Episodic Memory
Comprehension Judgments Task
Also called: metacomprehension, comprehension monitoring, text-learning judgments
Measure how well people predict their understanding of text. Compare predictions with later question accuracy to score relative accuracy and bias.
7-10 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
- 43
- Evidence grade
- D
- Registered clinical studies
- 0
Two gears
What percentage will you answer correctly?
What it measures
Measures how well people predict their understanding of text. Read six short texts, predict test performance for each, then answer questions.
Constructs
- Metacomprehension
- Relative monitoring accuracy
- Prediction bias
- Text comprehension
Research uses
- Metacognition research
- Learning research
- Reading comprehension research
How the paradigm works
Metacomprehension is judging how well you understand what you read. Participants read six original expository texts and predict the percentage of questions they will answer correctly for each, using 0, 20, 40, 60, 80 or 100%. Predictions follow each text by default, or all six texts in a delayed setting. Only the title is shown when making a prediction. All scored questions follow all reading and predictions. Each text has one retention and two application questions with three choices each. Text and answer orders are seeded. Reading and responses have no time limit. A separate practice text includes a prediction, a question and feedback.
Scores
- Relative Accuracyscore
- Goodman-Kruskal gamma across texts with a prediction and all three answers. Ties are excluded; null if no pair differs on both measures.
- Higher means better ordering of texts by performance
- Prediction Biaspoints
- Mean prediction minus actual percentage correct across the same paired texts, in percentage points.
- Positive means overprediction
- Mean Prediction%
- Mean predicted percentage correct across paired texts.
- Informational
- Observed Accuracy of Paired Texts%
- Mean actual percentage correct for the texts used in the prediction comparison.
- Informational
- Questions Correct%
- Accuracy across all recorded questions.
- Higher means more correct
- Retention Correct%
- Accuracy on retention questions.
- Informational
- Transfer Correct%
- Accuracy on application questions.
- Informational
- Texts Paired with Predictionscount
- Complete text tests matched to a prediction in the same run.
- 6 standard
- Complete Text Testscount
- Texts with all three question rows.
- 6 standard
- Predictions Recordedcount
- Unique valid predictions from the scored run.
- 6 standard
- Untied Text Pairscount
- Pairs that differ on both prediction and observed accuracy.
- Up to 15
- Predictions Under 100 mscount
- Predictions selected within 100 ms.
- 0 expected
- Predictions Under 200 mscount
- Predictions selected within 200 ms, for validity.
- 0 expected
- Questions Recordedcount
- Unique question rows in the most complete run.
- 18 standard
- Answers Under 100 mscount
- Question answers within 100 ms.
- 0 expected
- Median Answer Timems
- Median question response time, excluding times under 100 ms.
- Informational
Use and evidence
43 papers from 2021 to 2025 name this task in their title or abstract, up 7% compared with 2016 to 2020. That ranks it 382 of 582 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 7 |
| 2024 | 11 |
| 2023 | 9 |
| 2022 | 10 |
| 2021 | 6 |
| 2020 | 8 |
| 2019 | 8 |
| 2018 | 7 |
| 2017 | 7 |
| 2016 | 10 |
| 2015 | 5 |
| 2014 | 4 |
| 2013 | 8 |
| 2012 | 15 |
| 2011 | 12 |
| 2010 | 9 |
| 2009 | 7 |
| 2008 | 10 |
| 2007 | 11 |
| 2006 | 6 |
| 2005 | 6 |
| 2004 | 7 |
| 2003 | 1 |
| 2002 | 2 |
| 2001 | 2 |
| 2000 | 2 |
Use grade D
30 to 99 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
Measured on ConductCognition
Pooled across the studies that feed our platform norms, by device group. A figure needs 50 or more people.
No pooled figure yet.
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 0.39.0, counted 10 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Texts
- 6 original expository texts in all 8 languages; distinct from the Interesting Details lessons.
- Predictions
- 0, 20, 40, 60, 80 or 100% correct, with the text title as cue. After each text by default, or after all six texts.
- Questions
- 18 questions after all reading and predictions: 1 retention and 2 application questions per text, each with 3 choices.
- Practice
- 1 separate text, a prediction and a question with feedback.
- Order and timing
- Seeded reading and test order, with shuffled answer positions. Self-paced reading and untimed responses.
- Scores
- Text-level Goodman–Kruskal gamma and prediction minus observed accuracy in percentage points. Complete text tests only; tied pairs do not enter gamma.
These are our own brief texts and questions, with no restudy or keyword generation. Gamma is undefined when no text pair differs on both predictions and performance. Compare sessions using the same language and prediction timing. Scores support research and do not diagnose. No percentile output.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: unique text questions from the most complete run. Three choices give one-third chance. Fewer than half the questions invalidate. With at least four answers, half or more under 200 ms invalidate. Missing answers are flagged, and too many invalidate. Incomplete runs, repeated positions and a hidden tab are flagged. Missing text-prediction pairs and undefined gamma are flagged. With at least four predictions, half or more under 200 ms invalidate.
Why these rules
Every question has one answer supported by its text. Seeded answer positions separate content from button preference. Reading and answers are self-paced; response time is descriptive. The completion threshold is a ConductScience rule.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| When predictions are made | After each text (standard) | After each text (standard), After all texts |
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.
- 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 Comprehension Judgments Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Thiede KW, Anderson MCM, Therriault D, 2003; Nelson TO, 1984), with the standard settings (when predictions are made: After each text). The main score was relative accuracy (goodman-kruskal gamma across texts with a prediction and all three answers).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Thiede, K. W., Anderson, M. C. M., & Therriault, D. (2003). Accuracy of metacognitive monitoring affects learning of texts. Journal of Educational Psychology, 95(1), 66–73. https://doi.org/10.1037/0022-0663.95.1.66Text-level judgments matched with subsequent comprehension performance. Original texts; no training or restudy.
- Nelson, T. O. (1984). A comparison of current measures of the accuracy of feeling-of-knowing predictions. Psychological Bulletin, 95(1), 109–133. https://doi.org/10.1037/0033-2909.95.1.109Goodman–Kruskal gamma as a measure of relative monitoring accuracy, excluding ties.
Source papers
- Thiede, Anderson and Therriault (2003); Nelson (1984): monitoring of text learning and gamma.Paradigm source, Adults who read the session language
Questions
How long does the Comprehension Judgments Task take?
About 7 to 10 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?
Relative Accuracy, Prediction Bias, Mean Prediction, Observed Accuracy of Paired Texts, Questions Correct, Retention Correct, Transfer Correct, Texts Paired with Predictions, Complete Text Tests, Predictions Recorded, Untied Text Pairs, Predictions Under 100 ms, Predictions Under 200 ms, Questions Recorded, Answers Under 100 ms and Median Answer Time. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Comprehension Judgments Task used in research?
43 papers from 2021 to 2025 name it in their title or abstract, up 7% compared with 2016 to 2020. That ranks it 382 of 582 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.39.0, counted 10 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Comprehension Judgments 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?
When predictions are made (After each text (standard), After all texts). 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, with its words in the same language. 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.