Language
Picture Matching Task
Also called: sentence-picture matching, picture-sentence matching, reversible sentence comprehension, active and passive sentence comprehension
Measure how well people work out who does what to whom: participants choose which of two pictures an active or passive sentence describes, in their own language.
5-8 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
- 222
- Evidence grade
- B
- Registered clinical studies
- 5
What it measures
Measures comprehension of who does what to whom in a sentence. Participants see two pictures of the same two animals and the same action with the roles reversed, then read a sentence such as "The pig is pushed by the cat" and choose the picture it describes. Active and passive sentences, written in the participant's language, are compared for accuracy and response time.
Constructs
- Sentence comprehension
- Assigning who does what from word order and the passive
- Matching language to pictures
Research uses
- Language comprehension research
- Aphasia and child language research
- Within-study group comparison
How the paradigm works
Sentence-picture matching measures whether a reader or listener works out who does what to whom. In the reversible form used in aphasia and psycholinguistics research, either animal or person in the sentence could do the action, so word meanings alone cannot settle it: a sentence such as "The pig is pushed by the cat" comes with two pictures of the same action, one as the sentence says and one with the roles reversed. Active sentences name the doer first and passive sentences name the one acted on first, so the two kinds show how roles are read from the sentence's structure.
More about the paradigm
This task follows the two-picture design of Meyer, Mack and Thompson (2012) and Mack and colleagues (2016), with written sentences in the participant's language. Six animals (cat, rabbit, pig, goat, sheep and cow) do four actions: chase, push, pull and carry. The doer of a chase or a push is behind, of a pull ahead with a rope, and of a carry underneath, so no place in the picture gives the answer. 24 events, each animal doing each action once and having it done once, are each shown once as an active and once as a passive sentence (48 trials; a 12-event short form gives 24). The two pictures come first; a second later the sentence appears above them, and the participant clicks or taps the picture it describes within 20 seconds. The matching picture is on each side on half the trials, and half the scenes face each way. Four practice trials with feedback come first. Scores support research and do not diagnose.
The main score is the percentage of trials answered with the matching picture, and accuracy and the median time of correct choices are also reported for active and passive sentences, with the passive cost in each. In the spoken version of Mack and colleagues (2016), 20 young adults without language problems chose the matching picture on 99% or more of active and passive sentences, and in 12 people with aphasia accuracy and response times were reliable over about a week (intraclass correlations .64 to .92). With healthy adults near the maximum, response times carry most of the differences; here they include reading, and passive sentences are longer.
Scores
- Percent Correct%
- Test trials answered with the picture the sentence describes; a trial with no choice counts as wrong.
- Higher means more sentences understood
- Active Sentences Correct%
- Percent correct on active sentences, which name the doer first.
- Higher is better
- Passive Sentences Correct%
- Percent correct on passive sentences, which name the one acted on first.
- Higher is better
- Passive Accuracy Cost%
- Percent correct on active sentences minus on passive sentences, in percentage points.
- Positive means passives were harder
- Median Response Timems
- Median time from the sentence's onset to a correct choice, leaving out choices under 100 ms.
- Lower means faster; it includes reading the sentence. Compare within a device type
- Active Response Timems
- Median time of correct choices after active sentences.
- Lower means faster
- Passive Response Timems
- Median time of correct choices after passive sentences.
- Lower means faster
- Passive Time Costms
- Median time after passive sentences minus after active ones, on correct choices.
- Positive means passives took longer; they are also longer to read
- Choices Under 100 mscount
- Choices within 100 ms of the sentence appearing.
- Expected to be 0
- Missed Trialscount
- Trials with no choice within 20 seconds of the sentence.
- Expected to be 0
- Test Trials Recordedtrials
- Test trials in the scored run.
- Fewer than 90% of those set is invalid
- Test Trials Settrials
- Test trials the session set out to show.
- 24 or 48
Use and evidence
222 papers from 2021 to 2025 name this task in their title or abstract, down 1% compared with 2016 to 2020. That ranks it 115 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 56 |
| 2024 | 55 |
| 2023 | 38 |
| 2022 | 35 |
| 2021 | 38 |
| 2020 | 47 |
| 2019 | 41 |
| 2018 | 45 |
| 2017 | 53 |
| 2016 | 38 |
| 2015 | 39 |
| 2014 | 48 |
| 2013 | 32 |
| 2012 | 35 |
| 2011 | 20 |
| 2010 | 33 |
| 2009 | 30 |
| 2008 | 16 |
| 2007 | 14 |
| 2006 | 18 |
| 2005 | 13 |
| 2004 | 30 |
| 2003 | 10 |
| 2002 | 7 |
| 2001 | 11 |
| 2000 | 10 |
Use grade C
100 to 299 papers from 2021 to 2025.
Evidence grade B
At least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers.
- Papers on test-retest reliability
- 8
- Papers on reliability or validity
- 42
- Registered clinical studies using it as an outcome
- 5
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.
- Pictures
- 6 animals from Noto Emoji (cat, rabbit, pig, goat, sheep and cow) in scenes drawn by code for 4 actions: chase, push, pull and carry. The two pictures of a trial show the same action with the roles reversed.
- Sentences
- 24 events, each animal doing each action once and having it done once, each written as an active and a passive sentence: 48 sentences, in each of the eight participant languages.
- Trial
- The two pictures for 1,000 ms, then the sentence above them; a click or tap on a picture answers, within 20,000 ms of the sentence.
- Balance
- The matching picture is on the left on half the trials, with no more than 3 on one side in a row; half the scenes face left; no more than 3 of one voice in a row; an event's two sentences are at least 6 trials apart.
- Session
- 4 practice trials with feedback, then 48 trials (24 in the short form of 12 events) with none.
- Scoring
- Percent correct is the main score, with percent correct and median response time for active and passive sentences and the passive cost in each.
The pictures and sentences were made for this task from the two-picture design of Meyer, Mack and Thompson (2012); nothing comes from a published comprehension test. Both pictures show the same animals and the same action, so the choice turns on who does what, not on word knowledge.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: picture_matching_test from the scored run, a choice of the left or right picture within 20 seconds of the sentence (chance 50%; the matching picture is on each side on half the trials). Below chance invalidates and not above chance is flagged. One side on 90% of answers invalidates when accuracy is not above chance. A trial with no choice is a missed response: more than max(3, 25%) invalidate and any is flagged. Choices under 200 ms on at least half the trials invalidate. A run with fewer than 90% of its planned trials is invalid.
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. 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). A choice is due within each trial's window, so a trial with no choice is a missed response under the platform's rule for response windows. 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 sentences are shown in the session's language, so a session in any of the eight languages is scored the same way.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Test trials | 48 trials (standard) | 24 trials (short form), 48 trials (standard) |
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 Picture Matching Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Meyer, Mack & Thompson, 2012; Mack, Wei, Gutierrez & Thompson, 2016), with the standard settings (test trials: 48 trials). The main score was percent correct (test trials answered with the picture the sentence describes; a trial with no choice counts as wrong).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Meyer, A. M., Mack, J. E., & Thompson, C. K. (2012). Tracking passive sentence comprehension in agrammatic aphasia. Journal of Neurolinguistics, 25(1), 31-43. https://doi.org/10.1016/j.jneuroling.2011.08.001
- Mack, J. E., Wei, A. Z.-S., Gutierrez, S., & Thompson, C. K. (2016). Tracking sentence comprehension: Test-retest reliability in people with aphasia and unimpaired adults. Journal of Neurolinguistics, 40, 98-111. https://doi.org/10.1016/j.jneuroling.2016.06.001
- Salis, C., & Edwards, S. (2009). Tests of syntactic comprehension in aphasia: An investigation of task effects. Aphasiology, 23(10), 1215-1230. https://doi.org/10.1080/02687030802380165
Source papers
- Meyer AM, Mack JE, Thompson CK (2012). Tracking passive sentence comprehension in agrammatic aphasia. Journal of Neurolinguistics, 25(1):31-43.Paradigm source, Adults
- Mack JE, Wei AZS, Gutierrez S, Thompson CK (2016). Tracking sentence comprehension: Test-retest reliability in people with aphasia and unimpaired adults. Journal of Neurolinguistics, 40:98-111.Paradigm source, Adults
Questions
How long does the Picture Matching Task take?
About 5 to 8 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?
Percent Correct, Active Sentences Correct, Passive Sentences Correct, Passive Accuracy Cost, Median Response Time, Active Response Time, Passive Response Time, Passive Time Cost, Choices Under 100 ms, Missed Trials, Test Trials Recorded and Test Trials Set. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Picture Matching Task used in research?
222 papers from 2021 to 2025 name it in their title or abstract, down 1% compared with 2016 to 2020. That ranks it 115 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Picture Matching Task have?
In the published research, 8 papers report its test-retest reliability and 5 registered clinical studies use it as an outcome (evidence grade B in the ConductCognition test census).
Which settings can a researcher change?
Test trials (24 trials (short form), 48 trials (standard)). 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.