Episodic Memory
Verbal Overshadowing Task
Also called: verbal overshadowing effect, verbal overshadowing paradigm, face description and recognition task
Test whether describing a face makes it harder to recognize: participants see a face, describe it in writing or type the names of countries for a minute, then pick it out from four faces or answer that it is not there.
20-25 minUntimed
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
Tests whether describing a face in words makes it harder to recognize. In each round participants see a face, then either describe it in writing or type the names of countries and their capitals for the same time, then pick it out from four faces or answer that it is not there. The main score is how many fewer faces are picked out after describing.
Constructs
- Face recognition memory
- Verbal recoding of visual memories
- Response criterion in recognition
Research uses
- Face recognition and eyewitness memory research, at the group level
- Replication research on verbal overshadowing
- Studies of how language and visual memory interact
How the paradigm works
The verbal overshadowing task tests whether describing a face in words makes it harder to recognize. Schooler and Engstler-Schooler (1990) showed people a short film of a robbery; those who then wrote a description of the robber's face for 5 minutes picked him out of eight photographs less often than those who did an unrelated task, 38% against 64%. A registered replication in 22 laboratories (Alogna and colleagues, 2014) found a drop of 16 percentage points when the description came just before the choice, as in that study, and a replication in 31 laboratories one of 4 points when it came right after the film: the effect holds, smaller than first reported, and it is read across a group.
More about the paradigm
Here there is no crime story. Each round shows a photograph of a smiling face for 5 seconds; then the participant either describes that face in writing or types the names of countries and their capitals, for 2 minutes as standard (1 or 5 minutes in the settings); then chooses that person among four neutral photographs, or answers that the face is not there. In half the rounds it is not there. The 8 rounds come in two blocks of four, one for each activity, in an order drawn for each session, because describing one face can lower recognition of others (Dodson, Johnson and Schooler, 1997). Two practice rounds with drawn faces and feedback come first. The photographs come from the Face Research Lab London Set (DeBruine and Jones, 2017): 36 adults in 8 sets, each set of one gender and group and close in age.
The main score is the overshadowing effect: the percent of face-present rounds on which the face was picked out after typing countries minus after describing, in percentage points. The task also reports, for each activity, the percent of faces picked out, of face-absent rounds answered Not there and of all rounds answered Not there, since describing can make people more cautious to choose (Clare and Lewandowsky, 2004), with accuracy and the median choice time. The typed text is kept with the data for the researcher and is never scored. Scores support research and do not diagnose.
Scores
- Overshadowing Effect%
- Percent of face-present rounds with the face picked out after typing countries minus after describing, in percentage points. Positive means describing the face made it harder to pick out.
- Informational; read over a group, since a person has 2 face-present rounds of each activity
- Faces Picked After Describing%
- Percent of face-present rounds after describing on which the face was picked out.
- Higher means more faces recognized
- Faces Picked After Countries%
- Percent of face-present rounds after typing countries on which the face was picked out.
- Higher means more faces recognized
- Absent Faces Rejected After Describing%
- Percent of face-absent rounds after describing answered Not there.
- Higher means fewer wrong picks
- Absent Faces Rejected After Countries%
- Percent of face-absent rounds after typing countries answered Not there.
- Higher means fewer wrong picks
- Not There After Describing%
- Percent of all rounds after describing answered Not there.
- Informational; more than after countries means more caution after describing
- Not There After Countries%
- Percent of all rounds after typing countries answered Not there.
- Informational
- Accuracy After Describing%
- Percent of rounds after describing answered right: the face picked out when it was there, Not there when it was not.
- Quality check
- Accuracy After Countries%
- Percent of rounds after typing countries answered right.
- Quality check
- Median Choice Timems
- Median time from the faces appearing to the choice, over choices at or after 100 ms.
- Informational; the choice has no time limit
- Rounds Answeredtrials
- Rounds answered in the scored run, leaving out anticipations.
- Fewer than planned is flagged for review
- Planned Roundstrials
- Rounds the session planned: 8.
- Informational
- Anticipationscount
- Choices within 100 ms of the faces appearing, left out of the other scores.
- Quality check
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 368 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 6 |
| 2024 | 9 |
| 2023 | 3 |
| 2022 | 7 |
| 2021 | 3 |
| 2020 | 3 |
| 2019 | 3 |
| 2018 | 3 |
| 2017 | 5 |
| 2016 | 5 |
| 2015 | 12 |
| 2014 | 12 |
| 2013 | 10 |
| 2012 | 11 |
| 2011 | 15 |
| 2010 | 7 |
| 2009 | 6 |
| 2008 | 18 |
| 2007 | 7 |
| 2006 | 6 |
| 2005 | 3 |
| 2004 | 4 |
| 2003 | 4 |
| 2002 | 16 |
| 2001 | 3 |
| 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
- 0
- Papers on reliability or validity
- 10
- 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 3 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Faces
- Grayscale photographs of 36 adults from the Face Research Lab London Set in 8 sets, each set of one self-reported gender and group and close in age, 4 sets of women and 4 of men, aligned on the eyes and cropped to an oval that hides hair, ears and clothing.
- Each round
- A Start button, a 500 ms cross, then a smiling photograph of one person for 5 seconds. Then 2 minutes (standard; 1 or 5) of describing that face in writing, or of typing the names of countries and their capitals, which ends by itself. Then four neutral photographs up to 200 px wide, in a row on a wide screen and two to a row on a tall one, and Not there: the participant clicks or taps one, with no time limit. A 500 ms blank follows each choice.
- Rounds
- 8, in two blocks of 4, one for each activity, the first drawn for each session, with a rest between them. In each block the face is there in 2 rounds and not there in 2, one of each with women and one with men; which sets describe, the faces' places and the order within each block are drawn for each session.
- Practice
- 2 rounds with drawn faces, 20 seconds of describing with the face there, then 20 seconds of countries with it not there, each with feedback for 2,000 ms.
- Typed text
- Kept with the data as typed, with its length and the time to the first key. No score reads it.
The procedure follows Schooler and Engstler-Schooler (1990) and the registered replication of Alogna and colleagues (2014): a face, a written description or a control task of the same length, here countries and their capitals as in the replication, then a choice with a Not there answer. There is no crime story: the faces are photographs of people the participant does not know, the person is seen in a different photograph at the choice, and a choice shows four faces where the published lineups showed six or eight. The published studies gave one face to each person, with 5 minutes to describe it, and compared groups; here each person describes the faces of one block and lists countries after the faces of the other, 2 minutes each as standard. The effect is usually read across a group, and the replication found it smaller than the first study: correct picks fell by 16 percentage points with the description just before the lineup, against 25 in the original, and by 4 with the description right after the face was seen. With several faces, the effect was clearest on the first face (Fallshore and Schooler, 1995), and 15-second descriptions of many faces made recognition better (Brown and Lloyd-Jones, 2005), so compare the first block of each session between people as well as the two blocks within people. Describing can also make people answer Not there more often (Clare and Lewandowsky, 2004), which is why faces picked out and Not there answers are reported apart. Its people include 23 of the Face Memory Test's 24 distractor faces, 13 of the Face-name Memory Task's 24 foils, 29 of the Face Matching Task's 40, 13 of the Chimeric Faces Task's 18, 19 of the Part-Whole Face Task's 30 and 9 of the Thatcher Illusion Task's 14, so in a study with any of them, run this task in a separate session. The Two-Tone Pictures Task makes its pictures from the 3/4 photographs of 22 of these people, which this task never shows. No screen names or labels a face. 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: verbal_overshadowing_test from the scored run, one choice among 4 faces and Not there per round with no time limit (chance 1 in 5). Accuracy significantly below chance invalidates and not above chance is flagged once a run has enough rounds to judge. Choices under 200 ms on at least half the rounds invalidate. A run short of its planned rounds is flagged for review, and so is Not there in every round.
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. 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. The face was there in half the rounds, so Not there in every round means the faces were not compared, and the study face shows for a set time, so a hidden tab withholds the percentile.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Writing time per round | 2 minutes (standard) | 1 minute, 2 minutes (standard), 5 minutes |
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 and tablets. 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 Verbal Overshadowing Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Schooler and Engstler-Schooler, 1990), with the standard settings (writing time per round: 2 minutes). The main score was overshadowing effect (percent of face-present rounds with the face picked out after typing countries minus after describing, in percentage points).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Schooler, J. W., & Engstler-Schooler, T. Y. (1990). Verbal overshadowing of visual memories: Some things are better left unsaid. Cognitive Psychology, 22(1), 36-71. https://doi.org/10.1016/0010-0285(90)90003-M
- Alogna, V. K., Attaya, M. K., Aucoin, P., Bahník, Š., Birch, S., Birt, A. R., Bornstein, B. H., Bouwmeester, S., Brandimonte, M. A., Brown, C., ... Zwaan, R. A. (2014). Registered Replication Report: Schooler and Engstler-Schooler (1990). Perspectives on Psychological Science, 9(5), 556-578. https://doi.org/10.1177/1745691614545653The replication in 31 and 22 laboratories: correct picks lower by 4.01 and 16.31 percentage points after describing, with countries and capitals as the control.
- Meissner, C. A., & Brigham, J. C. (2001). A meta-analysis of the verbal overshadowing effect in face identification. Applied Cognitive Psychology, 15(6), 603-616. https://doi.org/10.1002/acp.728A small, reliable effect over 29 comparisons, more likely when the choice follows the description at once.
- Dodson, C. S., Johnson, M. K., & Schooler, J. W. (1997). The verbal overshadowing effect: Why descriptions impair face recognition. Memory & Cognition, 25(2), 129-139. https://doi.org/10.3758/BF03201107Describing one face also lowered recognition of another, the reason for the two blocks.
- Fallshore, M., & Schooler, J. W. (1995). Verbal vulnerability of perceptual expertise. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(6), 1608-1623. https://doi.org/10.1037/0278-7393.21.6.1608Four faces per person, 5 seconds each, chosen from a different photograph: the effect was clearest on the first face.
- Brown, C., & Lloyd-Jones, T. J. (2005). Verbal facilitation of face recognition. Memory & Cognition, 33(8), 1442-1456. https://doi.org/10.3758/BF03193377With 15-second descriptions of 12 faces, describing made recognition better.
- Clare, J., & Lewandowsky, S. (2004). Verbalizing facial memory: Criterion effects in verbal overshadowing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30(4), 739-755. https://doi.org/10.1037/0278-7393.30.4.739After describing, people chose a face less often from lineups with and without the face.
- Wilson, B. M., Seale-Carlisle, T. M., & Mickes, L. (2018). The effects of verbal descriptions on performance in lineups and showups. Journal of Experimental Psychology: General, 147(1), 113-124. https://doi.org/10.1037/xge0000354Lineups with and without the face: describing just before the lineup lowered correct picks from 62% to 38%, and responding was more cautious.
- DeBruine, L., & Jones, B. (2017). Face Research Lab London Set [Data set]. figshare. https://doi.org/10.6084/m9.figshare.5047666.v5
Source papers
- Schooler JW, Engstler-Schooler TY (1990). Verbal overshadowing of visual memories: Some things are better left unsaid. Cognitive Psychology, 22(1):36-71.N 88, Adults
- Alogna VK, Attaya MK, Aucoin P, et al. (2014). Registered Replication Report: Schooler and Engstler-Schooler (1990). Perspectives on Psychological Science, 9(5):556-578.Paradigm source, Adults, 18 to 25
- DeBruine L, Jones B (2017). Face Research Lab London Set. figshare. Dataset. doi:10.6084/m9.figshare.5047666.v5Paradigm source, Adults
Questions
How long does the Verbal Overshadowing Task take?
About 20 to 25 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops and tablets. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Overshadowing Effect, Faces Picked After Describing, Faces Picked After Countries, Absent Faces Rejected After Describing, Absent Faces Rejected After Countries, Not There After Describing, Not There After Countries, Accuracy After Describing, Accuracy After Countries, Median Choice Time, Rounds Answered, Planned Rounds and Anticipations. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Verbal Overshadowing 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 368 of 495 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Verbal Overshadowing Task have?
In the published research, 0 papers report 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?
Writing time per round (1 minute, 2 minutes (standard), 5 minutes). 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.