Working Memory
Letter Memory Task
Also called: letter memory updating task, letter updating task, running letter memory
Measure working memory updating: letters appear one at a time in lists of unpredictable length, and participants keep the last four in mind and enter them when the list stops.
7-9 minUntimed
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
- 18
- Evidence grade
- C
- Registered clinical studies
- 10
What it measures
Measures how well people update working memory. Letters appear one at a time in lists of unpredictable length. Participants keep the last four in mind, adding each new letter and dropping the oldest, and enter them when the list stops.
Constructs
- Working memory updating
- Serial order memory
- Executive function
Research uses
- Executive function research in aging and clinical groups
- Latent-variable studies of updating, shifting and inhibition
- Within-study comparison of short and long lists
How the paradigm works
The Letter Memory Task measures working memory updating: keeping the last few items of a stream in mind and replacing the oldest whenever a new one arrives. Letters appear one at a time at the center of the screen, each for 2,000 ms, in lists of 5, 7, 9 or 11 letters. A list can stop after any of these lengths, so the participant keeps the last four letters in mind throughout: after each letter they say the last four to themselves, adding the new letter and dropping the oldest. When the list stops, they enter the last four letters in order, by typing or on an on-screen keypad.
More about the paradigm
The task follows the letter memory task of Miyake and colleagues (2000), adapted from Morris and Jones (1990): letters shown for 2,000 ms each, the last four reported, lists of 5, 7, 9 or 11 letters, two practice lists of 5 and 7 letters, then 12 lists, 48 letters to report. A setting gives the form of Friedman and colleagues (2008): the last three, lists of 5, 7 or 9 letters at 2,500 ms each, three practice lists, then 12 lists. In both, each length comes once in every cycle of the lengths, as Friedman and colleagues ordered them, and a fixation cross of 1,000 ms comes before each list. The letters are 13 consonants that read the same way in all eight app languages; no letter comes twice in a list, and no letters side by side read as a syllable or an abbreviation.
Both sources had participants rehearse the letters aloud after each new one. Here the rehearsal is silent: the test records no voice and cannot check that a participant rehearsed. Every place needs a letter, so the entry screen asks for a best guess, where Friedman and colleagues let participants say "blank". Practice lists give feedback; test lists do not. Miyake and colleagues' 137 students reported 83% of the letters (SD 7%), with modest internal consistency over their 12 lists (Cronbach's alpha .42); Friedman and colleagues found alpha .61 in 568 twins aged about 17.
The Running Span also asks for the last items of lists that stop without warning, but it varies how many to report (3 to 7), runs lists on by at most 3 items, shows each digit for 300 ms, gives no strategy and scores digits in their place. Here the number to report stays the same, lists run on by up to seven letters, each letter stays on screen for its whole time, and the instructions ask for updating after every letter, as in the studies that use this task to measure updating. It is not the Letter Memory Test of Inman and colleagues (1998), a forced-choice test of effort.
The main score is the share of the letters to report that were entered, in any place, as Friedman and colleagues scored it (the proportion of letters recalled in Miyake and colleagues). The share in their own place and the share at each list length are also given. Wrong letters are split into letters shown earlier in the list, which should have been dropped, and letters not shown. Scores support research and do not diagnose.
Scores
- Letters Reported%
- Share of the letters to report that were entered, in any place, over all lists.
- Higher is better
- Letters in Place%
- Share of the letters to report entered in their own place.
- Higher is better
- Lists of 5%
- Share of the letters to report that were entered, on lists of 5 letters.
- Higher is better
- Lists of 7%
- Share of the letters to report that were entered, on lists of 7 letters.
- Higher is better
- Lists of 9%
- Share of the letters to report that were entered, on lists of 9 letters.
- Higher is better
- Lists of 11%
- Share of the letters to report that were entered, on lists of 11 letters. Null when the last 3 are reported, which uses lists of 5 to 9.
- Higher is better
- Outdated Letters%
- Share of the letters entered that were shown earlier in the list and should have been dropped.
- Lower is better
- Letters Not Shown%
- Share of the letters entered that were not shown in the list.
- Lower is better
- Lists All Correct%
- Share of lists with every letter to report entered.
- Higher is better
- Median Entry Timems
- Median time from the entry screen to Submit.
- Informational
- Lists Completedtrials
- Test lists recorded in the scored run.
- Under 90% of the planned lists is invalid
- Planned Liststrials
- Test lists the session planned to give.
- Informational
Use and evidence
18 papers from 2021 to 2025 name this task in their title or abstract, up 12% compared with 2016 to 2020. That ranks it 332 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 5 |
| 2024 | 6 |
| 2023 | 2 |
| 2022 | 5 |
| 2021 | 0 |
| 2020 | 5 |
| 2019 | 5 |
| 2018 | 4 |
| 2017 | 2 |
| 2016 | 0 |
| 2015 | 4 |
| 2014 | 2 |
| 2013 | 4 |
| 2012 | 0 |
| 2011 | 2 |
| 2010 | 1 |
| 2009 | 1 |
| 2008 | 0 |
| 2007 | 0 |
| 2006 | 0 |
| 2005 | 1 |
| 2004 | 0 |
| 2003 | 1 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 2 |
Use grade E
Fewer than 30 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
- 2
- Papers on reliability or validity
- 4
- Registered clinical studies using it as an outcome
- 10
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.
- Letters
- 13 consonants (B C D F G L M N P R T V Z), one at a time at the center, each for 2,000 ms with no gap (2,500 ms when the last 3 are reported).
- Lists
- 5, 7, 9 or 11 letters, three of each, 12 lists, each length once in every four lists, in a seeded order. Reporting the last 3: 5, 7 or 9 letters, four of each.
- List rules
- No letter twice in a list, and no letters side by side that are neighbors in the alphabet, start words or spell an abbreviation.
- Updating
- After each letter, participants say the last four to themselves, adding the new letter and dropping the oldest. Only the letters entered are recorded.
- Trial
- A fixation cross for 1,000 ms, the letters, then entry.
- Entry
- The last letters in order, typed, or clicked or tapped on a keypad of the 13 letters, with Delete and Submit. Untimed; every place needs a letter.
- Practice
- Lists of 5 and 7 letters (5, 7 and 9 when reporting the last 3), each followed by the letters to report, the letters entered and how many were right.
- Score
- The share of the letters to report that were entered, in any place, overall and at each list length; letters in place and outdated letters as well.
Miyake and colleagues and Friedman and colleagues had participants rehearse aloud; here the rehearsal is silent and only the letters entered are recorded. Every place needs a letter. This is not the Letter Memory Test of effort (Inman and colleagues 1998). No published norms are applied.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: test-phase entries of the scored run, one per list, each letter to report counted once. Chance is 4 in 13 for each letter at the standard form (3 in 13 when the last 3 are reported). Accuracy at or below chance is flagged for review and never invalidates. Fewer than 90% of the planned lists invalidates. A hidden tab withholds the percentile.
Why these rules
Keeping letters that should have been dropped is the error this test measures, and a participant who keeps the first four letters of every list enters a quarter of the letters to report, below the chance of 4 in 13; accuracy at or below chance is therefore flagged and never invalidates. Entry is untimed and needs every letter, so nothing can be missed and no answer can come too fast. A hidden tab lets the study, delay or response timers run without the participant, so a percentile is withheld after one.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Letters to report | The last 4 (standard) | The last 4 (standard), The last 3 |
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, tablets and phones. 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 Letter Memory Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Morris and Jones, 1990; Miyake et al., 2000), with the standard settings (letters to report: The last 4). The main score was letters reported (share of the letters to report that were entered, in any place, over all lists).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Morris, N., & Jones, D. M. (1990). Memory updating in working memory: The role of the central executive. British Journal of Psychology, 81(2), 111-121. https://doi.org/10.1111/j.2044-8295.1990.tb02349.xThe method: the last items of lists of unknown length, updated as each new item arrives.
- Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000). The unity and diversity of executive functions and their contributions to complex "frontal lobe" tasks: A latent variable analysis. Cognitive Psychology, 41(1), 49-100. https://doi.org/10.1006/cogp.1999.0734The standard form: the last 4 letters, lists of 5 to 11 letters at 2,000 ms each, 12 lists, rehearsal after every letter and the proportion of letters recalled.
- Friedman, N. P., Miyake, A., Young, S. E., DeFries, J. C., Corley, R. P., & Hewitt, J. K. (2008). Individual differences in executive functions are almost entirely genetic in origin. Journal of Experimental Psychology: General, 137(2), 201-225. https://doi.org/10.1037/0096-3445.137.2.201The last 3 form: lists of 5 to 9 letters at 2,500 ms each, each length once in every three lists, and letters scored in any order.
Source papers
- Miyake A, Friedman NP, Emerson MJ, Witzki AH, Howerter A, Wager TD (2000). The unity and diversity of executive functions and their contributions to complex "frontal lobe" tasks: a latent variable analysis. Cognitive Psychology, 41(1):49-100.Paradigm source, Undergraduates
- Friedman NP, Miyake A, Young SE, DeFries JC, Corley RP, Hewitt JK (2008). Individual differences in executive functions are almost entirely genetic in origin. Journal of Experimental Psychology: General, 137(2):201-225.Paradigm source, Twin study, about age 17
Questions
How long does the Letter Memory Task take?
About 7 to 9 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops, tablets and phones. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Letters Reported, Letters in Place, Lists of 5, Lists of 7, Lists of 9, Lists of 11, Outdated Letters, Letters Not Shown, Lists All Correct, Median Entry Time, Lists Completed and Planned Lists. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Letter Memory Task used in research?
18 papers from 2021 to 2025 name it in their title or abstract, up 12% compared with 2016 to 2020. That ranks it 332 of 395 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Letter Memory Task have?
In the published research, 2 papers report its test-retest reliability and 10 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Which settings can a researcher change?
Letters to report (The last 4 (standard), The last 3). 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