Social Cognition
Game of Chicken
Also called: chicken game, hawk-dove game, hawk-dove task, snowdrift game
Measure yielding and conflict: participants choose A or B in each round with one programmed partner, where holding out pays most against a partner who yields and least against one who holds out.
4-7 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Preview any test free after you sign up.
- Papers, 2021 to 2025
- 109
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures yielding and conflict: in repeated rounds, the participant and a programmed partner each choose to yield or to hold out, where holding out against a partner who yields pays most but both holding out pays least.
Constructs
- Conflict and appeasement
- Risk taking
- Strategic decision making
- Social decision making
Research uses
- Behavioral economics research
- Conflict and negotiation studies
- Social decision-making research in clinical groups
- Within-study group comparison
How the paradigm works
The Game of Chicken measures yielding and conflict when two sides want the same thing. In each round the participant and a partner each choose one of two options at the same time: yield or hold out. Holding out against a partner who yields gives 10 points and the partner 3; both yielding gives 7 each; both holding out gives nothing. So the best reply to either choice is the other one, and each side gains by holding out only if the other gives way. The partner is a computer program that follows set rules, and by default the instructions say so.
More about the paradigm
The game is the game of chicken that Rapoport and Chammah (1966) studied as brinksmanship and appeasement, also known as the hawk-dove game. The points are ConductScience's own: as in chicken, holding out against yielding pays more than both yielding, which pays more than yielding against holding out, and both holding out pays least. Two rounds of both yielding pay more than taking turns, so both yielding is the cooperative outcome, and in the mixed equilibrium each side yields half the time. The participant plays 40 rounds as standard against one partner that answers the way people answered in a hawk-dove study (Neugebauer, Poulsen and Schram, 2008): it yields in round 1 with a 60% chance, then with a 40% chance after the participant yielded, since most people answered a yielder by holding out, and with an 80% chance after the participant held out, since most answered holding out by yielding. The number of rounds is not shown, so no round is known to be the last. The options are labelled A and B, and which letter is yield is drawn from the session's seed. A comprehension question with feedback comes first, and both choices and the points show after each round. Researchers can set 10 to 100 rounds, show the number of rounds, choose a partner that always yields, always holds out or chooses at random, use a neutral partner wording, and choose the payment line.
The main score is the share of rounds in which the participant yielded; the share held out is the rest. The task also reports the first choice, made before seeing the partner, the share of rounds in which both held out (mutual conflict), yielding after the partner yielded and after it held out, and the total points. In Neugebauer and colleagues' study, 53% of people answered each choice with the other, 30% always yielded, 6% always held out and 5% matched the other's choice: a majority chose what paid them more, and few reciprocated. In repeated two-person games of chicken, people played less competitively than teams did (Bornstein, Budescu and Zamir, 1997), and in a repeated game of the same kind (the snowdrift game) people cooperated more than in a repeated prisoner's dilemma (Kümmerli and colleagues, 2007). Scores support research and do not diagnose.
Scores
- Yield Rate%
- Share of rounds in which the participant yielded. The share held out is 1 minus this.
- Higher means giving way more often
- First-Round Yield0/1
- 1 when the participant yielded in round 1, before seeing the partner's choice, 0 when not.
- Informational
- Conflict Rate%
- Share of rounds in which both held out, the outcome that pays both least.
- Higher means more mutual conflict
- Yield After Partner Yielded%
- Share of rounds after the partner yielded in which the participant yielded.
- Lower means taking advantage of a partner who gave way
- Yield After Partner Held Out%
- Share of rounds after the partner held out in which the participant yielded.
- Higher means giving way after the partner held out
- Total Pointspoints
- Points the participant earned over all rounds.
- Informational
- Comprehension Check%
- Share of the comprehension questions answered correctly.
- 1 supports interpretation
- Median Decision Timems
- Median time from a round's start 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. They still count in the shares.
- 0 expected
- Roundscount
- Rounds scored.
- Informational
Use and evidence
109 papers from 2021 to 2025 name this task in their title or abstract, up 16% compared with 2016 to 2020. That ranks it 178 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 25 |
| 2024 | 23 |
| 2023 | 16 |
| 2022 | 18 |
| 2021 | 27 |
| 2020 | 21 |
| 2019 | 19 |
| 2018 | 17 |
| 2017 | 22 |
| 2016 | 15 |
| 2015 | 28 |
| 2014 | 23 |
| 2013 | 22 |
| 2012 | 22 |
| 2011 | 14 |
| 2010 | 17 |
| 2009 | 14 |
| 2008 | 15 |
| 2007 | 15 |
| 2006 | 15 |
| 2005 | 9 |
| 2004 | 5 |
| 2003 | 11 |
| 2002 | 12 |
| 2001 | 6 |
| 2000 | 6 |
Use grade C
100 to 299 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 15
- Registered clinical studies using it as an outcome
- 0
Note on the counts: includes game-theory and politics papers that run no task.
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 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Partner
- One computer program that follows set rules, for the whole game. The standard instructions say so; a neutral wording says only "your partner".
- Points
- Both yield: 7 each. Both hold out: 0 each. Yield against hold out: 3 for yield and 10 for hold out. In the mixed equilibrium each side yields with a 50% chance.
- Options
- A and B, chosen at the same time. Which letter is yield is drawn from the session's seed and saved; the screens show only the letters.
- Rounds
- 40 (standard; 10 to 100) against the same partner. The number is not shown unless a setting shows it.
- Partner rule
- Typical replies (standard): yields with a 60% chance in round 1, then 40% after the participant yielded and 80% after the participant held out, from the replies people gave in a hawk-dove study. Or always yield, always hold out, or random with a 50% chance of yielding in each round.
- Outcome
- Both choices and the points show for 2,000 ms after each round.
- Check
- One question with feedback before the first round: the points for a pair of different choices drawn from the seed, with 0, 3, 7 and 10 as the answers.
- Money
- Points are hypothetical by default; a setting changes the instruction line to say the study pays.
The points are ConductScience's own, in the order of the game of chicken. The standard partner's chances come from the replies 94 students gave in Neugebauer, Poulsen and Schram's (2008) hawk-dove game, which asked for one reply to the other's choice; here the partner answers the participant's choice in the round before. The points are hypothetical unless the study pays. 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: the scored run's choices in each round. No choice is correct, so accuracy is not judged against chance, and choosing to yield or to hold out in every round is accepted. A wrong comprehension answer is 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. Choosing one option in every round is a strategy, so repeated choices are not checked. A participant who misreads the points may choose the opposite of what they intend, so a wrong comprehension answer is flagged for review and never invalidates. 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 |
|---|---|---|
| Partner strategy | Typical replies (standard) | Typical replies (standard), Always yield, Always hold out, Random: 50% chance of yielding each round |
| Rounds | 40 | 10 to 100 |
| Number of rounds shown | Off | On or off |
| Partner wording | A computer program (standard) | A computer program (standard), Neutral: your partner |
| 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 Game of Chicken in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Rapoport A, Chammah AM, 1966; Neugebauer T, Poulsen A, Schram A, 2008), with the standard settings (partner strategy: Typical replies; rounds: 40; number of rounds shown: Off; partner wording: A computer program; payment line: Hypothetical points). The main score was yield rate (share of rounds in which the participant yielded).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Rapoport, A., & Chammah, A. M. (1966). The game of chicken. American Behavioral Scientist, 10(3), 10-28. https://doi.org/10.1177/000276426601000303The game of chicken in the laboratory, as brinksmanship and appeasement.
- Neugebauer, T., Poulsen, A., & Schram, A. (2008). Fairness and reciprocity in the Hawk-Dove game. Journal of Economic Behavior & Organization, 66(2), 243-250. https://doi.org/10.1016/j.jebo.2005.12.008How people reply to yielding and holding out, the standard partner's rule.
- Bornstein, G., Budescu, D., & Zamir, S. (1997). Cooperation in intergroup, N-person, and two-person games of chicken. Journal of Conflict Resolution, 41(3), 384-406. https://doi.org/10.1177/0022002797041003003The repeated two-person game beside team versions.
- Kümmerli, R., Colliard, C., Fiechter, N., Petitpierre, B., Russier, F., & Keller, L. (2007). Human cooperation in social dilemmas: Comparing the Snowdrift game with the Prisoner's Dilemma. Proceedings of the Royal Society B, 274(1628), 2965-2970. https://doi.org/10.1098/rspb.2007.0793More cooperation in a repeated game of the same kind than in the prisoner's dilemma.
Source papers
- Neugebauer, Poulsen and Schram (2008): replies of 94 students in a hawk-dove game, which set the standard partner's rule. Rapoport and Chammah (1966): the game of chicken in the laboratory.Paradigm source, Adults
Questions
How long does the Game of Chicken take?
About 4 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?
Yield Rate, First-Round Yield, Conflict Rate, Yield After Partner Yielded, Yield After Partner Held Out, Total Points, Comprehension Check, 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 Game of Chicken used in research?
109 papers from 2021 to 2025 name it in their title or abstract, up 16% compared with 2016 to 2020. That ranks it 178 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Game of Chicken 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?
Partner strategy (Typical replies (standard), Always yield, Always hold out, Random: 50% chance of yielding each round), rounds (10 to 100), number of rounds shown (On or off), partner wording (A computer program (standard), Neutral: your partner) 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.