Language
Agreement Attraction Task
Also called: agreement attraction in comprehension, subject-verb agreement attraction
Measure how an intervening noun affects English agreement errors, with word reading times and grammar judgments.
5-8 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
- 64
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measure how an intervening noun affects subject-verb agreement in English. Participants read sentences word by word and judge their grammar. Compare reading times at and after the verb, and acceptance of agreement errors. A study can use quick whole-sentence judgments instead.
Constructs
- Sentence processing
- Agreement attraction
- Cue-based retrieval
Research uses
- Psycholinguistic research
- Language comprehension research
How the paradigm works
Agreement attraction measures how an intervening noun changes the processing of subject-verb agreement. Original English sentences cross a singular or plural intervening noun with a grammatical or ungrammatical verb, while the subject stays singular. Each sentence frame appears once, with its condition rotated across four seeded lists.
More about the paradigm
The standard run has 32 agreement sentences, 32 other sentences and four practice sentences. Participants press the space bar to uncover one word at a time, then use F for correct English or J for a grammar error. A study can instead show each whole sentence for a quick judgment. Each judgment has a four-second limit. Word-by-word mode reports times at the verb and the next two words; both settings report judgments.
Scores
- Attraction at the Verbms
- Mean verb time with a singular attractor minus a plural attractor in ungrammatical sentences. Each cell needs three usable sentences.
- Positive values mean agreement errors were read faster with a plural attractor
- Attraction after the Verbms
- Same contrast for the mean of the two words after the verb, with both words required.
- Positive values indicate facilitation after the verb
- Agreement Interactionms
- Ungrammatical attraction contrast minus the corresponding grammatical contrast at the verb.
- Separates the agreement-error effect from the grammatical control
- Attraction in Judgments%
- Acceptance rate of ungrammatical plural-attractor sentences minus singular-attractor sentences.
- Positive values mean more agreement errors were accepted with a plural attractor
- Judgment Accuracy%
- Correct judgments at 100 to 4000 ms divided by all presented test sentences.
- Read with omissions and the condition scores
- Acceptance by Condition%
- Share judged correct English, keyed by grammaticality and attractor number; at least three usable judgments per cell.
- Compare all four conditions
- Verb Reading Timesms
- Mean word times from 100 ms to the configured trim maximum, by condition.
- Null in whole-sentence judgment mode
- Following-Word Reading Timesms
- Mean of both following words per sentence, then mean by condition.
- Null in whole-sentence judgment mode
- Judgment Times by Conditionms
- Median correct judgment times from 100 to 4000 ms, with three per cell.
- Compare conditions within the same presentation setting
- Correct Judgment Timems
- Median correct judgment time across test sentences.
- Time from the judgment screen, not the start of word-by-word reading
- Early Judgmentscount
- Judgments under 100 ms.
- Excluded from accuracy, acceptance and response-time measures
- Early Word Advancescount
- Word advances under 100 ms.
- Excluded from reading-time means
- Missing Judgmentscount
- Sentences without an F or J response.
- Retained in the accuracy denominator
- Sentences Presentedcount
- Judgment rows in the selected run.
- Includes omissions
- Sentences Plannedcount
- Test sentences set before the run.
- Compare with presented sentences
Use and evidence
64 papers from 2021 to 2025 name this task in their title or abstract, up 2% compared with 2016 to 2020. That ranks it 322 of 582 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 16 |
| 2024 | 14 |
| 2023 | 11 |
| 2022 | 14 |
| 2021 | 9 |
| 2020 | 10 |
| 2019 | 13 |
| 2018 | 16 |
| 2017 | 11 |
| 2016 | 13 |
| 2015 | 7 |
| 2014 | 4 |
| 2013 | 1 |
| 2012 | 2 |
| 2011 | 2 |
| 2010 | 0 |
| 2009 | 1 |
| 2008 | 1 |
| 2007 | 0 |
| 2006 | 1 |
| 2005 | 0 |
| 2004 | 0 |
| 2003 | 0 |
| 2002 | 0 |
| 2001 | 1 |
| 2000 | 0 |
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
- 2
- 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 8 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Sentences
- 32 original agreement frames and 32 other sentences by default. Agreement frames cross grammaticality with singular or plural intervening nouns; the subject is singular. One version per frame appears in each of four seeded lists.
- Reading
- Press the space bar for each word in a moving window, then judge the sentence with F or J. Word times at the verb and both following words are saved. The sentence is not shown again during the judgment.
- Quick judgments
- The optional setting shows a 500 ms fixation and then the whole sentence. Answer with F for correct English or J for a grammar error, within 4000 ms. Reading-time contrasts are null in this setting.
- Practice
- Four sentences with feedback, then 32 to 64 test sentences without feedback. Agreement count ranges from 16 to 32 and rounds down to a multiple of four. Other sentences balance correct and incorrect answers.
- Scores
- Attraction at the verb and after it is singular-attractor minus plural-attractor mean time in ungrammatical sentences. The interaction subtracts the same contrast in grammatical sentences. Judgment attraction is plural-attractor minus singular-attractor error acceptance.
- Timing
- Reading times under 100 ms or above the study trim limit (2500 ms by default) leave the means. Both spillover words must be usable. Each condition needs three usable sentences. Judgment measures exclude responses below 100 ms.
- Interpretation
- Use these contrasts for group research and keep the two presentation settings separate. English reading is required. No percentile is computed.
Adapted from Wagers, Lau and Phillips with original items, four practice sentences, shorter runs, a 50% error rate, grammar judgments after reading and fixed trimming. These settings are ConductScience choices; the source used comprehension questions.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scores judgments from the run with the most sentences, ties going to the latest run, and word times from its completed sentences. Chance is 1 in 2. Each contrast needs three usable sentences per condition. Short runs and missing judgment contrasts are flagged. A session run in another language is flagged. Omissions and responses under 100 ms follow the timed-test rule.
Why these rules
Agreement attraction is a within-task condition contrast. Word times outside 100 ms to the study's trim limit leave the reading means; each spillover mean needs both following words. Correct judgments determine judgment times, while acceptance rates use all usable judgments. These are ConductScience quality rules. 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.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Presentation | Word-by-word reading | Word-by-word reading, Quick judgments |
| Agreement sentences | 32 | 16 to 32 |
| Longest reading time kept (ms) | 2,500 | 1,500 to 4,000 |
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
- 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.
- 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 Agreement Attraction Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Wagers, Lau and Phillips, 2009), with the standard settings (presentation: Word-by-word reading; agreement sentences: 32; longest reading time kept (ms): 2,500). The main score was attraction at the verb (mean verb time with a singular attractor minus a plural attractor in ungrammatical sentences).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Wagers, M. W., Lau, E. F., & Phillips, C. (2009). Agreement attraction in comprehension: Representations and processes. Journal of Memory and Language, 61(2), 206-237. https://doi.org/10.1016/j.jml.2009.04.002Factorial agreement-attraction design and word-by-word moving-window reading.
Source papers
Questions
How long does the Agreement Attraction Task take?
About 5 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?
Attraction at the Verb, Attraction after the Verb, Agreement Interaction, Attraction in Judgments, Judgment Accuracy, Acceptance by Condition, Verb Reading Times, Following-Word Reading Times, Judgment Times by Condition, Correct Judgment Time, Early Judgments, Early Word Advances, Missing Judgments, Sentences Presented and Sentences Planned. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Agreement Attraction Task used in research?
64 papers from 2021 to 2025 name it in their title or abstract, up 2% compared with 2016 to 2020. That ranks it 322 of 582 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.39.0, counted 8 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Agreement Attraction 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?
Presentation (Word-by-word reading, Quick judgments), agreement sentences (16 to 32) and longest reading time kept (ms) (1,500 to 4,000). 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.