 3 convictions in 10) and the offer (2 or 6 penalty points) within each person. The session seed deals the eight cells to the eight cases, one each, and shuffles the case order, and the two buttons keep one order for the whole session. A rule check with feedback comes first.
Scores are acceptance overall and by guilt, evidence and offer, the differences between levels, and the share of decisions that chose the lower expected penalty, the deal when its penalty is below the chance of conviction times the trial penalty. The penalties are points, and the end screen tells participants that the cases were invented. Instructions and cases are available in eight languages.
Scores
- Acceptance Rate%
- Share of the eight cases in which the deal was accepted.
- Higher means more deals accepted.
- Guilty Acceptance%
- Share of deals accepted in the four cases where the participant did it.
- Compare with Innocent Acceptance.
- Innocent Acceptance%
- Share of deals accepted in the four cases where the participant did not do it.
- Deals accepted despite innocence.
- Guilt Effect%
- Guilty acceptance minus innocent acceptance.
- Positive values mean deals were taken more often when guilty.
- Strong Evidence Acceptance%
- Share of deals accepted when the lawyer gives 8 convictions in 10.
- Compare with weak evidence.
- Weak Evidence Acceptance%
- Share of deals accepted when the lawyer gives 3 convictions in 10.
- Compare with strong evidence.
- Evidence Effect%
- Strong-evidence acceptance minus weak-evidence acceptance.
- Positive values mean deals were taken more often when conviction was likely.
- Large Discount Acceptance%
- Share of deals accepted when the deal costs 2 penalty points against 10 at trial.
- Compare with the small discount.
- Small Discount Acceptance%
- Share of deals accepted when the deal costs 6 penalty points against 10 at trial.
- Compare with the large discount.
- Discount Effect%
- Large-discount acceptance minus small-discount acceptance.
- Positive values mean larger discounts drew more deals.
- Lower Expected Penalty Share%
- Share of decisions that chose the lower expected penalty: the deal when its penalty is below the chance of conviction times the trial penalty, trial otherwise.
- 1 means every decision followed expected penalties; guilt plays no part in this benchmark.
- Decisionscount
- Decisions scored in the selected run, one per case.
- 8 in a complete run.
- Planned Decisionscount
- Decisions the run plans: 8.
- Compare with Decisions.
- Rule Check Correct0/1
- 1 when the rule check was answered correctly, 0 when not; null without a check.
- 0 is flagged for review.
- Median Decision Timems
- Median time from a case appearing to the decision, including reading, over times of at least 100 ms.
- Decision pace.
- Anticipationscount
- Decisions made in under 100 ms.
- Should be 0.
Use and evidence
70 papers from 2021 to 2025 name this task in their title or abstract, up 71% compared with 2016 to 2020. That ranks it 331 of 632 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 22 |
| 2024 | 12 |
| 2023 | 14 |
| 2022 | 9 |
| 2021 | 13 |
| 2020 | 10 |
| 2019 | 15 |
| 2018 | 6 |
| 2017 | 5 |
| 2016 | 5 |
| 2015 | 3 |
| 2014 | 4 |
| 2013 | 10 |
| 2012 | 6 |
| 2011 | 7 |
| 2010 | 2 |
| 2009 | 1 |
| 2008 | 2 |
| 2007 | 1 |
| 2006 | 2 |
| 2005 | 3 |
| 2004 | 0 |
| 2003 | 1 |
| 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
- 8
- 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: Invented minor cases imagined from the accused's side; penalties in points; the laboratory accusation of the source is not staged..
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 0.44.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Cases
- 8 original cases of minor, non-violent offenses, imagined from the accused's side. Each says whether the participant did it, the evidence, the lawyer's estimate of convictions in 10 at trial and the penalties.
- Penalties
- 10 penalty points if found guilty at trial and none if found not guilty; the deal costs 2 or 6 penalty points with no trial. Points only.
- Conditions
- Guilt (guilty or innocent), evidence (8 or 3 convictions in 10) and offer (2 or 6 penalty points) crossed: each of the 8 cells once per person, dealt to the cases by the session's seed.
- Order
- Seeded case order and one button order for the whole session, untimed. One rule check with feedback comes first, and no outcome follows a choice.
- Scores
- Acceptance overall and by guilt, evidence and offer, the differences between levels, and the share of choices with the lower expected penalty.
- Cases and debrief
- Everyday minor offenses written for this task. The end screen tells participants the cases were invented.
Choices are hypothetical: no accusation is staged, as it was in the laboratory version of Dervan and Edkins, and penalties are points.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: one decision per case in the selected run. No decision is correct, and repeated decisions are allowed. Missing decisions or a wrong rule check are flagged. Decisions under 200 ms on at least half the cases invalidate.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. Decisions describe choices about invented cases, not the person who makes them. 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
This test runs with fixed settings.
Need a different version of this test?
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 Plea Decision Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Bordens, 1984; Tor, Gazal-Ayal & Garcia, 2010). The main score was acceptance rate (share of the eight cases in which the deal was accepted).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Bordens, K. S. (1984). The effects of likelihood of conviction, threatened punishment, and assumed role on mock plea bargaining decisions. Basic and Applied Social Psychology, 5(1), 59-74. https://doi.org/10.1207/s15324834basp0501_4
- Tor, A., Gazal-Ayal, O., & Garcia, S. M. (2010). Fairness and the willingness to accept plea bargain offers. Journal of Empirical Legal Studies, 7(1), 97-116. https://doi.org/10.1111/j.1740-1461.2009.01171.x
- Dervan, L. E., & Edkins, V. (2012). The innocent defendant's dilemma: An innovative empirical study of plea bargaining's innocence problem. Working paper, published in the Journal of Criminal Law and Criminology, 103(1), 2013. https://doi.org/10.2139/ssrn.2071397Guilty and innocent students facing a real accusation in the laboratory; this task asks for hypothetical choices.
- Edkins, V. A., & Dervan, L. E. (2013). Pleading innocents: Laboratory evidence of plea bargaining's innocence problem. Current Research in Social Psychology, 21(2). https://doi.org/10.1037/e571212013-065Methods and results of the same laboratory study.
- Redlich, A. D., Bibas, S., Edkins, V. A., & Madon, S. (2017). The psychology of defendant plea decision making. American Psychologist, 72(4), 339-352. https://doi.org/10.1037/a0040436Review of plea decision research.
Source papers
Questions
How long does the Plea Decision Task take?
About 4 to 6 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?
Acceptance Rate, Guilty Acceptance, Innocent Acceptance, Guilt Effect, Strong Evidence Acceptance, Weak Evidence Acceptance, Evidence Effect, Large Discount Acceptance, Small Discount Acceptance, Discount Effect, Lower Expected Penalty Share, Decisions, Planned Decisions, Rule Check Correct, Median Decision Time and Anticipations. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Plea Decision Task used in research?
70 papers from 2021 to 2025 name it in their title or abstract, up 71% compared with 2016 to 2020. That ranks it 331 of 632 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.44.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Plea Decision 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 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.