Decision Making
Decoy Choice Task
Also called: decoy effect task, attraction effect task, asymmetric dominance task, compromise effect task, context effects in choice
Measure how a third option moves a choice between two others, with price and quality pairs shown alone and beside a decoy or an extreme option, scored as the decoy effect.
5-7 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 251
- Evidence grade
- C
- Registered clinical studies
- 0
What it measures
Measures the decoy effect: how a third option changes a choice between two others. Participants choose which version of a pictured everyday thing they would buy, each version with a price in points and a quality score out of 100, in 40 choices over 8 kinds of thing. Each kind's pair is shown alone and beside a third version that one of the pair beats on both scores (the attraction effect) or that makes one of the pair the middle choice (the compromise effect). The main score is the change in that version's share of the choices.
Constructs
- Context effects in choice
- Attraction effect
- Compromise effect
- Multi-attribute decision-making
Research uses
- Judgment and decision-making research
- Consumer choice research
- Individual differences in context effects
- Within-study group comparison
How the paradigm works
The decoy choice task measures the decoy effect: a choice between two options moves when a third is added. Huber, Payne and Puto (1982) added an option that one of two others beats on both of its attributes, and that option's share of the choices rose, the attraction effect. Simonson (1989) found that an option gains share when a third makes it the middle choice, the compromise effect. Each of 8 kinds of everyday thing, pictured with Noto Emoji, has a pair of versions that trade a price in points against a quality score out of 100: one cheaper and lower in quality, one dearer and higher. The pair is shown alone, beside a decoy that one of the two beats on both scores, and beside an extreme version that makes one of the two the middle choice, each third version once for each of the pair: 40 choices in five rounds, each kind once a round. The numbers follow a rule: each kind's versions lie on a line of equal steps, a decoy is a quarter of a step worse than its target on both scores, an extreme version is a full step beyond it, and the 8 kinds trade quality for points at rates from 0.5 to 2. Every screen moves all its numbers by a small offset, so the same numbers do not come back, and the order of the screens and the places of the versions come from the session's seed.
More about the paradigm
One check question on reading the scores, with feedback, and two practice choices come first. There is no time limit, nothing is bought, and the points are not exchanged for money.
The main score is the decoy effect: on each screen with a third version, whether the version it favors was chosen, minus whether that version was chosen when its pair was alone, averaged over the screens, from -1 to 1. With every screen answered and no third version chosen, it is the favored version's share of the pair's choices minus one half, the relative share Berkowitsch, Scheibehenne and Rieskamp (2014) used to find both effects within the same people. The attraction and compromise effects are also given on their own, with the effect when the favored version is the dearer or the cheaper one, how often the decoy and the extreme version were chosen, and how often the dearer version was chosen from the pair alone. The decoy should almost never be chosen (about 2% in Huber and colleagues' studies; Huber, Payne and Puto, 2014). The effect is found with attributes shown as numbers and is weak or absent when an attribute is shown as a picture (Frederick, Lee and Baskin, 2014), so here a picture only names the kind and the choice rests on the numbers. Scores describe how choices moved in this task; they are research measures and do not diagnose.
Scores
- Decoy Effectscore
- On screens with a third version: whether the version it favors was chosen, minus whether that version was chosen when its pair was alone, averaged; from -1 to 1. Choices of the third version are left out.
- Positive means a third version drew choices to the version it favors; 0 means no change
- Attraction Effectscore
- The decoy effect on the screens with a decoy, a version that one of the pair beats on both price and quality.
- Positive is the classic attraction effect
- Compromise Effectscore
- The decoy effect on the screens with an extreme version, which makes one of the pair the middle choice.
- Positive is the classic compromise effect
- Effect, Dearer Version Favoredscore
- The decoy effect on the screens where the third version favors the pair's dearer, higher-quality version.
- Informational
- Effect, Cheaper Version Favoredscore
- The decoy effect on the screens where the third version favors the pair's cheaper, lower-quality version.
- Informational
- Decoy Chosen%
- Share of the decoy screens where the decoy, beaten on both price and quality, was chosen.
- Near 0 expected; higher values suggest the scores were not compared
- Extreme Version Chosen%
- Share of the compromise screens where the extreme version was chosen.
- Informational
- Dearer Version Chosen Alone%
- Share of the pairs shown alone where the dearer, higher-quality version was chosen.
- Informational: how quality was weighed against price
- Check Correct0/1
- 1 when the check question on reading the scores was answered correctly.
- 1 supports interpretation
- Median Choice Timems
- Median time from a screen appearing to the choice.
- Informational
- Clicks Under 100 mscount
- Clicks within 100 ms of a screen appearing, which are not taken as choices.
- 0 expected
- Choicescount
- Choices made, out of 40.
- Informational
Use and evidence
251 papers from 2021 to 2025 name this task in their title or abstract, up 5% compared with 2016 to 2020. That ranks it 111 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 83 |
| 2024 | 47 |
| 2023 | 41 |
| 2022 | 37 |
| 2021 | 43 |
| 2020 | 53 |
| 2019 | 62 |
| 2018 | 44 |
| 2017 | 44 |
| 2016 | 35 |
| 2015 | 44 |
| 2014 | 35 |
| 2013 | 20 |
| 2012 | 34 |
| 2011 | 30 |
| 2010 | 23 |
| 2009 | 13 |
| 2008 | 26 |
| 2007 | 30 |
| 2006 | 19 |
| 2005 | 15 |
| 2004 | 9 |
| 2003 | 8 |
| 2002 | 8 |
| 2001 | 4 |
| 2000 | 8 |
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
- 34
- Registered clinical studies using it as an outcome
- 0
Note on the counts: includes marketing papers.
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 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Kinds
- 8 everyday things, each a pair of versions: one cheaper and lower in quality, one dearer and higher. Each version shows a price in points and a quality score out of 100.
- Screens
- Each pair alone, beside a decoy that one of the pair beats on both scores, and beside an extreme version that makes one of the pair the middle choice, each third version once for each of the pair: 40 choices in 5 rounds, each kind once a round.
- Numbers
- A kind's versions lie on a line of equal steps in price and quality. A decoy is a quarter of a step worse than its target on both scores; an extreme version is a full step beyond it. The 8 kinds trade quality for points at rates from 0.5 to 2, two of them at 1.
- Order
- Each screen moves all its numbers by one offset from -3 to 3 on each score, different on each of a kind's screens. The order of the screens and the places of the versions come from the session's seed.
- Practice
- One check question on reading the scores, with feedback, then two practice choices. No time limit.
- Points
- Nothing is bought, and the points are not exchanged for money.
- Scores
- The decoy effect: whether the favored version was chosen beside its third version, minus whether it was chosen from the pair alone. Also the attraction and compromise effects, and how often the decoy was chosen.
Huber, Payne and Puto (1982) and Simonson (1989) compared groups who saw different choice sets. This test shows every set to each person, as Berkowitsch, Scheibehenne and Rieskamp (2014) did with pairs matched to each person beforehand; here the pairs are fixed, and their rates of trade are spread so that most people are close to undecided on some of them. The effect is found with attributes shown as numbers (Frederick, Lee and Baskin, 2014), so the pictures only name the kind. One person's score rests on 32 screens, so compare groups. 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: decoy_choice_choice from the scored run, one per screen. A choice has no correct answer, so accuracy is not judged against chance. Choices under 200 ms on at least half the screens invalidate. Fewer than half the planned screens invalidate, and any missing screen is flagged. The option beaten on both price and quality chosen on 25% or more of the screens that show one, the same place on the screen chosen on 90% or more of the screens, and a wrong answer to the check question are flagged.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. 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. The decoy effect compares each pair alone with the same pair beside a third option, so a run that stopped early is not comparable. A decoy that is beaten on both scores should almost never be chosen (Huber, Payne and Puto, 2014, put its share at about 2% in their studies), so choosing it often, or always choosing the same place on the screen, means the scores may not have been compared; both are flagged and never exclude.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
This test runs with fixed settings.
Need a different version of this test?
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 Decoy Choice Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Huber, Payne & Puto, 1982; Simonson, 1989). The main score was decoy effect (on screens with a third version: whether the version it favors was chosen, minus whether that version was chosen when its pair was alone, averaged; from -1 to 1).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Huber, J., Payne, J. W., & Puto, C. (1982). Adding asymmetrically dominated alternatives: Violations of regularity and the similarity hypothesis. Journal of Consumer Research, 9(1), 90-98. https://doi.org/10.1086/208899
- Simonson, I. (1989). Choice based on reasons: The case of attraction and compromise effects. Journal of Consumer Research, 16(2), 158-174. https://doi.org/10.1086/209205
- Berkowitsch, N. A. J., Scheibehenne, B., & Rieskamp, J. (2014). Rigorously testing multialternative decision field theory against random utility models. Journal of Experimental Psychology: General, 143(3), 1331-1348. https://doi.org/10.1037/a0035159Both effects within each person, the relative share of the target, and the place of the extreme option.
- Huber, J., Payne, J. W., & Puto, C. P. (2014). Let's be honest about the attraction effect. Journal of Marketing Research, 51(4), 520-525. https://doi.org/10.1509/jmr.14.0208Why the decoy should almost never be chosen, and when the effect weakens.
- Frederick, S., Lee, L., & Baskin, E. (2014). The limits of attraction. Journal of Marketing Research, 51(4), 487-507. https://doi.org/10.1509/jmr.12.0061The effect with numbers and its absence with pictured attributes.
Source papers
- Berkowitsch, Scheibehenne and Rieskamp (2014): attraction and compromise effects within the same people choosing consumer products, relative target shares of .63 and .58 (48 participants).Paradigm source, Adults
Questions
How long does the Decoy Choice Task take?
About 5 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?
Decoy Effect, Attraction Effect, Compromise Effect, Effect, Dearer Version Favored, Effect, Cheaper Version Favored, Decoy Chosen, Extreme Version Chosen, Dearer Version Chosen Alone, Check Correct, Median Choice Time, Clicks Under 100 ms and Choices. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Decoy Choice Task used in research?
251 papers from 2021 to 2025 name it in their title or abstract, up 5% compared with 2016 to 2020. That ranks it 111 of 495 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Decoy Choice 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 C in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
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.