Attention
Attentional Blink Task
Also called: AB task, RSVP dual-target task, rapid serial visual presentation task
Measure the attentional blink with two digits in a rapid stream of letters, second-digit accuracy at lags 1 to 8 and lag-1 sparing.
8-10 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
- 332
- Evidence grade
- C
- Registered clinical studies
- 12
What it measures
Measures the attentional blink: after spotting one target in a rapid stream, people often miss a second target that follows within about half a second. Letters flash at 10 per second; participants report the two digits hidden in the stream.
Constructs
- Temporal attention
- Selective attention
- Visual awareness
Research uses
- Attention and awareness research
- Within-study group comparison
- Individual differences in temporal attention
How the paradigm works
The attentional blink task measures a brief gap in awareness: after spotting one target in a rapid stream, people often miss a second target that follows within about half a second, while a second target right after the first is often still seen. Capital letters flash one at a time in the middle of the screen, 10 per second, and two of the items are digits. After each stream, participants type the two digits they saw, in any order, or press SPACEBAR for a digit they missed.
More about the paradigm
The stream follows Chun and Potter (1995, Experiment 1), with digits as targets among letters in place of their letters among digits. Each stream has 16 items shown for 100 ms each with no gap, after a 400 ms plus sign and a 100 ms blank, and ends with a 100 ms ampersand mask. The first digit appears at position 3 to 7 and the second 1 to 8 items later (the lag). The digits are 2 to 9 and the letters are all capitals except I and O, so no digit looks like a letter. The test runs 10 trials at each lag by default after 8 practice trials with feedback, the first three at half speed. Each item lasts a whole number of screen frames at the measured refresh rate, and each trial records the measured onset of every item. Sizes are set in screen pixels without calibration. Every session runs a recorded random order.
The blink size is the rate of reporting the second digit when the first was reported (T2|T1), averaged over lags 6 to 8, minus the same rate averaged over lags 2 and 3, where Chun and Potter found it lowest. Larger values mean a deeper blink. T1 accuracy, T2|T1 at each lag and lag-1 sparing are also reported.
Scores
- Blink Size%
- Second-digit report rate given the first (T2|T1) at lags 6 to 8 minus lags 2 and 3.
- Larger means a deeper blink
- T1 Accuracy%
- Share of reports that include the first digit.
- Higher is better
- T2|T1 at Lags 2 and 3%
- Second-digit report rate given the first, averaged over lags 2 and 3.
- Lower means a deeper blink
- T2|T1 at Lags 6 to 8%
- Second-digit report rate given the first, averaged over lags 6 to 8.
- Higher is better
- Lag-1 Sparing%
- T2|T1 at lag 1 minus T2|T1 at lags 2 and 3.
- Positive means the second digit was spared right after the first
- T2|T1 at Lag 1%
- Second-digit report rate given the first at lag 1 (100 ms).
- Higher is better
- T2|T1 at Lag 2%
- Second-digit report rate given the first at lag 2 (200 ms).
- Higher is better
- T2|T1 at Lag 3%
- Second-digit report rate given the first at lag 3 (300 ms).
- Higher is better
- T2|T1 at Lag 4%
- Second-digit report rate given the first at lag 4 (400 ms).
- Higher is better
- T2|T1 at Lag 5%
- Second-digit report rate given the first at lag 5 (500 ms).
- Higher is better
- T2|T1 at Lag 6%
- Second-digit report rate given the first at lag 6 (600 ms).
- Higher is better
- T2|T1 at Lag 7%
- Second-digit report rate given the first at lag 7 (700 ms).
- Higher is better
- T2|T1 at Lag 8%
- Second-digit report rate given the first at lag 8 (800 ms).
- Higher is better
- Mean Item Durationms
- Measured mean duration of each item in the stream.
- Planned at 100 ms; off by more than 10 ms is flagged
- Dropped Frame Rate%
- Share of stream frames the browser dropped.
- Above 0.05 is flagged for review
Use and evidence
332 papers from 2021 to 2025 name this task in their title or abstract, down 10% compared with 2016 to 2020. That ranks it 76 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 83 |
| 2024 | 54 |
| 2023 | 52 |
| 2022 | 75 |
| 2021 | 68 |
| 2020 | 66 |
| 2019 | 68 |
| 2018 | 77 |
| 2017 | 81 |
| 2016 | 77 |
| 2015 | 100 |
| 2014 | 78 |
| 2013 | 87 |
| 2012 | 118 |
| 2011 | 113 |
| 2010 | 220 |
| 2009 | 107 |
| 2008 | 87 |
| 2007 | 95 |
| 2006 | 90 |
| 2005 | 63 |
| 2004 | 67 |
| 2003 | 37 |
| 2002 | 38 |
| 2001 | 43 |
| 2000 | 24 |
Use grade B
300 to 999 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
- 52
- Registered clinical studies using it as an outcome
- 12
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 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Display
- One capital letter at a time in the middle of the screen; two items in each stream are digits.
- Stream
- 16 items of 100 ms each with no gap, after a 400 ms plus sign and a 100 ms blank, ending with a 100 ms ampersand mask.
- Targets
- Two different digits from 2 to 9; the first at position 3 to 7, the second 1 to 8 items later.
- Letters
- All capitals except I and O, with no letter repeated within four items.
- Sizes
- Screen pixels without calibration, so the visual angle depends on the screen and the viewing distance.
- Report
- Type the two digits in any order after the stream, or SPACEBAR for a digit not seen; unspeeded, within 20 seconds.
- Length
- 10 trials at each of the 8 lags (6 to 30), with a break after every 40.
- Practice
- 8 trials with feedback, the first 3 at half speed.
- Frames
- Each item lasts a whole number of screen frames at the measured refresh rate; more than 5% dropped frames or items off by more than 10 ms is flagged for review.
This version follows the stream of Chun and Potter (1995, Experiment 1) with digits as targets among letters in place of letters among digits, and typed reports in place of spoken ones.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: attentional_blink_test from the scored run; each trial counts as correct when the report includes the first digit (chance 25%: two of eight digits). Below chance invalidates and not above chance is flagged. One report on 90% of trials invalidates when accuracy is not above chance. Most reports finished under 200 ms are flagged. More than 5% of stream frames dropped, items off their planned duration by more than 10 ms on average, or too few trials to estimate the blink is flagged for review.
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). The blink is measured over tenths of a second, so a stream that drops frames or runs off its 100 ms pace changes the lags, and the second-digit rates are computed only on trials where the first digit was found (Martens and Wyble 2010).
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Trials per lag | 10 | 6 to 30 |
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 Attentional Blink Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Raymond, Shapiro and Arnell, 1992; Chun and Potter, 1995), with the standard settings (trials per lag: 10). The main score was blink size (second-digit report rate given the first (T2|T1) at lags 6 to 8 minus lags 2 and 3).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Raymond, J. E., Shapiro, K. L., & Arnell, K. M. (1992). Temporary suppression of visual processing in an RSVP task: An attentional blink? Journal of Experimental Psychology: Human Perception and Performance, 18(3), 849-860. https://doi.org/10.1037/0096-1523.18.3.849
- Chun, M. M., & Potter, M. C. (1995). A two-stage model for multiple target detection in rapid serial visual presentation. Journal of Experimental Psychology: Human Perception and Performance, 21(1), 109-127. https://doi.org/10.1037/0096-1523.21.1.109Stream design, lags 1 to 8 and scoring by lag.
- MacLean, M. H., & Arnell, K. M. (2012). A conceptual and methodological framework for measuring and modulating the attentional blink. Attention, Perception, & Psychophysics, 74(6), 1080-1097. https://doi.org/10.3758/s13414-012-0338-4Blink size from a short lag other than lag 1 and a long lag.
Source papers
- Raymond JE, Shapiro KL, Arnell KM (1992). Temporary suppression of visual processing in an RSVP task: an attentional blink? Journal of Experimental Psychology: Human Perception and Performance, 18(3):849-860.Paradigm source, Adults
- Chun MM, Potter MC (1995). A two-stage model for multiple target detection in rapid serial visual presentation. Journal of Experimental Psychology: Human Perception and Performance, 21(1):109-127.Paradigm source, Adults
Questions
How long does the Attentional Blink Task take?
About 8 to 10 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?
Blink Size, T1 Accuracy, T2|T1 at Lags 2 and 3, T2|T1 at Lags 6 to 8, Lag-1 Sparing, T2|T1 at Lag 1, T2|T1 at Lag 2, T2|T1 at Lag 3, T2|T1 at Lag 4, T2|T1 at Lag 5, T2|T1 at Lag 6, T2|T1 at Lag 7, T2|T1 at Lag 8, Mean Item Duration and Dropped Frame Rate. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Attentional Blink Task used in research?
332 papers from 2021 to 2025 name it in their title or abstract, down 10% compared with 2016 to 2020. That ranks it 76 of 395 test families in the ConductCognition test census (use grade B). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Attentional Blink Task have?
In the published research, 4 papers report its test-retest reliability and 12 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Which settings can a researcher change?
Trials per lag (6 to 30). 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.