Social Cognition
Trust Game
Also called: investment game
Measure trust and trustworthiness: participants send points to programmed partners and choose how much to send back, scored by the shares sent and returned.
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
- 794
- Evidence grade
- C
- Registered clinical studies
- 26
What it measures
Measures trust and trustworthiness: the participant sends points to programmed partners, who receive three times the amount, and decides how much to send back when a partner sends to them.
Constructs
- Trust
- Reciprocity
- Social preferences
- Social decision making
Research uses
- Behavioral economics research
- Trust and reciprocity studies
- Social decision-making research in clinical groups
- Within-study group comparison
How the paradigm works
The Trust Game measures trust and trustworthiness with points. The participant plays both roles with partners that are computer programs following set rules, and by default the instructions say so.
More about the paradigm
The task follows Berg, Dickhaut and McCabe (1995). As investor, the participant has 10 points in each round and sends any whole number of them to a new partner, who receives three times the amount. Each partner sends back nothing, the amount sent or twice the amount sent, by a rule drawn from the session's seed and saved with the round. What came back is shown only at the end, so every send is a one-shot decision. As trustee, the participant answers for each amount a partner could send, 1 to 10 points as standard, by choosing how many of the tripled points to send back (the strategy method). The partner's actual send, drawn from the seed, picks the answer that counts. Each role starts with one comprehension question with feedback. The standard task has 5 investor rounds and plays the investor role first. Researchers can change or counterbalance the role order, set 1 to 10 investor rounds, ask about the even amounts only, use a neutral partner wording, and choose the payment line.
Scores
- Share Sent%
- Points sent as investor as a proportion of the 10 points, averaged over rounds.
- Higher means more trust
- Amount Sentpoints
- Points sent as investor, averaged over rounds.
- Higher means more trust
- Sent Nothing%
- Share of investor rounds that sent nothing.
- Informational
- Sent Everything%
- Share of investor rounds that sent all 10 points.
- Informational
- Share Returned%
- Points sent back as trustee as a proportion of the points received, averaged over the amounts a partner could send.
- Higher means more trustworthiness
- Share Returned at 30 Points%
- Share sent back when a partner sent all 10 points, so 30 points were received. Shares are also reported for every other amount received.
- Higher means more trustworthiness
- Comprehension Check%
- Share of the two comprehension questions, one per role, answered correctly.
- 1 supports interpretation
Use and evidence
794 papers from 2021 to 2025 name this task in their title or abstract, up 3% compared with 2016 to 2020. That ranks it 31 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 184 |
| 2024 | 129 |
| 2023 | 146 |
| 2022 | 147 |
| 2021 | 188 |
| 2020 | 161 |
| 2019 | 162 |
| 2018 | 156 |
| 2017 | 139 |
| 2016 | 153 |
| 2015 | 139 |
| 2014 | 117 |
| 2013 | 129 |
| 2012 | 124 |
| 2011 | 120 |
| 2010 | 111 |
| 2009 | 93 |
| 2008 | 81 |
| 2007 | 82 |
| 2006 | 60 |
| 2005 | 48 |
| 2004 | 32 |
| 2003 | 17 |
| 2002 | 18 |
| 2001 | 13 |
| 2000 | 12 |
Use grade B
300 to 999 papers from 2021 to 2025.
Evidence grade C
At least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies.
- Papers on test-retest reliability
- 7
- Papers on reliability or validity
- 109
- Registered clinical studies using it as an outcome
- 26
Note on the counts: may include finance papers on investment games.
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 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Partners
- Computer programs that follow set rules, named by letter. The standard instructions say so; a neutral wording says only "your partner".
- Investor role
- 5 rounds (standard; 1 to 10), each with 10 points and a new partner. Points sent are tripled.
- Partner rules
- Each partner sends back nothing, the amount sent or twice the amount sent, equally often, in an order drawn from the session's seed. The results show only at the end.
- Trustee role
- One answer for each amount a partner could send, 1 to 10 points (standard) or the even amounts, in ascending order. The partner's actual send, drawn from the seed, picks the answer that counts.
- Input
- A bar to tap or drag, with buttons that change the amount by 1. Confirm stays off until an amount is chosen.
- Check
- One question with feedback at the start of each role.
- Role order
- Investor first (standard), trustee first, or counterbalanced across sessions.
- Money
- Points are hypothetical by default; a setting changes the instruction line to say the study pays.
Berg and colleagues paired participants once, paid in dollars, and the second mover decided after seeing the amount sent. Here partners are programmed, the investor role is repeated with new partners, and the trustee role answers for every amount a partner could send.
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 decisions in both roles. No amount is correct, so accuracy is not judged against chance, and the same amount on every decision is accepted. A wrong answer to either comprehension question is flagged for review. Decisions confirmed under 200 ms on at least half the decisions invalidate.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. Sending or returning the same share every time is a stable preference, so repeated amounts are not checked. A participant who misreads how points are tripled or kept may choose the opposite of what they intend, so a wrong answer to a comprehension question 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 |
|---|---|---|
| Role order | Investor first (standard) | Investor first (standard), Trustee first, Counterbalanced across sessions |
| Investor rounds | 5 | 1 to 10 |
| Trustee answers | Each amount from 1 to 10 (standard) | Each amount from 1 to 10 (standard), Even amounts only: 2, 4, 6, 8 and 10 |
| 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 Trust Game in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Berg J, Dickhaut J, McCabe K, 1995), with the standard settings (role order: Investor first; investor rounds: 5; trustee answers: Each amount from 1 to 10; partner wording: A computer program; payment line: Hypothetical points). The main score was share sent (points sent as investor as a proportion of the 10 points, averaged over rounds).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Berg, J., Dickhaut, J., & McCabe, K. (1995). Trust, reciprocity, and social history. Games and Economic Behavior, 10(1), 122-142. https://doi.org/10.1006/game.1995.1027
- Johnson, N. D., & Mislin, A. A. (2011). Trust games: A meta-analysis. Journal of Economic Psychology, 32(5), 865-889. https://doi.org/10.1016/j.joep.2011.05.007Protocol choices that change how much people send and return.
- Brandts, J., & Charness, G. (2011). The strategy versus the direct-response method: A first survey of experimental comparisons. Experimental Economics, 14(3), 375-398. https://doi.org/10.1007/s10683-011-9272-xThe strategy method in the trustee role.
Source papers
- Berg, Dickhaut and McCabe (1995): the trust game. Johnson and Mislin (2011): meta-analysis of 162 replications with more than 23,000 participants.Paradigm source, Adults
Questions
How long does the Trust Game 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?
Share Sent, Amount Sent, Sent Nothing, Sent Everything, Share Returned, Share Returned at 30 Points and Comprehension Check. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Trust Game used in research?
794 papers from 2021 to 2025 name it in their title or abstract, up 3% compared with 2016 to 2020. That ranks it 31 of 395 test families in the ConductCognition test census (use grade B). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Trust Game have?
In the published research, 7 papers report its test-retest reliability and 26 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Role order (Investor first (standard), Trustee first, Counterbalanced across sessions), investor rounds (1 to 10), trustee answers (Each amount from 1 to 10 (standard), Even amounts only: 2, 4, 6, 8 and 10), 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.