Executive Function
Multi-Source Interference Task
Also called: MSIT, multisource interference task, multi-source interference test, Bush MSIT, digit interference task
Measure interference control: how much slower people name the digit that differs when its place and the other digits point to a wrong answer.
7-8 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
- 63
- Evidence grade
- C
- Registered clinical studies
- 32
What it measures
Measures interference from several sources at once. Three digits appear and one differs from the other two; participants press 1, 2 or 3 for its value. In control blocks it sits in its own place among zeros, and in interference blocks it sits elsewhere among other possible answers. The score is the slowing in interference blocks.
Constructs
- Interference control
- Selective attention
- Response selection
Research uses
- Cognitive control research with brain imaging
- Psychiatric research, such as ADHD and posttraumatic stress
- Treatment and intervention studies
How the paradigm works
The Multi-Source Interference Task measures how well a person answers one feature of a display while other features pull toward a different answer. Three digits from 0 to 3 appear in a row, and one differs from the other two. Participants press the key for the value of the different digit, 1, 2 or 3, on keys that run from left to right. On control trials the different digit sits in its own place (1 on the left, 2 in the middle, 3 on the right) between two zeros, which are never an answer. On interference trials it sits in another place, between two copies of another possible answer, so its place and its neighbors both point to a wrong key. Bush and colleagues (2003) built the task to combine Stroop, flanker and Simon interference in one display, and it is widely used with brain imaging.
More about the paradigm
The standard session follows the block design of Bush and Shin (2006): blocks of 24 trials alternate, a control block first, 4 of each, for 192 trials. Each trial lasts 1,750 ms whatever the answer: the digits show for 1,250 ms, then a small dot until the next display, and an answer counts until the trial ends. Each block lasts 42 seconds and the run under 6 minutes, after 12 practice trials with feedback (6 control, then 6 interference). Desktop and laptop only, with the 1, 2 and 3 keys on the row above the letters or on the number pad.
The main score is the interference effect: mean correct response time on interference trials minus control trials, in ms; lower means less slowing. Bush and colleagues (2003) report a mean of 312 ms in their volunteers, and Bush and Shin (2006) a range of 200 to 350 ms. The error interference effect (control accuracy minus interference accuracy), the condition means and accuracies and the difference of the medians are also reported. Responses under 100 ms count as anticipations. Scores support research and do not diagnose.
Scores
- Interference Effectms
- Mean correct response time on interference trials minus control trials.
- Lower means less slowing under interference; positive values are typical
- Control RTms
- Mean correct response time on control trials, the different digit in its own place among zeros.
- Informational
- Interference RTms
- Mean correct response time on interference trials, the different digit elsewhere among other possible answers.
- Informational
- Interference Effect (Medians)ms
- Median correct response time on interference trials minus control trials, less sensitive to slow outliers.
- Lower means less slowing
- Accuracy Cost%
- Control accuracy minus interference accuracy.
- Lower means fewer extra errors under interference
- Accuracy%
- Share of trials answered with the right key; a trial with no answer counts as wrong.
- Higher is better
- Control Accuracy%
- Share of control trials answered right.
- Informational
- Interference Accuracy%
- Share of interference trials answered right.
- Informational
- Mean Correct RTms
- Mean response time over all correct trials.
- Lower is faster
- Control Trials Usedtrials
- Correct control trials at or after 100 ms that the effect uses.
- Fewer than 20 in either condition is flagged for review
- Interference Trials Usedtrials
- Correct interference trials at or after 100 ms that the effect uses.
- Fewer than 20 in either condition is flagged for review
- Anticipationscount
- Answers within 100 ms of the digits.
- Many invalidate the session
- Missed Trialscount
- Trials with no answer within 1,750 ms.
- Many invalidate the session
- Total Trialstrials
- Scored trials in the run.
- Informational
Use and evidence
63 papers from 2021 to 2025 name this task in their title or abstract, up 12% compared with 2016 to 2020. That ranks it 237 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 10 |
| 2024 | 12 |
| 2023 | 9 |
| 2022 | 14 |
| 2021 | 18 |
| 2020 | 11 |
| 2019 | 12 |
| 2018 | 11 |
| 2017 | 13 |
| 2016 | 9 |
| 2015 | 4 |
| 2014 | 15 |
| 2013 | 7 |
| 2012 | 14 |
| 2011 | 4 |
| 2010 | 4 |
| 2009 | 3 |
| 2008 | 3 |
| 2007 | 6 |
| 2006 | 3 |
| 2005 | 1 |
| 2004 | 1 |
| 2003 | 1 |
| 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
- 18
- Registered clinical studies using it as an outcome
- 32
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.
- Display
- Three digits from 0 to 3 in a row on a dark panel, one different from the other two, in 56 px type.
- Keys
- 1, 2 or 3 for the value of the different digit, on the row above the letters or the number pad.
- Trial types
- Control: the digit in its own place between two zeros (100, 020, 003). Interference: in another place between two copies of another answer, all 12 such displays.
- Timing
- Each trial lasts 1,750 ms: the digits for 1,250 ms, then a dot. An answer counts until the trial ends.
- Blocks
- 24 trials per block, control and interference in turn, control first: 4 pairs (standard; 2 to 6), 192 trials.
- Practice
- 6 control trials, then 6 interference trials, with feedback for 800 ms.
- Score
- Mean correct response time on interference trials minus control trials. Answers under 100 ms count as anticipations and leave the means.
The run keeps the published block design: blocks of 42 seconds at a fixed pace with no rest between them. All three digits share one size, and answers come from the keyboard rather than a button box.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: msit_test from the scored run, the 1, 2 or 3 key (chance 33%, each a third of the answers) within 1,750 ms, correctness from the recorded target digit. Below chance invalidates and not above chance is flagged. One key on 90% of answers invalidates when accuracy is not above chance. Missed responses and responses under 100 ms follow the timed-test rule. Fewer than 20 correct trials in either condition is flagged and withholds the percentile.
Why these rules
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). Missed responses and responses under 100 ms keep the platform's timed-test rule: more than max(3, 25%) missed responses or more than max(2, 20%) responses under 100 ms invalidate, and any is flagged. A response under 100 ms cannot be a reaction to the stimulus. The effect is a difference between two means, so it needs enough correct trials in each condition; the limit of 20 is a ConductScience choice.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Pairs of blocks | 4 | 2 to 6 |
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.
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Multi-Source Interference Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Bush, Shin, Holmes, Rosen and Vogt, 2003; Bush and Shin, 2006), with the standard settings (pairs of blocks: 4). The main score was interference effect (mean correct response time on interference trials minus control trials).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Bush, G., Shin, L. M., Holmes, J., Rosen, B. R., & Vogt, B. A. (2003). The Multi-Source Interference Task: Validation study with fMRI in individual subjects. Molecular Psychiatry, 8(1), 60-70. https://doi.org/10.1038/sj.mp.4001217The task, combining Stroop, flanker and Simon interference, and its interference effect of 312 ms on average.
- Bush, G., & Shin, L. M. (2006). The Multi-Source Interference Task: An fMRI task that reliably activates the cingulo-frontal-parietal cognitive/attention network. Nature Protocols, 1(1), 308-313. https://doi.org/10.1038/nprot.2006.48The protocol, with an interference effect of 200 to 350 ms.
- Sheth, S. A., Mian, M. K., Patel, S. R., Asaad, W. F., Williams, Z. M., Dougherty, D. D., Bush, G., & Eskandar, E. N. (2012). Human dorsal anterior cingulate cortex neurons mediate ongoing behavioural adaptation. Nature, 488(7410), 218-221. https://doi.org/10.1038/nature11239Answers by value on buttons for 1, 2 and 3 from left to right, and blocks of 24 trials of one type.
- Gruber, S. A., Sagar, K. A., Dahlgren, M. K., Gonenc, A., Smith, R. T., Lambros, A. M., Cabrera, K. B., & Lukas, S. E. (2018). The grass might be greener: Medical marijuana patients exhibit altered brain activity and improved executive function after 3 months of treatment. Frontiers in Pharmacology, 8, 983. https://doi.org/10.3389/fphar.2017.00983The block timing: four control and four interference blocks of 42 seconds, 24 trials each, 1.25 s displays and 0.5 s between them.
- Dzianok, P., Antonova, I., Wojciechowski, J., Dreszer, J., & Kublik, E. (2022). The Nencki-Symfonia electroencephalography/event-related potential dataset: Multiple cognitive tasks and resting-state data collected in a sample of healthy adults. GigaScience, 11, giac015. https://doi.org/10.1093/gigascience/giac015The control and interference displays, answered on keys 1, 2 and 3.
Source papers
- Bush G, Shin LM, Holmes J, Rosen BR, Vogt BA (2003). The Multi-Source Interference Task: validation study with fMRI in individual subjects. Molecular Psychiatry, 8(1):60-70.Paradigm source, Adults
Questions
How long does the Multi-Source Interference Task take?
About 7 to 8 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?
Interference Effect, Control RT, Interference RT, Interference Effect (Medians), Accuracy Cost, Accuracy, Control Accuracy, Interference Accuracy, Mean Correct RT, Control Trials Used, Interference Trials Used, Anticipations, Missed Trials and Total Trials. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Multi-Source Interference Task used in research?
63 papers from 2021 to 2025 name it in their title or abstract, up 12% compared with 2016 to 2020. That ranks it 237 of 395 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Multi-Source Interference Task have?
In the published research, 1 paper reports its test-retest reliability and 32 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Which settings can a researcher change?
Pairs of blocks (2 to 6). 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