Social Cognition
Advice Taking Task
Also called: judge-advisor system, JAS, weight of advice, advice utilization, egocentric discounting
Measure how far people revise their own estimate toward advice: participants estimate the dots in a picture, see a simulated adviser's estimate and give a final one, scored as the weight of advice.
6-9 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
- 174
- Evidence grade
- C
- Registered clinical studies
- 1
What it measures
Measures how much people revise their own estimate toward advice. Pictures of dots are shown for 2 seconds; the participant estimates the number, sees a simulated adviser's estimate and gives a final estimate. The main score is the weight of advice, from 0 (advice ignored) to 1 (advice adopted).
Constructs
- Advice taking
- Social influence
- Egocentric discounting
- Judgment revision
Research uses
- Judgment and decision making research
- Social influence and trust in advice
- Advice from people and from algorithms
- Within-study group comparison
How the paradigm works
The Advice Taking Task measures how far people revise their own judgment toward someone else's, in the judge-advisor system (Sniezek and Buckley, 1995). The participant makes an estimate, sees an adviser's estimate for the same question, then gives a final estimate. The weight of advice is the share of the distance to the advice that the final estimate covers: 0 keeps the first estimate, 1 takes the advice, and 0.5 averages the two (Harvey and Fischer, 1997; Yaniv, 2004). People typically move about 30% of the way, though averaging would often make them more accurate, a pattern called egocentric discounting (Yaniv and Kleinberger, 2000). Across 129 data sets, people moved 32%, 37% and 48% of the way toward advisers described as low, neutral and high in quality (Bailey, Leon, Ebner, Moustafa and Weidemann, 2023).
More about the paradigm
Here each estimate is the number of dots in a picture shown for 2 seconds, too briefly to count, as with the pictures of animals in Molleman, Kanngiesser and van den Bos (2019). The true number is known by construction and needs no general knowledge, units or language. A session has 20 pictures (10 to 30 as a setting) of about 40 to 120 dots, after 2 practice pictures that show the true number. The adviser is simulated: its estimate is the participant's first estimate moved toward the true number by 15%, 20% or 25%, as in Molleman and colleagues, so the advice always points the right way but is not always closer. A first estimate below a quarter or above four times the number of dots is taken for a slip: the advice is then the true number give or take 15%, and the picture is left out, as Schultze, Rakotoarisoa and Schulz-Hardt (2015) did. By default the instructions say the adviser is a computer program that follows set rules; a neutral wording leaves that line out. One check question with feedback comes first, and no accuracy is shown during the task. At the end one estimate, first or final, is drawn and earns 100 points minus 1 for each dot it is off. The study decides what points are worth, and the payment line says whether they are exchanged for money.
The main score is the weight of advice: the mean over pictures of the final estimate minus the first, over the advice minus the first, truncated to 0 and 1 (Schultze and colleagues, 2015). The shares of pictures that keep the first estimate, take the advice or land between the two separate staying, copying and compromising (Molleman and colleagues, 2019). The median errors of the first and final estimates, as a percent of the number of dots, show how much the advice helped. Estimates under 100 ms are counted as anticipations and left out. Scores support research and do not diagnose.
Scores
- Weight of Advice%
- Mean over pictures of the final estimate minus the first, over the advice minus the first, truncated to 0 and 1.
- 0 keeps the first estimate, 1 takes the advice
- Kept First Estimate%
- Share of pictures whose final estimate equals the first estimate.
- Informational
- Took the Advice%
- Share of pictures whose final estimate equals the advice.
- Informational
- Between the Two%
- Share of pictures whose final estimate lies strictly between the first estimate and the advice.
- Informational
- First Estimate Error%
- Median distance of the first estimates from the number of dots, as a percent of it.
- Lower is closer
- Final Estimate Error%
- Median distance of the final estimates from the number of dots, as a percent of it.
- Lower is closer
- Error Reduction%
- First estimate error minus final estimate error, in percentage points.
- Above 0 means the advice helped
- Median Revision Timems
- Median time from the advice appearing to the final estimate.
- Informational
- Anticipationscount
- Pictures with an estimate submitted under 100 ms after its screen appeared, left out of the scores.
- 0 expected
- Slipscount
- Pictures whose first estimate was under a quarter or over four times the number of dots; their advice was not built from the estimate, and they are left out.
- More than a quarter of the pictures is flagged
- Pictures Answeredtrials
- Pictures with both estimates recorded, out of 20 as standard.
- Informational
- Check Correctflag
- 1 when the check question about which estimates earn points was answered correctly, 0 when not.
- 1 supports interpretation
- Points Wonpoints
- Points from the drawn estimate: 100 minus 1 for each dot it is off, at least 0.
- Informational
Use and evidence
174 papers from 2021 to 2025 name this task in their title or abstract, up 39% compared with 2016 to 2020. That ranks it 134 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 60 |
| 2024 | 41 |
| 2023 | 26 |
| 2022 | 27 |
| 2021 | 20 |
| 2020 | 29 |
| 2019 | 34 |
| 2018 | 21 |
| 2017 | 29 |
| 2016 | 12 |
| 2015 | 20 |
| 2014 | 18 |
| 2013 | 21 |
| 2012 | 20 |
| 2011 | 8 |
| 2010 | 10 |
| 2009 | 10 |
| 2008 | 6 |
| 2007 | 7 |
| 2006 | 5 |
| 2005 | 2 |
| 2004 | 5 |
| 2003 | 2 |
| 2002 | 1 |
| 2001 | 4 |
| 2000 | 2 |
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
- 4
- Papers on reliability or validity
- 47
- Registered clinical studies using it as an outcome
- 1
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.
- Pictures
- 20 (standard; 10 to 30), from about 40 to about 120 dots, evenly spread on a ratio scale and drawn by our code. The dots show for 2 seconds, too briefly to count, in a circle 300 pixels across.
- Estimates
- A typed whole number from 1 to 999, then the adviser's estimate beside the participant's, then a typed final estimate. No time limit for typing and no feedback during the task.
- Adviser
- Simulated: the first estimate moved toward the true number by 15%, 20% or 25%, about a third of the pictures each, as in Molleman and colleagues (2019). The standard instructions say the adviser is a computer program that follows set rules; a neutral wording leaves that line out.
- Slips
- A first estimate under a quarter or over 4 times the number of dots gets advice of the true number times 0.85 to 1.15, and that picture is left out of the scores.
- Practice
- 2 pictures (30 and 70 dots), each followed by the true number, then one check question with feedback: both estimates can earn points.
- Points
- One estimate, first or final, drawn from the session's seed earns 100 points minus 1 for each dot it is off, never below 0. The study decides what points are worth, and the payment line says whether they are exchanged for money.
- Scores
- The weight of advice: the mean of the final estimate minus the first, over the advice minus the first, truncated to 0 and 1. Also the shares of pictures that kept the first estimate, took the advice or landed between them, and the median errors of the first and final estimates. Estimates under 100 ms are left out as anticipations.
- Checks
- Fewer than half the pictures answered is invalid. Slips on more than a quarter of the pictures, no weight of advice and a wrong check answer are flagged and never exclude.
Many studies ask for estimates such as distances, weights or calories and give advice from other people or a named source. Here the estimates are of dot pictures whose true number the code knows, and the adviser is a rule, as with the briefly shown pictures and selected advice of Molleman and colleagues (2019). The weight of advice depends on how the adviser is described and how far the advice is from the first estimate (Bailey and colleagues, 2023; Schultze and colleagues, 2015), so compare people who saw the same instructions. 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 rounds, each a first estimate, the advice and a final estimate. No weight of advice is correct, so accuracy is not judged against chance, and keeping every first estimate or taking every piece of advice does not invalidate. Fewer than half the planned rounds answered is invalid, and any other incomplete run is flagged. A first estimate under a quarter or over 4 times the number of dots is a slip, and its round is left out; slips on more than a quarter of the rounds, no weight of advice and a wrong check answer are flagged for review and never exclude. Estimates under 100 ms are counted as anticipations and left out, and final estimates in under 200 ms on at least half the rounds invalidate.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. The weight of advice is how far a person moves toward another estimate, which has no right value: taking more advice is not better or worse in itself. A first estimate far outside the picture's range is most often a typing slip, and advice built from it would look strange, so its round gets plausible advice and is left out, as Schultze, Rakotoarisoa and Schulz-Hardt (2015) did; many slips can still be a real way of estimating, so they flag and never exclude. 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 |
|---|---|---|
| Pictures | 20 | 10 to 30 |
| Adviser wording | A computer program (standard) | A computer program (standard), Neutral: the adviser |
| 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 Advice Taking Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Sniezek JA, Buckley T, 1995), with the standard settings (pictures: 20; adviser wording: A computer program; payment line: Hypothetical points). The main score was weight of advice (mean over pictures of the final estimate minus the first, over the advice minus the first, truncated to 0 and 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.
- Sniezek, J. A., & Buckley, T. (1995). Cueing and cognitive conflict in judge-advisor decision making. Organizational Behavior and Human Decision Processes, 62(2), 159-174. https://doi.org/10.1006/obhd.1995.1040
- Yaniv, I. (2004). Receiving other people's advice: Influence and benefit. Organizational Behavior and Human Decision Processes, 93(1), 1-13. https://doi.org/10.1016/j.obhdp.2003.08.002The weight of advice.
- Molleman, L., Kanngiesser, P., & van den Bos, W. (2019). Social information use in adolescents: The impact of adults, peers and household composition. PLOS ONE, 14(11), e0225498. https://doi.org/10.1371/journal.pone.0225498Briefly shown pictures, advice set from the first estimate, and keeping, copying and compromising.
- Schultze, T., Rakotoarisoa, A.-F., & Schulz-Hardt, S. (2015). Effects of distance between initial estimates and advice on advice utilization. Judgment and Decision Making, 10(2), 144-171. https://doi.org/10.1017/S1930297500003922Advice built from the first estimate, slips and the truncated weight.
- Bailey, P. E., Leon, T., Ebner, N. C., Moustafa, A. A., & Weidemann, G. (2023). A meta-analysis of the weight of advice in decision-making. Current Psychology, 42(28), 24516-24541. https://doi.org/10.1007/s12144-022-03573-2Pooled weights of advice and what changes them.
Source papers
- Bailey, Leon, Ebner, Moustafa and Weidemann (2023): across 129 data sets (N = 17,296), people moved 32%, 37% and 48% of the way toward advisers described as low, neutral and high in quality. Molleman, Kanngiesser and van den Bos (2019): adolescents estimating animals in briefly shown pictures kept their estimate on 27% of rounds, copied the advice on 18% and compromised on 51%.Paradigm source, Adolescents and adults
Questions
How long does the Advice Taking Task take?
About 6 to 9 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?
Weight of Advice, Kept First Estimate, Took the Advice, Between the Two, First Estimate Error, Final Estimate Error, Error Reduction, Median Revision Time, Anticipations, Slips, Pictures Answered, Check Correct and Points Won. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Advice Taking Task used in research?
174 papers from 2021 to 2025 name it in their title or abstract, up 39% compared with 2016 to 2020. That ranks it 134 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Advice Taking Task have?
In the published research, 4 papers report its test-retest reliability and 1 registered clinical study uses 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?
Pictures (10 to 30), adviser wording (A computer program (standard), Neutral: the adviser) 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.
Related tests