Language
Word Superiority Task
Also called: word superiority effect, WSE, Reicher-Wheeler task, Reicher-Wheeler paradigm, word superiority paradigm, letter identification in words, two-alternative letter identification
Measure the word superiority effect: a flashed letter is picked out better inside a word than alone or in a made-up string, read across a group.
6-8 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. Try it here first, with no account.
- Papers, 2021 to 2025
- 28
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures the word superiority effect: a letter flashed briefly is identified better inside a word than alone or inside a made-up string. Participants pick which of two letters was shown, and both letters would make a word. The advantage is read across a group.
Constructs
- Visual word recognition
- Letter perception
- Orthographic processing
Research uses
- Reading and word recognition research, at the group level
- Letter perception research
- Within-study group comparison
How the paradigm works
The word superiority task measures the word superiority effect: a letter flashed briefly is identified better when it is part of a word than when it is part of a string that is not a word, or alone (Reicher, 1969; Wheeler, 1970). After the flash, two letters are offered for one place, and both make a word with the rest of the word, so knowing the word cannot help a guess; the same two letters are offered when the item is shown as a made-up string or as a single letter (Jordan and colleagues, 2024).
More about the paradigm
On a dark panel, two short marks show where to look. After 500 to 1,000 ms, four letters, or one letter in its place, flash between them for 50 ms in whole screen frames (the standard; 33, 67 or 83 ms is a setting fixed before the session, the range of exposures skilled readers needed in Lally and Rastle, 2023). Hash marks then cover the four places, and two letters appear above and below the place asked about; the up arrow key picks the letter above and the down arrow key the letter below, with no time limit. Each of 48 word pairs is shown once as a word, once as a made-up string of the same four letters with the other three moved, and once as a letter alone, in three parts of 48 trials. The words are written for this task in English, German, Spanish, French, Italian and Portuguese; Chinese and Japanese sessions show the English pairs. Six practice trials with feedback come first, the first three flashed for 150 ms. The primary score is the share of right choices for letters in words minus made-up strings. One person's difference is noisy, so it is read across a group.
Scores
- Word Advantage%
- Share of right choices for letters in words minus made-up strings.
- Positive values mean letters were seen better in words; read across a group
- Word Advantage over Letters Alone%
- Share of right choices for letters in words minus letters alone.
- Positive values mean letters were seen better in words than alone
- Words%
- Share of right choices when the letter was in a word.
- Higher is better; guessing gives 0.5
- Made-Up Strings%
- Share of right choices when the letter was in a made-up string.
- Higher is better; guessing gives 0.5
- Letters Alone%
- Share of right choices when the letter was shown alone.
- Higher is better; guessing gives 0.5
- Accuracy%
- Share of right choices over every counted answer.
- Higher is better; guessing gives 0.5
- Median Answer Timems
- Median time of right answers from the onset of the letters.
- Informational; only accuracy is asked for
- Anticipationscount
- Answers under 100 ms from the onset of the letters, left out of the shares.
- Lower is better
- Keys Before the Letterscount
- Trials with an arrow key pressed before the letters appeared.
- Lower is better
- Flash Durationms
- Median measured time the letters were on the screen.
- Informational; the standard plan is 50 ms
- Frame Intervalms
- Median time between screen frames, about 16.7 ms on a 60 Hz screen.
- Informational
- Long Flashes%
- Share of trials whose letters stayed a frame or more too long, a sign of missed screen frames.
- Over 10% is flagged
- Counted Answerstrials
- Test trials answered at or after 100 ms.
- 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, up 47% compared with 2016 to 2020. That ranks it 369 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 8 |
| 2024 | 7 |
| 2023 | 3 |
| 2022 | 8 |
| 2021 | 2 |
| 2020 | 3 |
| 2019 | 3 |
| 2018 | 4 |
| 2017 | 4 |
| 2016 | 5 |
| 2015 | 3 |
| 2014 | 4 |
| 2013 | 7 |
| 2012 | 12 |
| 2011 | 11 |
| 2010 | 14 |
| 2009 | 3 |
| 2008 | 7 |
| 2007 | 6 |
| 2006 | 3 |
| 2005 | 4 |
| 2004 | 6 |
| 2003 | 5 |
| 2002 | 4 |
| 2001 | 3 |
| 2000 | 4 |
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
- 6
- 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
- On a dark panel, four places 44 px apart on the middle row, with two short marks above and below the middle. Letters in capitals, 44 px bold.
- Wait
- 500 to 1,000 ms of the marks alone, drawn for each trial.
- Flash
- Four letters, or one letter in its place, for 50 ms in whole screen frames, three frames of a 60 Hz screen (standard; 33, 67 or 83 ms as a setting), then hash marks over all four places.
- Choice
- With the hash marks, two letters appear above and below the place asked about. Both make a word with the rest of the word, so knowing the word cannot help a guess. The up arrow key picks the letter above and the down arrow key the letter below, with no time limit.
- Forms
- Each letter is shown inside a word, inside a made-up string of the same four letters with the other three moved, or alone in its place.
- Items
- 48 pairs of 4-letter words that differ in one letter, 12 with the change in each place, written for this task in English, German, Spanish, French, Italian and Portuguese. Chinese and Japanese sessions show the English pairs.
- Parts
- 3 parts of 48 trials: each part shows every pair once, a third as words, a third as made-up strings and a third as letters alone, and over the three parts each pair appears once in each form. 144 scored trials.
- Balance
- In each form, half the pairs show the first word and half the second, and in each part the right letter is above on half the trials of each form. No more than three trials of one form in a row.
- Practice
- 6 trials with their own pairs, each form twice, the first three flashed for 150 ms, with feedback for 800 ms. No feedback in the test.
- Score
- The share of right choices for letters in words minus made-up strings, which hold the same letters in the same places under the same mask. Also words minus letters alone, and the share right in each form.
The advantage is read across a group: it is a difference between two shares of right choices over 48 trials each, which is noisy in one person. The flash is a whole number of screen frames, three on a 60 Hz screen, and each trial saves how long the letters were shown; sessions whose letters stayed a frame or more too long on over a tenth of trials are flagged. A flash that leaves both words and made-up strings near the top leaves no room for an advantage, so such sessions are flagged too, and a shorter flash setting suits them. The letters appear on whatever screen the participant has, so their size in degrees varies between people. The scores support research and do not diagnose reading difficulties.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: word_superiority_test from the scored run, the up or down arrow for the letter above or below (chance 50%, the right letter above on half the trials), correctness from the key for the right letter's place. Below chance invalidates and not above chance is flagged. One key on 90% of answers invalidates when accuracy is not above chance. Responses under 100 ms follow the timed-test rule. Words and made-up strings both chosen right on 95% of trials or more, and letters shown a frame or more too long on over 10% of trials, are flagged.
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 advantage is a difference between two shares of right choices, so it cannot show when both are near the top, and letters kept on the screen too long are no longer brief. The limits of 95% and 10% of trials are ConductScience choices, the second as in the Divided Visual Field Task.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Flash duration (ms) | 50 (standard) | 33, 50 (standard), 67, 83 |
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 and Portuguese (Brazil). Chinese (Simplified) and Japanese sessions use English words.
- 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 Word Superiority Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Reicher, 1969; Wheeler, 1970), with the standard settings (flash duration (ms): 50). The main score was word advantage (share of right choices for letters in words minus made-up strings).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Reicher, G. M. (1969). Perceptual recognition as a function of meaningfulness of stimulus material. Journal of Experimental Psychology, 81(2), 275-280. https://doi.org/10.1037/h0027768The two-letter choice after a brief, masked display: letters in words were picked out better than in nonwords or alone.
- Wheeler, D. D. (1970). Processes in word recognition. Cognitive Psychology, 1(1), 59-85. https://doi.org/10.1016/0010-0285(70)90005-8The same comparison, which with Reicher's gave the paradigm its name.
- Jordan, T. R., Akkaya, A., Göçmüş, F., Kalan, A., Morgul, E., Önalan, K., & Sheen, M. (2024). The Reicher-Wheeler paradigm in word recognition research: A cautionary note on its actual contributions and published misconceptions. Frontiers in Psychology, 15, 1399237. https://doi.org/10.3389/fpsyg.2024.1399237Why both letters offered must fit the rest of the string, and the same two letters must serve every form of the item.
- Lally, C., & Rastle, K. (2023). Orthographic and feature-level contributions to letter identification. Quarterly Journal of Experimental Psychology, 76(5), 1111-1119. https://doi.org/10.1177/17470218221106155A recent Reicher-Wheeler task with words, pseudowords and consonant strings, in which skilled readers' flashes ran from 33 to 83 ms, the range of this test's setting.
- Massol, S., & Grainger, J. (2025). Word superiority and sentence superiority effects in post-cued letter-in-string identification. Attention, Perception, & Psychophysics, 87(4), 1342-1352. https://doi.org/10.3758/s13414-025-03059-wLetter identification run online on participants' own computers showed the word superiority effect.
Source papers
- Reicher (1969), Journal of Experimental Psychology, 81(2), 275-280, and Wheeler (1970), Cognitive Psychology, 1(1), 59-85: the two-letter choice for words, nonwords and single letters.Paradigm source, Adults
Questions
How long does the Word Superiority Task take?
About 6 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?
Word Advantage, Word Advantage over Letters Alone, Words, Made-Up Strings, Letters Alone, Accuracy, Median Answer Time, Anticipations, Keys Before the Letters, Flash Duration, Frame Interval, Long Flashes, Counted Answers and Total Trials. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Word Superiority Task used in research?
28 papers from 2021 to 2025 name it in their title or abstract, up 47% compared with 2016 to 2020. That ranks it 369 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 Word Superiority 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?
Flash duration (ms) (33, 50 (standard), 67, 83). 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 and Portuguese (Brazil), with its words in the same language. In Chinese (Simplified) and Japanese, the instructions are translated and the words stay in English. 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.