Decision Making
Demand Selection Task
Also called: DST, cognitive demand selection task, demand avoidance task, effort avoidance task
Measure avoidance of mental effort with free choices between two decks, one of which switches between two number questions far more often, scored as the share of choices from the easier deck.
9-11 minTimed: desktop or laptop with a mouse or trackpad
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
- 26
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures how much people avoid mental effort when free to choose. In each round participants pick a card from one of two decks and say whether its number is odd or even, or lower or higher than 5, as its color and question ask. One deck changes the question on 1 card in 10 and the other on 9 in 10; nothing else differs. The main score is the share of choices from the easier deck, over 4 parts of 50 rounds.
Constructs
- Cognitive effort avoidance
- Task switching
- Cost of cognitive control
- Motivation
Research uses
- Motivation and apathy research
- Schizophrenia and other clinical groups
- Brain imaging of decision costs
- Within-study group comparison
How the paradigm works
The demand selection task measures how much people avoid mental effort when they are free to choose (Kool, McGuire, Rosen and Botvinick, 2010). In each round the participant picks a card from one of two decks. The card shows a number from 1 to 9, except 5, in orange or blue: orange asks whether it is odd or even, blue whether it is lower or higher than 5, and the question is also written above the number. The left and right arrow keys answer, as in the task switching test. The decks differ in one way that is not announced: from one deck the question changes from the round before on 1 card in 10, from the other on 9 cards in 10, so keeping to it means switching between the two questions most of the time. Each of 4 parts has a new pair of decks and 50 rounds, and the easier deck is on the left in half the parts.
More about the paradigm
Practice with feedback comes first: 16 numbers on their own, then 8 rounds with practice decks. The main task gives no feedback. A setting adds a line saying that one deck changes the question more often, after Gold and colleagues (2015).
The main score is the share of choices from the easier deck. The share in the first and second half of each part, accuracy with each deck, the switch cost in response time and accuracy, and choice time are also given. In the source's task switching versions, the easier deck took 61% to 68% of choices on average, and participants who did not notice the difference avoided the harder deck too (Kool and colleagues, 2010). Larger switch costs went with stronger avoidance (Kool and colleagues, 2010, Experiment 5). In 811 people tested online, the easier deck took 65% of choices on average (Patzelt, Kool, Millner and Gershman, 2019). In people with schizophrenia and community volunteers, avoidance appeared only when the instructions said the decks differ, and it was weaker in the patients (Gold and colleagues, 2015). Scores support research and do not diagnose.
Scores
- Easier Deck Chosen%
- Share of choices from the deck whose question changes on 1 card in 10.
- Higher means more avoidance of mental effort
- Easier Deck, First Half%
- The same share over the first half of each part's rounds.
- Higher means more avoidance of mental effort
- Easier Deck, Second Half%
- The same share over the second half of each part's rounds.
- Higher means more avoidance of mental effort
- Accuracy%
- Share of numbers answered correctly.
- Higher is better
- Accuracy, Easier Deck%
- Share of numbers answered correctly on cards from the easier deck.
- Higher is better
- Accuracy, Harder Deck%
- Share of numbers answered correctly on cards from the harder deck.
- Higher is better
- Switch Costms
- Median response time on rounds where the question changed minus rounds where it stayed the same, over correct answers that follow a correct answer, leaving out each part's first round.
- Higher means switching took longer
- Switch Cost in Accuracyscore
- Accuracy on rounds where the question stayed the same minus where it changed, from -1 to 1.
- Higher means more errors after a change
- Response Timems
- Median time from the number appearing to a correct answer.
- Lower is faster
- Question Changes%
- Share of rounds, after each part's first, on which the question changed. Picking the easier deck lowers it.
- Informational
- Choice Timems
- Median time from the decks appearing to the choice of a deck.
- Informational
- Same Deck Again%
- Share of choices, after each part's first, of the same deck as the round before.
- Informational
- Same Screen Side%
- Share of choices on the screen side picked more often. The easier deck changes side between parts.
- Near 1 suggests the decks were not told apart
- Early Pressescount
- Presses under 100 ms after the decks or the number appeared, not taken as choices or answers.
- Lower is better
- Roundscount
- Scored rounds played.
- Informational
Use and evidence
26 papers from 2021 to 2025 name this task in their title or abstract, up 73% compared with 2016 to 2020. That ranks it 311 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 6 |
| 2024 | 7 |
| 2023 | 5 |
| 2022 | 6 |
| 2021 | 2 |
| 2020 | 5 |
| 2019 | 3 |
| 2018 | 4 |
| 2017 | 1 |
| 2016 | 2 |
| 2015 | 0 |
| 2014 | 1 |
| 2013 | 0 |
| 2012 | 0 |
| 2011 | 0 |
| 2010 | 1 |
| 2009 | 0 |
| 2008 | 0 |
| 2007 | 0 |
| 2006 | 0 |
| 2005 | 0 |
| 2004 | 0 |
| 2003 | 0 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 0 |
Use grade E
Fewer than 30 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 0
- Registered clinical studies using it as an outcome
- 0
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.
- Decks
- Two decks drawn as patterned cards, a new pair in each part. From one deck the question changes from the round before on 1 card in 10, from the other on 9 in 10; nothing else differs, and the difference is not announced.
- Cards
- A number from 1 to 9, except 5. Orange asks odd or even, blue asks lower or higher than 5, and the question is written above the number. The left and right arrow keys answer, as in the task switching test.
- Choice
- Click a deck or press the left or right arrow key, with no time limit for the choice or the answer. Presses under 100 ms are not taken.
- Parts
- 4 parts (standard, 2 to 8) of 50 rounds (standard, 40 to 75). The easier deck is on the left in half the parts, in an order from the session's seed.
- Practice
- 16 numbers on their own, then 8 rounds with practice decks, each answer followed by 800 ms of feedback. The main task gives none.
- Instructions
- The standard says that either deck may be picked. A setting adds that one deck changes the question more often.
The sources answered aloud, with mouse buttons or with keys, and showed the question by the number's color alone. This test also writes the question and uses its own decks and numbers of rounds, so compare its shares within a study. 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: demand_selection_trial from the scored run, one per round. The deck chosen has no correct answer, so it is not judged; the number's answer is a choice of two keys judged against chance. Accuracy significantly below chance invalidates and accuracy not above chance is flagged, as is one key on 90% of answers when accuracy is not above chance (invalid then). Presses within 100 ms of the decks or the number are not taken: more than max(2, 20%) of rounds with one invalidate, and any is flagged. A run with fewer than half its planned rounds is invalid and any other shortfall is flagged. Picking decks on one side of the screen on at least 90% of at least 40 rounds over two or more parts is flagged.
Why these rules
Choosing the easier or the harder deck throughout is a preference, so deck choices are not judged against chance. The demand avoided is that of switching between the two questions, so the answers must follow the questions: Accuracy significantly below chance (one-sided exact test, p < .05) cannot come from guessing; it means the response mapping was reversed or the task was not followed (Slick, Sherman and Iverson 1999 treat below-chance forced-choice performance as definite response bias). Chance-level accuracy can be a real result in impaired participants, so it is only flagged. One option on at least 90% of answered trials, when the correct options are balanced and accuracy is not above chance, means the answers did not depend on the items (the task counterpart of long-string responding in surveys; Curran 2016). A press within 100 ms cannot be a response to the decks or the number, so such presses follow the platform's timed-test limits. The easier deck changes side between parts, so picking one side almost every time means the decks were not told apart.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Parts | 4 | 2 to 8 |
| Rounds per part | 50 | 40 to 75 |
| What the instructions say about the decks | No word on a difference (standard) | No word on a difference (standard), Say that one deck changes the question more often |
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
- Timed test: participants need a desktop or laptop with a mouse or trackpad. 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 Demand Selection Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Kool et al., 2010), with the standard settings (parts: 4; rounds per part: 50; what the instructions say about the decks: No word on a difference). The main score was easier deck chosen (share of choices from the deck whose question changes on 1 card in 10).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Kool, W., McGuire, J. T., Rosen, Z. B., & Botvinick, M. M. (2010). Decision making and the avoidance of cognitive demand. Journal of Experimental Psychology: General, 139(4), 665-682. https://doi.org/10.1037/a0020198The task: decks that switch the question on 10% or 90% of cards, and a preference for the easier deck.
- McGuire, J. T., & Botvinick, M. M. (2010). Prefrontal cortex, cognitive control, and the registration of decision costs. Proceedings of the National Academy of Sciences, 107(17), 7922-7926. https://doi.org/10.1073/pnas.0910662107Lateral prefrontal activity was higher in people who more strongly avoided demanding choices.
- Gold, J. M., Kool, W., Botvinick, M. M., Hubzin, L., August, S., & Waltz, J. A. (2015). Cognitive effort avoidance and detection in people with schizophrenia. Cognitive, Affective, & Behavioral Neuroscience, 15(1), 145-154. https://doi.org/10.3758/s13415-014-0308-5Avoidance in community volunteers only when the instructions said the decks differ, and less of it in schizophrenia.
- Patzelt, E. H., Kool, W., Millner, A. J., & Gershman, S. J. (2019). The transdiagnostic structure of mental effort avoidance. Scientific Reports, 9, 1689. https://doi.org/10.1038/s41598-018-37802-1An online version in 811 people, and how avoidance related to psychiatric symptoms.
Source papers
- Kool et al. (2010): the task, 61% to 68% easier-deck choices in its task switching versions. Patzelt et al. (2019): 65% in 811 people online. Gold et al. (2015): schizophrenia and community volunteers, with and without saying the decks differ.Paradigm source, Adults
Questions
How long does the Demand Selection Task take?
About 9 to 11 minutes, as listed in the test library.
What device does a participant need?
Timed test: participants need a desktop or laptop with a mouse or trackpad. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Easier Deck Chosen, Easier Deck, First Half, Easier Deck, Second Half, Accuracy, Accuracy, Easier Deck, Accuracy, Harder Deck, Switch Cost, Switch Cost in Accuracy, Response Time, Question Changes, Choice Time, Same Deck Again, Same Screen Side, Early Presses and Rounds. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Demand Selection Task used in research?
26 papers from 2021 to 2025 name it in their title or abstract, up 73% compared with 2016 to 2020. That ranks it 311 of 395 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Demand Selection 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 settings can a researcher change?
Parts (2 to 8), rounds per part (40 to 75) and what the instructions say about the decks (No word on a difference (standard), Say that one deck changes the question more often). 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.