Processing Speed
Redundant Signals Task
Also called: redundant signals paradigm, redundant signals effect, redundant target effect, redundant targets task, bilateral redundant target paradigm, redundancy gain, race model inequality test
Measure how much faster people press for two flashes at once than for one, the redundancy gain, with a race model test of whether the two signals were combined.
7-9 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. Try it here first, with no account.
- Papers, 2021 to 2025
- 28
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures how much faster people press when two signals come at once than when one does, the redundancy gain. A small square flashes to the left of a cross, to the right, or on both sides at the same time, and participants press the space bar as soon as they see any square. A race model test asks whether the gain is larger than two separate detections racing each other could give.
Constructs
- Redundancy gain
- Integration of signals
- Interhemispheric integration
- Simple reaction time
Research uses
- Perception and attention research
- Interhemispheric integration research, at the group level
- Aging research, at the group level
- Within-study group comparison
How the paradigm works
The redundant signals task measures how much faster people respond when two signals come at once than when one does, the redundancy gain (Miller, 1982), here in its bilateral visual form: one square to the left of a cross, one to the right, or both at once (Miniussi, Girelli and Marzi, 1998). In healthy young adults two squares usually bring the press 12 to 30 ms sooner than one; Westerhausen's (2023) meta-analysis put control samples at 13.9 ms against the faster single side and 19.1 ms against the mean of the two. Part of any gain comes for free: if two separate detections race, the faster one wins, and Miller's race model inequality sets the most that such a race can give. Faster responses than that mean the two signals were combined on the way to the response.
More about the paradigm
A cross marks the middle of a dark panel. After 500 to 1,500 ms a small white square flashes for 50 ms, 300 px to the left of the cross, to the right, or on both sides at once, and the participant presses the space bar as fast as possible, always with the hand chosen at the start, which is recorded. The standard session has 80 trials of each kind and one catch trial with no square for every nine, 267 trials in 4 parts in a seeded order, after 11 practice trials with feedback. Each trial waits up to 1,000 ms for a press. The displays switch on screen frames, and the measured wait and flash duration are saved. Both squares sit on the cross's row, so a screen draws them at the same moment.
Scores
- Redundancy Gainms
- Mean response time to a square on the faster single side minus to two squares.
- Positive values mean faster responses to two squares
- Redundancy Gain, Mean of Sidesms
- Mean of the two single sides' mean response times minus the mean to two squares.
- Positive values mean faster responses to two squares
- Redundancy Gain, Hand's Sidems
- Mean response time to a single square on the side of the hand that pressed minus to two squares.
- Positive values mean faster responses to two squares
- Redundancy Gain, Mediansms
- Median response time on the faster single side minus the median to two squares.
- Positive values mean faster responses to two squares
- Response Time, Left Squarems
- Mean response time to a single square on the left.
- Lower is faster
- Response Time, Right Squarems
- Mean response time to a single square on the right.
- Lower is faster
- Response Time, Two Squaresms
- Mean response time to two squares at once.
- Lower is faster
- Race Model Gapms
- Mean of the four race gaps below: the time by which two-square responses beat the bound for two racing detections, less the catch trials' presses.
- Positive values mean responses faster than a race allows; read across a group
- Race Model Gap, 10th Percentilems
- The race gap at the 10th percentile of the distributions.
- Positive values mean responses faster than a race allows
- Race Model Gap, 15th Percentilems
- The race gap at the 15th percentile.
- Positive values mean responses faster than a race allows
- Race Model Gap, 20th Percentilems
- The race gap at the 20th percentile.
- Positive values mean responses faster than a race allows
- Race Model Gap, 25th Percentilems
- The race gap at the 25th percentile.
- Positive values mean responses faster than a race allows
- Squares Answered%
- Share of trials with a square answered at or after 100 ms, with no press before the square.
- Higher is better
- Presses Without a Square%
- Share of catch trials with a press.
- Lower is better; over a quarter is flagged
- Missed Squarescount
- Trials with a square and no press within 1,000 ms.
- Lower is better
- Presses Before the Squarecount
- Trials with a press before the square appeared, or before the moment it would have appeared on a catch trial.
- Lower is better
- Anticipationscount
- Presses under 100 ms from the square's onset.
- Lower is better
- Fewest Responses of a Kindtrials
- Usable responses in the kind of trial (left, right or both) with the fewest.
- Fewer than 20 is flagged
- Flash Durationms
- Median measured time the squares were on the screen.
- Informational; the plan is 50 ms
- Trials With a Squaretrials
- Trials with one or two squares in the scored run.
- Informational
- Catch Trialstrials
- Trials with no square in the scored run.
- Informational
- Total Trialstrials
- Trials in the scored run.
- Informational
Use and evidence
28 papers from 2021 to 2025 name this task in their title or abstract, down 22% compared with 2016 to 2020. That ranks it 366 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 5 |
| 2024 | 6 |
| 2023 | 3 |
| 2022 | 5 |
| 2021 | 9 |
| 2020 | 6 |
| 2019 | 11 |
| 2018 | 6 |
| 2017 | 7 |
| 2016 | 6 |
| 2015 | 13 |
| 2014 | 6 |
| 2013 | 10 |
| 2012 | 3 |
| 2011 | 6 |
| 2010 | 18 |
| 2009 | 4 |
| 2008 | 9 |
| 2007 | 6 |
| 2006 | 3 |
| 2005 | 8 |
| 2004 | 3 |
| 2003 | 4 |
| 2002 | 2 |
| 2001 | 2 |
| 2000 | 1 |
Use grade E
Fewer than 30 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 1
- Papers on reliability or validity
- 9
- 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 0.22.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Display
- A cross in the middle of a dark panel, and a 40 px white square 300 px to its left, to its right or on both sides at once, on the cross's row.
- Wait
- 500 to 1,500 ms of the cross alone, drawn for each trial.
- Flash
- The square or squares for 50 ms in whole screen frames, then the cross alone until a press or 1,000 ms.
- Key
- The space bar, always with the hand chosen at the start, which is recorded.
- Trial mix
- 80 trials of each kind (standard; 40 to 160), plus one catch trial with no square for every nine: 267 trials in 4 parts, in a seeded order.
- Practice
- 11 trials, 2 of them with no square, with feedback for 800 ms. No feedback in the test.
- Score
- Mean response time to the faster single side minus to two squares. The race model test gives, at the 10th, 15th, 20th and 25th percentiles, the time by which two squares beat the bound for two racing detections, less the catch trials' presses.
Sight only: a sight-and-sound version would rest on audio delays a browser cannot measure. Delays of the screen and the keyboard are the same for one square and two, so they cancel out of the gain and the race gaps. One person's race gaps are noisy, so compare groups.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: redundant_signals_test from the scored run, press the space bar for a square on the left, on the right or on both sides, and nothing on the trials with no square (one for every nine with a square), within 1,000 ms. Responding mainly on trials with no square invalidates; not separating squares from empty trials is flagged. Missed responses and responses under 100 ms follow the timed-test rule. Presses before the square are held to the same limit as responses under 100 ms. Presses on over half the trials with no square invalidate and on over a quarter are flagged. Fewer than 20 usable responses for any kind of trial is flagged.
Why these rules
Responses concentrated on trials that needed none (one-sided Fisher exact test, p < .05) mean the instructions were reversed and invalidate; responses that do not separate the two trial types are flagged. Responding is the common, prepotent response here, so missing more than max(3, 25%) of response trials signals non-engagement (the platform's timed-test rule); any miss is flagged. 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 gain is a difference between means and the race model test compares the fast part of three distributions, which is biased with fewer than 20 trials of each kind (Kiesel, Miller and Ulrich, 2007). The catch-trial limits follow the inhibition of return task's and are ConductScience choices.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Trials of each kind | 80 | 40 to 160 |
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 Redundant Signals Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Miller, 1982; Miniussi, Girelli and Marzi, 1998), with the standard settings (trials of each kind: 80). The main score was redundancy gain (mean response time to a square on the faster single side minus to two squares).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Miller, J. (1982). Divided attention: Evidence for coactivation with redundant signals. Cognitive Psychology, 14(2), 247-279. https://doi.org/10.1016/0010-0285(82)90010-XThe race model inequality: the most that two racing detections can give.
- Miniussi, C., Girelli, M., & Marzi, C. A. (1998). Neural site of the redundant target effect: Electrophysiological evidence. Journal of Cognitive Neuroscience, 10(2), 216-230. https://doi.org/10.1162/089892998562663The bilateral design: one checkerboard to the left or right of fixation, or two at once.
- Miller, J., & Lopes, A. (1991). Bias produced by fast guessing in distribution-based tests of race models. Perception & Psychophysics, 50(6), 584-590. https://doi.org/10.3758/BF03207544Presses made without seeing anything bias the test toward a race, and catch trials correct it.
- Kiesel, A., Miller, J., & Ulrich, R. (2007). Systematic biases and Type I error accumulation in tests of the race model inequality. Behavior Research Methods, 39(3), 539-551. https://doi.org/10.3758/BF03193024At least 20 trials of each kind, and the 10th to 25th percentiles.
- Rach, S., Diederich, A., Steenken, R., & Colonius, H. (2010). The race model inequality for censored reaction time distributions. Attention, Perception, & Psychophysics, 72(3), 839-847. https://doi.org/10.3758/APP.72.3.839Why no response is cut from either end of the distributions.
- Gondan, M., & Minakata, K. (2016). A tutorial on testing the race model inequality. Attention, Perception, & Psychophysics, 78(3), 723-735. https://doi.org/10.3758/s13414-015-1018-y
- Westerhausen, R. (2023). Interhemispheric integration after callosotomy: A meta-analysis of Poffenberger and redundant-target paradigms. Neuropsychology Review, 33(4), 872-890. https://doi.org/10.1007/s11065-022-09569-wThe size of the gain in control samples: 13.9 ms against the faster single side and 19.1 ms against the mean of the two.
Source papers
- Westerhausen R (2023). Interhemispheric integration after callosotomy: a meta-analysis of Poffenberger and redundant-target paradigms. Neuropsychology Review, 33(4):872-890.Paradigm source, Adults
Questions
How long does the Redundant Signals Task take?
About 7 to 9 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?
Redundancy Gain, Redundancy Gain, Mean of Sides, Redundancy Gain, Hand's Side, Redundancy Gain, Medians, Response Time, Left Square, Response Time, Right Square, Response Time, Two Squares, Race Model Gap, Race Model Gap, 10th Percentile, Race Model Gap, 15th Percentile, Race Model Gap, 20th Percentile, Race Model Gap, 25th Percentile, Squares Answered, Presses Without a Square, Missed Squares, Presses Before the Square, Anticipations, Fewest Responses of a Kind, Flash Duration, Trials With a Square, Catch Trials and Total Trials. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Redundant Signals Task used in research?
28 papers from 2021 to 2025 name it in their title or abstract, down 22% compared with 2016 to 2020. That ranks it 366 of 495 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.22.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Redundant Signals Task have?
In the published research, 1 paper reports 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?
Trials of each kind (40 to 160). 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.