Decision Making
Willingness to Wait Task
Also called: WTW task, willingness-to-wait task, voluntary persistence task, token waiting task
Measure persistence when the wait is uncertain: participants wait for tokens that pay once ready or sell them early, in one part where waiting pays and one where it does not.
11-13 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
- 102
- Evidence grade
- C
- Registered clinical studies
- 4
What it measures
Measures how long people keep waiting for a reward whose timing is uncertain. Tokens pay points once they are ready, and participants sell or wait in two timed parts: one where waiting pays and one where selling early pays.
Constructs
- Persistence
- Delay of gratification
- Temporal expectation
- Value-based decision making
Research uses
- Decision-making research
- Self-control and impulsivity studies
- Computational psychiatry batteries
- Within-study group comparison
How the paradigm works
The willingness to wait task measures how long people keep waiting for a reward when they cannot know how long it will take, and whether they fit that persistence to the timing they meet (McGuire and Kable, 2012). Tokens appear one at a time, each worth 0 points. After a delay a token turns gold and is worth 10 points. The participant sells a token with a button or the space bar: a gold token pays its points, and an earlier sale pays nothing but brings the next token sooner, after a 2 second gap. Each part lasts a fixed time, so time spent waiting for one token cannot be spent on others.
More about the paradigm
The task has two parts of 5 minutes each as standard (3 to 12), after a 3 token practice, in an order drawn from the session's seed. In one part the delays are spread evenly from 0 to 12 seconds, so the best policy is to wait for every token. In the other they follow a long-tailed distribution: half the tokens are ready within about 3 seconds, but a token that has not paid by then is likely to take much longer, so the best policy sells any token not ready after 2.75 seconds. The two best policies earn about the same points per second, so the parts differ in the persistence they reward, not in what they pay. Each run of 8 tokens takes one delay from each eighth of its part's distribution, so every participant meets the distributions as they are. The parts' tokens differ in color, and participants are not told how the delays differ.
The main score is the difference in willingness to wait between the parts. Willingness to wait is the area under a part's survival curve of waiting times over the first 12 seconds: how many of those seconds the participant was willing to wait, with a token that turned gold, or was still waiting when the part ended, counted as waited at least that long. In McGuire and Kable's experiments, people waited longer where waiting paid. Scores support research and do not diagnose.
Scores
- Willingness to Wait Differenceseconds
- Willingness to wait where waiting pays minus willingness to wait where selling early pays, out of the first 12 seconds.
- Positive means waiting longer where waiting pays; the two best policies give 9.25
- Willingness to Wait, Waiting Paysseconds
- Area under the survival curve of waiting times over the first 12 seconds, in the part where waiting pays.
- Higher means waiting longer; the best policy gives 12
- Willingness to Wait, Selling Early Paysseconds
- Area under the survival curve of waiting times over the first 12 seconds, in the part where selling early pays.
- Lower means selling sooner; the best policy gives 2.75
- Tokens Sold Early, Waiting Payscount
- Tokens sold before they turned gold, in the part where waiting pays.
- Informational
- Tokens Sold Early, Selling Early Payscount
- Tokens sold before they turned gold, in the part where selling early pays.
- Informational
- Tokens, Waiting Payscount
- Tokens shown in the part where waiting pays.
- A part with none makes the session invalid
- Tokens, Selling Early Payscount
- Tokens shown in the part where selling early pays.
- A part with none makes the session invalid
- Pointspoints
- Points from tokens sold after they turned gold, both parts.
- Higher means more reward earned
- Time to Sell a Gold Tokenms
- Median time from a token turning gold to its sale.
- Above 2,000 ms is flagged for review
- Tokens with Focus Losscount
- Tokens during which the page lost focus; the clock kept running.
- Above 0 is flagged for review
Use and evidence
102 papers from 2021 to 2025 name this task in their title or abstract, up 52% compared with 2016 to 2020. That ranks it 184 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 26 |
| 2024 | 26 |
| 2023 | 17 |
| 2022 | 18 |
| 2021 | 15 |
| 2020 | 11 |
| 2019 | 18 |
| 2018 | 18 |
| 2017 | 13 |
| 2016 | 7 |
| 2015 | 13 |
| 2014 | 11 |
| 2013 | 3 |
| 2012 | 8 |
| 2011 | 12 |
| 2010 | 7 |
| 2009 | 4 |
| 2008 | 7 |
| 2007 | 3 |
| 2006 | 2 |
| 2005 | 0 |
| 2004 | 0 |
| 2003 | 1 |
| 2002 | 3 |
| 2001 | 2 |
| 2000 | 1 |
Use grade C
100 to 299 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
- 0
- Papers on reliability or validity
- 8
- 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.
- Tokens
- One at a time, worth 0 points until a delay ends, then gold and worth 10 points (standard, 1 to 100). A sale by button or space bar pays a gold token's points and an earlier sale nothing; the next token appears 2,000 ms after a sale.
- Parts
- Two of 5 minutes each (standard, 3 to 12), in an order drawn from the session's seed (standard) or fixed. The time runs whatever the choices.
- Delays
- Where waiting pays: spread evenly from 0 to 12 seconds. Where selling early pays: a generalized Pareto distribution with shape 5 and scale 2.4 seconds, cut at 90 seconds. Each run of 8 tokens takes one delay from each eighth of the distribution.
- Best policy
- Wait for every token where waiting pays; sell any token not ready after 2.75 seconds where selling early pays. At 10 points a token, both earn about 1.25 points per second.
- Practice
- 3 tokens with delays of 1.5, 3 and 2 seconds.
- Score
- The area under each part's Kaplan-Meier survival curve of waiting times over the first 12 seconds. A token that turned gold, or was still waiting when the part ended, counts as waited at least that long.
- Points
- Points only (standard), or paid when the study pays for them.
This version follows the design of McGuire and Kable (2012, 2015) with its own delays, chosen so the two best policies earn about the same points per second, and 5 minute parts as standard. Participants are not told how the delays differ. 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: willingness_to_wait_token from the scored run, one per token. Waiting or selling has no correct answer, so neither accuracy nor a repeated choice is judged. A part with no tokens invalidates. Sales within 100 ms of a token appearing on at least half the sales invalidate, and any is flagged. No points in the whole task, a median of more than 2 seconds from a token turning gold to its sale, and a focus loss are flagged.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. Calibration compares the two parts, so a part with no tokens leaves nothing to compare. A sale within 100 ms of a token appearing cannot follow from the screen, so at least half the sales that fast means the button or key was pressed without looking. Selling every token before it was ready earns nothing, a ready token left unsold wastes time the delays do not account for, and a part's clock keeps running when the page loses focus, so each is flagged.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Minutes per part | 5 | 3 to 12 |
| Points per token | 10 | 1 to 100 |
| Order of parts | Drawn from the session's seed (standard) | Drawn from the session's seed (standard), Waiting pays first, Selling early pays first |
| Payment line | No mention of money (standard) | No mention of money (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 Willingness to Wait Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; McGuire & Kable, 2012; McGuire & Kable, 2015), with the standard settings (minutes per part: 5; points per token: 10; order of parts: Drawn from the session's seed; payment line: No mention of money). The main score was willingness to wait difference (willingness to wait where waiting pays minus willingness to wait where selling early pays, out of the first 12 seconds).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- McGuire, J. T., & Kable, J. W. (2012). Decision makers calibrate behavioral persistence on the basis of time-interval experience. Cognition, 124(2), 216-226. https://doi.org/10.1016/j.cognition.2012.03.008
- McGuire, J. T., & Kable, J. W. (2013). Rational temporal predictions can underlie apparent failures to delay gratification. Psychological Review, 120(2), 395-410. https://doi.org/10.1037/a0031910Why giving up on a delayed reward can be rational.
- McGuire, J. T., & Kable, J. W. (2015). Medial prefrontal cortical activity reflects dynamic re-evaluation during voluntary persistence. Nature Neuroscience, 18(5), 760-766. https://doi.org/10.1038/nn.3994Two environments that favor more or less persistence.
Source papers
- McGuire and Kable (2012): persistence fitted to the timing of delays. McGuire and Kable (2015): waiting for rewards in two environments that favor more or less persistence.Paradigm source, Adults
Questions
How long does the Willingness to Wait Task take?
About 11 to 13 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?
Willingness to Wait Difference, Willingness to Wait, Waiting Pays, Willingness to Wait, Selling Early Pays, Tokens Sold Early, Waiting Pays, Tokens Sold Early, Selling Early Pays, Tokens, Waiting Pays, Tokens, Selling Early Pays, Points, Time to Sell a Gold Token and Tokens with Focus Loss. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Willingness to Wait Task used in research?
102 papers from 2021 to 2025 name it in their title or abstract, up 52% compared with 2016 to 2020. That ranks it 184 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Willingness to Wait Task have?
In the published research, 0 papers report its test-retest reliability and 4 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?
Minutes per part (3 to 12), points per token (1 to 100), order of parts (Drawn from the session's seed (standard), Waiting pays first, Selling early pays first) and payment line (No mention of money (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.