Decision Making
Game of Dice Task
Also called: GDT, dice gambling task, dice betting task, decision making under risk task, decisions under explicit risk
Measure decisions under risk with known rules: before each roll of a die, bet on one to four numbers, scored as safe minus risky bets.
4-6 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
- 51
- Evidence grade
- C
- Registered clinical studies
- 4
What it measures
Measures decisions under risk when the rules are known. Before each of 18 rolls of a die, participants bet on one number or on two, three or four numbers. A bet on fewer numbers wins and loses more points but wins less often, so over many rolls it loses points. The main score is the net score: safe bets (three or four numbers) minus risky bets (one or two).
Constructs
- Decision-making under risk
- Use of feedback
- Risk taking
Research uses
- Decision-making in addiction and gambling research
- Neurological and psychiatric research
- Aging and executive function research
How the paradigm works
The Game of Dice Task measures decisions under risk when the rules are known. Brand and colleagues (2005) built it so that the chance of winning and the points at stake on every option can be read or worked out from the start, unlike the Iowa Gambling Task, where they must be learned. Before each of 18 rolls of a die, participants bet on one number or on two, three or four numbers. A bet wins when the die shows one of its numbers and loses the same points when it does not, and a bet on fewer numbers wins and loses more. Bets on one or two numbers lose points on average and are called risky; bets on three or four are called safe.
More about the paradigm
This version uses its own stakes: 400 points on one number, 250 on two, 150 on three and 80 on four, so each roll loses 266.7, 83.3 or 0 points on average, or gains 26.7. The game starts at 500 points, and the points can fall below zero. Participants click or tap any 1 to 4 numbers, then press Roll. The stakes, the points and the rolls left stay on screen, and each outcome shows for 2 seconds. The 18 rolls (12 to 36 in the settings) hold each face equally often, in an order drawn for each session. One check question with feedback comes first; there are no practice rolls. About 4 to 6 minutes, on desktops, laptops, tablets and phones.
The main score is the net score: safe bets minus risky bets, from -18 to 18 at the standard length. The bets of each size, the final points and the use of feedback are reported too: how often a risky bet that lost was followed by a safe one, and the other switches after wins and losses, as Starcke and colleagues (2011) measure it. Bets placed under 100 ms count as anticipations and are left out.
Scores
- Net Scorecount
- Safe bets (on three or four numbers) minus risky bets (on one or two numbers), over the rolls not placed as anticipations.
- Higher means more bets that keep points on average
- Safe Betscount
- Bets on three or four numbers.
- Informational
- Risky Betscount
- Bets on one or two numbers, which lose points on average.
- Lower means fewer disadvantageous bets
- Bets on One Numbercount
- Bets on a single number, the riskiest bet.
- Informational
- Bets on Two Numberscount
- Bets on two numbers.
- Informational
- Bets on Three Numberscount
- Bets on three numbers.
- Informational
- Bets on Four Numberscount
- Bets on four numbers, the bet with the best average.
- Informational
- Final Pointspoints
- Points after the last roll; the game starts at 500.
- Higher is better, though luck plays a part
- Safe Bet After a Risky Loss%
- Share of risky bets that lost and were followed by a safe bet.
- Higher means more use of losses to change the bet
- Safe Bet After a Risky Win%
- Share of risky bets that won and were followed by a safe bet.
- Informational
- Risky Bet After a Safe Loss%
- Share of safe bets that lost and were followed by a risky bet.
- Informational
- Risky Bet After a Safe Win%
- Share of safe bets that won and were followed by a risky bet.
- Informational
- Median Decision Timems
- Median time from a roll's screen to the press of Roll.
- Informational
- Check Question Correct0/1
- 1 when the check question on losing a bet was answered correctly.
- 0 is flagged for review
- Anticipationscount
- Rolls pressed under 100 ms after their screen, left out of the scores.
- Lower is better
- Rolls Playedtrials
- Rolls in the scored run.
- Fewer than planned is flagged for review
- Rolls Plannedtrials
- Rolls the run set out to play.
- Informational
Use and evidence
51 papers from 2021 to 2025 name this task in their title or abstract, down 29% compared with 2016 to 2020. That ranks it 254 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 14 |
| 2024 | 10 |
| 2023 | 10 |
| 2022 | 8 |
| 2021 | 9 |
| 2020 | 9 |
| 2019 | 10 |
| 2018 | 15 |
| 2017 | 27 |
| 2016 | 11 |
| 2015 | 13 |
| 2014 | 19 |
| 2013 | 21 |
| 2012 | 10 |
| 2011 | 15 |
| 2010 | 6 |
| 2009 | 7 |
| 2008 | 11 |
| 2007 | 10 |
| 2006 | 3 |
| 2005 | 3 |
| 2004 | 1 |
| 2003 | 0 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 0 |
Use grade D
30 to 99 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
- 1
- Papers on reliability or validity
- 9
- Registered clinical studies using it as an outcome
- 4
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.
- Rolls
- A die is rolled 18 times (standard; 12 to 36 in sets of six). The rolls hold each face equally often, 3 times in 18, in an order drawn from the session's seed before the first roll.
- Bets
- Before each roll the participant clicks or taps 1 to 4 of the six numbers and presses Roll, with no time limit. A bet wins when the die shows one of its numbers and loses the same points when it does not.
- Stakes
- 400 points on 1 number, 250 on 2, 150 on 3 and 80 on 4, shown on screen throughout. The game starts at 500 points, and the points can fall below zero.
- Outcome
- The die, the win or loss and the new points show for 2,000 ms, beside the rolls left.
- Practice
- One check question on losing a bet, with feedback. There are no practice rolls, so no early outcome shapes the first bets.
- Score
- The net score: bets on 3 or 4 numbers minus bets on 1 or 2. On average a bet on 1 number loses 266.7 points per roll, on 2 loses 83.3, on 3 breaks even and on 4 gains 26.7. Also reported: the bets of each size, the final points and the share of bets switched after a win or a loss.
- Points
- The study decides what points are worth, and the payment line says whether they are exchanged for money. The standard line says they are not.
Brand and colleagues (2005) offered 14 fixed sets of numbers, with stakes of their own and a sound after each throw. This test lets the participant choose any 1 to 4 numbers, uses its own stakes and shows each outcome without sound, so its values are not directly comparable with published ones. 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 rolls. No bet is correct on a single roll, so answers are not judged against chance, and the same kind of bet on every roll is not judged either. Bets placed under 200 ms after the roll's screen on at least half the rolls invalidate. Rolls pressed under 100 ms are counted as anticipations and left out of the scores. A run short of its planned rolls is flagged and withholds the percentile; a wrong check answer is flagged.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. Betting on four numbers on every roll is the best strategy, so a repeated bet is a valid result. 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.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Rolls, in sets of six | 3 | 2 to 6 |
| 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.
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Game of Dice Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Brand, Fujiwara, Borsutzky, Kalbe, Kessler and Markowitsch, 2005), with the standard settings (rolls, in sets of six: 3; payment line: Hypothetical points). The main score was net score (safe bets (on three or four numbers) minus risky bets (on one or two numbers), over the rolls not placed as anticipations).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Brand, M., Fujiwara, E., Borsutzky, S., Kalbe, E., Kessler, J., & Markowitsch, H. J. (2005). Decision-making deficits of Korsakoff patients in a new gambling task with explicit rules: Associations with executive functions. Neuropsychology, 19(3), 267-277. https://doi.org/10.1037/0894-4105.19.3.267The task: 18 throws of a die, bets on one to four numbers, stakes that fall as the odds rise, and risky against safe choices.
- Brand, M., Labudda, K., & Markowitsch, H. J. (2006). Neuropsychological correlates of decision-making in ambiguous and risky situations. Neural Networks, 19(8), 1266-1276. https://doi.org/10.1016/j.neunet.2006.03.001Decisions under risk set apart from decisions under ambiguity.
- Brand, M. (2008). Does the feedback from previous trials influence current decisions? A study on the role of feedback processing in making decisions under explicit risk conditions. Journal of Neuropsychology, 2(2), 431-443. https://doi.org/10.1348/174866407X220607More disadvantageous choices when no feedback followed each throw.
- Starcke, K., Pawlikowski, M., Wolf, O. T., Altstötter-Gleich, C., & Brand, M. (2011). Decision-making under risk conditions is susceptible to interference by a secondary executive task. Cognitive Processing, 12(2), 177-182. https://doi.org/10.1007/s10339-010-0387-3The net score, safe minus risky choices, and the use of feedback as switching between risky and safe bets after a gain or a loss.
- Schiebener, J., & Brand, M. (2015). Decision making under objective risk conditions: A review of cognitive and emotional correlates, strategies, feedback processing, and external influences. Neuropsychology Review, 25(2), 171-198. https://doi.org/10.1007/s11065-015-9285-x
Source papers
- Brand, Fujiwara, Borsutzky, Kalbe, Kessler and Markowitsch (2005): 35 healthy adults made a median of 2 risky choices in 18 throws (range 0 to 18), with a median final outcome of 500 fictive euros.Paradigm source, Adults
Questions
How long does the Game of Dice 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?
Net Score, Safe Bets, Risky Bets, Bets on One Number, Bets on Two Numbers, Bets on Three Numbers, Bets on Four Numbers, Final Points, Safe Bet After a Risky Loss, Safe Bet After a Risky Win, Risky Bet After a Safe Loss, Risky Bet After a Safe Win, Median Decision Time, Check Question Correct, Anticipations, Rolls Played and Rolls Planned. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Game of Dice Task used in research?
51 papers from 2021 to 2025 name it in their title or abstract, down 29% compared with 2016 to 2020. That ranks it 254 of 395 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Game of Dice Task have?
In the published research, 1 paper reports its test-retest reliability and 4 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Which settings can a researcher change?
Rolls, in sets of six (2 to 6) 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.