Episodic Memory
Levels of Processing Task
Also called: depth of processing task, Craik and Tulving task, incidental learning task
Measure the depth effect: questions about each word's letters, sound or meaning, then a surprise recognition test.
8-11 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
- 278
- Evidence grade
- C
- Registered clinical studies
- 1
What it measures
Measures how the way a word is processed affects memory. Participants answer questions about each word's letters, sound or meaning, then take a surprise recognition test.
Constructs
- Depth of processing
- Incidental learning
- Recognition memory
- Episodic memory
Research uses
- Encoding and memory research
- Aging and memory research
- Within-study comparison of encoding effects
How the paradigm works
The levels of processing task measures how the way a word is processed affects memory for it. Participants are told the task is about judging words. On each of 60 trials a question appears, and 2 seconds later an English word appears below it, answered Yes or No as quickly and accurately as possible. A third of the questions ask whether the word is in capital letters, a third whether it rhymes with a given word, and a third whether it names a kind of thing, such as an animal; half of each kind are answered yes. Which words get which question, and the order, are drawn for each session.
More about the paradigm
After the questions and 30 seconds of true-or-false arithmetic, a recognition test the participant was not told about shows the 60 words and 60 new words matched on length and frequency, one at a time, answered Old or New without a time limit. One question then asks whether the participant expected a memory test. The task follows Craik and Tulving (1975), Experiment 2, with category questions as the meaning questions, as in their Experiment 1. The words, rhyme words and new words were chosen for this task by stated rules from word norms and a pronouncing dictionary.
The main score is the depth effect: the share of words studied with category questions later called old minus the share of words studied with letter case questions. In Craik and Tulving's Experiment 2, recognition of words whose question was answered yes rose from 15% after case questions to 81% after sentence questions. Hit rates by question and answer, the false alarm rate, corrected recognition, d-prime and study answer times are also reported. Scores support research and do not diagnose.
Scores
- Depth Effect%
- Hit rate for words studied with category questions minus hit rate for words studied with letter case questions.
- Higher means processing meaning helped memory more
- Hit Rate, Case Questions%
- Share of words studied with letter case questions called old.
- Higher is better
- Hit Rate, Rhyme Questions%
- Share of words studied with rhyme questions called old.
- Higher is better
- Hit Rate, Category Questions%
- Share of words studied with category questions called old.
- Higher is better
- Hit Rate, Yes Answers%
- Share of studied words whose question was answered yes called old.
- Higher is better
- Hit Rate, No Answers%
- Share of studied words whose question was answered no called old.
- Higher is better
- Congruity Effect%
- Hit rate for yes answers minus hit rate for no answers.
- Positive values mean words that fit their question were remembered better
- Hit Rate%
- Share of all studied words called old.
- Higher is better
- False Alarm Rate%
- Share of new words called old.
- Lower is better
- Corrected Recognition%
- Hit rate minus false alarm rate.
- Higher is better
- d-Primed'
- z(hit rate) minus z(false alarm rate), with the log-linear correction.
- Higher means better recognition
- Study Accuracy%
- Share of study questions answered correctly.
- Context for the recognition scores
- Case Question Timems
- Median time of correct answers to letter case questions, from the word's appearance.
- Usually the shortest of the three
- Rhyme Question Timems
- Median time of correct answers to rhyme questions, from the word's appearance.
- Context for the depth effect
- Category Question Timems
- Median time of correct answers to category questions, from the word's appearance.
- Usually the longest of the three
- Expected a Memory Test0/1
- 1 if the participant said they expected a memory test before it began, 0 if not.
- 1 means the words may have been learned on purpose
Use and evidence
278 papers from 2021 to 2025 name this task in their title or abstract, down 20% compared with 2016 to 2020. That ranks it 91 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 65 |
| 2024 | 51 |
| 2023 | 35 |
| 2022 | 60 |
| 2021 | 67 |
| 2020 | 76 |
| 2019 | 86 |
| 2018 | 70 |
| 2017 | 55 |
| 2016 | 61 |
| 2015 | 70 |
| 2014 | 92 |
| 2013 | 71 |
| 2012 | 69 |
| 2011 | 77 |
| 2010 | 65 |
| 2009 | 44 |
| 2008 | 58 |
| 2007 | 35 |
| 2006 | 40 |
| 2005 | 32 |
| 2004 | 45 |
| 2003 | 36 |
| 2002 | 36 |
| 2001 | 37 |
| 2000 | 31 |
Use grade C
100 to 299 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
- 69
- Registered clinical studies using it as an outcome
- 1
Note on the counts: includes general uses of the term.
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.
- Words
- 60 concrete English nouns of 3 to 8 letters, each with a word that rhymes with it, a word that does not, a category it belongs to and one it does not. 60 new words are matched one to one on length and word frequency.
- Questions
- 20 letter case questions ("Is the word in capital letters?"), 20 rhyme questions and 20 category questions ("Is it an animal?"), 10 answered yes and 10 no of each, in an order with no more than 4 yes or no answers and no more than 3 questions of one type in a row. Words, questions and answers are assigned from the session's seed.
- Study trial
- The question alone for 2,000 ms, then the word below it, answered Yes or No without a time limit, then 500 ms blank. The instructions describe a word judgment task and never mention memory.
- Letter case
- Words of case questions answered yes are shown in capitals and those answered no in small letters; half the words of each other question and answer are shown in capitals.
- Practice
- One question of each type, each answer followed by feedback for 1,500 ms.
- Delay
- 30 seconds (standard; 0 to 120) of true-or-false arithmetic.
- Test
- A surprise test of 120 words, the 60 studied words and the 60 new words, one at a time in small letters, answered Old or New without a time limit. No more than 4 old or new words in a row.
- Last question
- Whether the participant expected a memory test. A yes is flagged for review and withholds the percentile.
The words, rhyme words and new words were chosen for this task by stated rules from word norms and a pronouncing dictionary; no published word list is used. The words are English; sessions run in other languages are flagged for review.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: levels_of_processing_test from the scored run, one old or new answer per word, with the study answers, delay arithmetic and the expectation answer from the same run. Old or new accuracy significantly below chance (one-sided exact binomial, p < .05, chance 50%) invalidates; accuracy not above chance over 10 or more answers is flagged. The same answer on 90% of the words invalidates when accuracy is not above chance. Answers under 200 ms on at least half of 4 or more answered words invalidate. Study answers below chance over 10 or more questions invalidate, and study answers not above chance are flagged. A participant who says they expected a memory test is flagged and the percentile withheld. No arithmetic answered in the delay, or delay arithmetic at or below chance over at least 10 answers, is flagged. A session run in another language is flagged. A hidden tab withholds the percentile.
Why these rules
Half the test words were studied and half are new, so guessing gives 50%: 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). 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 study questions have clear answers, so answering them below chance means the questions were not followed, and at chance means the words may not have been processed as the questions asked. The depth effect assumes incidental learning: the participant does not know a memory test will follow (Craik and Tulving 1975), so an expected test is flagged and its percentile withheld. Skipping the delay arithmetic means that part of the procedure may not have been followed. The word lists are English. A session run in another language shows English words to a participant who may not read English fluently, so it is flagged for review. 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 |
|---|---|---|
| Arithmetic before the word test (seconds) | 30 | 0 to 120 |
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
- Instructions and screens: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. Words and sentences: English.
- 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 Levels of Processing Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Craik & Lockhart, 1972; Craik & Tulving, 1975), with the standard settings (arithmetic before the word test (seconds): 30). The main score was depth effect (hit rate for words studied with category questions minus hit rate for words studied with letter case questions).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3), 268-294. https://doi.org/10.1037/0096-3445.104.3.268
- Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11(6), 671-684. https://doi.org/10.1016/S0022-5371(72)80001-X
- Hautus, M. J. (1995). Corrections for extreme proportions and their biasing effects on estimated values of d'. Behavior Research Methods, Instruments, & Computers, 27(1), 46-51. https://doi.org/10.3758/BF03203619
Source papers
- Craik FIM, Tulving E (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3):268-294.Paradigm source, Young adults
- Craik FIM, Lockhart RS (1972). Levels of processing: a framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11(6):671-684.Paradigm source
Questions
How long does the Levels of Processing Task take?
About 8 to 11 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?
Depth Effect, Hit Rate, Case Questions, Hit Rate, Rhyme Questions, Hit Rate, Category Questions, Hit Rate, Yes Answers, Hit Rate, No Answers, Congruity Effect, Hit Rate, False Alarm Rate, Corrected Recognition, d-Prime, Study Accuracy, Case Question Time, Rhyme Question Time, Category Question Time and Expected a Memory Test. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Levels of Processing Task used in research?
278 papers from 2021 to 2025 name it in their title or abstract, down 20% compared with 2016 to 2020. That ranks it 91 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Levels of Processing Task have?
In the published research, 2 papers report its test-retest reliability and 1 registered clinical study uses it as an outcome (evidence grade C in the ConductCognition test census).
Which settings can a researcher change?
Arithmetic before the word test (seconds) (0 to 120). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
The instructions and screens are in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. The words and sentences stay in English, so sessions in other languages are flagged for review. 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.