Episodic Memory
Boundary Extension Task
Also called: camera distance paradigm, scene boundary memory, boundary extension paradigm, wide-angle memory
Measure how far people remember a scene as wider than it was: they study photographs, then judge each one again as the same view, closer up or farther away.
8-10 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
- 65
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Participants study 20 photographs of everyday scenes, half as close-ups and half as wider views, then see each again and judge whether it is the same view, closer up or farther away. Measures how far people remember a scene as wider than it was.
Constructs
- Scene memory
- Boundary extension
- Visual memory
- Scene construction
Research uses
- Research on scene memory and the hippocampus
- Research on aging and visual memory
- Within-study comparison of memory distortions
How the paradigm works
The Boundary Extension Task measures a memory error that almost everyone makes: people remember a scene as having shown more of its surroundings than it did. Intraub and Richardson (1989) showed photographs of a main object in a natural setting, some as close-ups and some as wider views, and later showed each again, the same view or the other one, to be rated from much closer up (-2) through the same (0) to much farther away (+2) than before. The same close-up looked closer up than the one remembered (a mean of -0.62), the same wide view barely did (-0.12), and the change from a close-up to its wide view looked smaller (+0.23) than the change from a wide view to its close-up (-1.28): the close-ups had been remembered as wider.
More about the paradigm
This version follows that camera distance paradigm with an immediate test, as Ménétrier, Iralde and Le Bohec (2019) ran it with younger and older adults. The person studies 20 photographs of everyday scenes, each for 10 s (standard; 4 to 15), after a fixation point and followed by visual noise and a blank screen. Four sets of five scenes take the four conditions in an order drawn from the session's seed: a close-up tested by itself, a wide view tested by itself, a close-up tested by its wide view, and a wide view tested by its close-up. Both views of a scene are cut from one photograph by the task's code: the close view is the middle of the wide view enlarged 1.2 times, within the 8% to 21% enlargements of Intraub and Dickinson (2008). At test each scene comes back in a new order and is rated on the five-point scale, with no time limit. The photographs are CC0 pictures from Unsplash, copied to Wikimedia Commons, with no person in them.
The main score is the mean rating of the views tested by themselves: below 0, the same view looked closer up than the one remembered, which is boundary extension. The changed views check that the scale was used as asked: a wide view of a studied close-up should look farther away than a close-up of a studied wide view. Extension is strongest for close views of objects; Bainbridge and Baker (2020) found it for 93.2% of object-oriented pictures but no more often than contraction for wide scenes, so the result depends on the pictures, and scores compare people who saw the same set. People with hippocampal amnesia made fewer of the closer-up errors than controls (Mullally, Intraub and Maguire 2012). The scores support research and do not diagnose.
Scores
- Rating of the Same Viewscore
- Mean rating, from -2 (much closer up) to +2 (much farther away), of scenes tested in the view that was studied, close-ups and wide views together.
- Below 0 means the same view looked closer up than remembered: boundary extension
- Rating of the Same Close-Upscore
- Mean rating of close-ups tested by themselves.
- Usually the most negative mean
- Rating of the Same Wide Viewscore
- Mean rating of wide views tested by themselves.
- Informational
- Rating of Close-Up to Wide Viewscore
- Mean rating when a studied close-up came back as its wide view.
- Above 0 means the wider view was seen as farther away
- Rating of Wide View to Close-Upscore
- Mean rating when a studied wide view came back as its close-up.
- Below 0 means the closer view was seen as closer up
- Changed Views Told Apartscore
- The close-up to wide view mean minus the wide view to close-up mean.
- Above 0 when the changed views were judged in the right directions; 0 or less is flagged
- Same Views Rated Closer Up%
- Share of the scenes tested in their studied view that were rated a little or much closer up.
- Informational: the share of boundary extension answers
- Same Views Rated the Same%
- Share of the scenes tested in their studied view that were rated the same.
- Informational
- Median Rating Timems
- Median time from a test picture to its rating, leaving out ratings under 100 ms.
- Informational; ratings have no time limit
- Rated Scenescount
- Test pictures rated in the scored run.
- Informational
- Anticipationscount
- Ratings under 100 ms from the test picture, still scored.
- Lower is better
Use and evidence
65 papers from 2021 to 2025 name this task in their title or abstract, up 20% compared with 2016 to 2020. That ranks it 273 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 17 |
| 2024 | 13 |
| 2023 | 10 |
| 2022 | 14 |
| 2021 | 11 |
| 2020 | 9 |
| 2019 | 12 |
| 2018 | 10 |
| 2017 | 10 |
| 2016 | 13 |
| 2015 | 12 |
| 2014 | 10 |
| 2013 | 15 |
| 2012 | 15 |
| 2011 | 17 |
| 2010 | 20 |
| 2009 | 3 |
| 2008 | 6 |
| 2007 | 7 |
| 2006 | 8 |
| 2005 | 11 |
| 2004 | 12 |
| 2003 | 4 |
| 2002 | 4 |
| 2001 | 2 |
| 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
- 3
- Registered clinical studies using it as an outcome
- 0
Note on the counts: a few counted papers use the words in other fields.
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 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Scenes
- 20 photographs of a main object in its everyday setting, with no person, in four sets of five, and one more scene for the example. They are CC0 pictures from Unsplash, copied to Wikimedia Commons.
- Views
- Both views of a scene are cut from one photograph by the task's code: the wide view is the whole 4:3 frame, and the close view is its middle enlarged 1.2 times.
- Study
- Each picture for 10 s (standard; 4 to 15), after a fixation point for 600 ms and followed by visual noise for 500 ms and a blank screen for 1,000 ms, in an order drawn from the session's seed. An example picture shows first how they appear.
- Conditions
- The four sets take the four conditions in one of four turns drawn from the seed, 5 scenes each: a close-up tested by itself, a wide view tested by itself, a close-up tested by its wide view, and a wide view tested by its close-up.
- Test
- Right after the study list, each scene again in a new order, rated on five steps from much closer up (-2) through the same (0) to much farther away (+2) than before, with no time limit. The instructions first show the example scene closer up and farther away.
- Scores
- The mean rating of the views tested by themselves, the main score, where below 0 is boundary extension; the mean of each condition; and how far the changed views were told apart.
Extension depends on the pictures: it is strongest for close views of objects and can give way to contraction for wide scenes (Bainbridge and Baker 2020), so compare people who saw the same set. Intraub and Richardson (1989) showed each picture for 15 s and tested 48 hours later; here the test follows within minutes, as in Ménétrier and colleagues (2019), and the standard study time is 10 s. Confidence ratings are left out. The scores support research and do not diagnose.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: boundary_extension_test from the scored run, one rating per scene on the five-point scale. The same view has no right answer to compare with chance, because remembering more than was shown is what the test measures. Fewer than 3 ratings in any of the four conditions, or the same answer on every picture, invalidate. Changed views rated in the wrong directions, or not apart, are flagged for review. Ratings under 200 ms on at least half the trials invalidate.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. A boundary extension is an error by design: a same view rated closer up is the expected result, not a mistake to count against the person. The changed views show whether the scale was used as asked: a wide view of a studied close-up should look farther away than a close-up of a studied wide view, as young and older participants alike rated them in Ménétrier and colleagues (2019). 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. A hidden tab lets the study, delay or response timers run without the participant, so a percentile is withheld after one.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Time per study picture (seconds) | 10 | 4 to 15 |
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 Boundary Extension Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Intraub & Richardson, 1989), with the standard settings (time per study picture (seconds): 10). The main score was rating of the same view (mean rating, from -2 (much closer up) to +2 (much farther away), of scenes tested in the view that was studied, close-ups and wide views together).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Intraub, H., & Richardson, M. (1989). Wide-angle memories of close-up scenes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15(2), 179-187. https://doi.org/10.1037/0278-7393.15.2.179The camera distance paradigm: close-ups and wide views, tested by themselves or by the other view, on a five-point scale.
- Intraub, H., Bender, R. S., & Mangels, J. A. (1992). Looking at pictures but remembering scenes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 18(1), 180-191. https://doi.org/10.1037/0278-7393.18.1.180Main objects against natural backgrounds, and extension for close-ups more than for wide views.
- Intraub, H., Gottesman, C. V., Willey, E. V., & Zuk, I. J. (1996). Boundary extension for briefly glimpsed photographs: Do common perceptual processes result in unexpected memory distortions? Journal of Memory and Language, 35(2), 118-134. https://doi.org/10.1006/jmla.1996.0007Extension after short views, of 250 ms and 4 s.
- Intraub, H., & Dickinson, C. A. (2008). False memory 1/20th of a second later: What the early onset of boundary extension reveals about perception. Psychological Science, 19(10), 1007-1014. https://doi.org/10.1111/j.1467-9280.2008.02192.xClose-ups made by enlarging wider views 8% to 21%, and the labels of the scale.
- Mullally, S. L., Intraub, H., & Maguire, E. A. (2012). Attenuated boundary extension produces a paradoxical memory advantage in amnesic patients. Current Biology, 22(4), 261-268. https://doi.org/10.1016/j.cub.2012.01.001Fewer closer-up errors in patients with hippocampal damage.
- Ménétrier, E., Iralde, L., & Le Bohec, L. (2019). Spatial layout extrapolation in aging: Underlying cognitive and executive mechanisms. Visual Cognition, 27(9-10), 668-686. https://doi.org/10.1080/13506285.2019.1634663The immediate test, 20 scenes in four conditions, and the fixation point, noise and blank around each picture.
- Bainbridge, W. A., & Baker, C. I. (2020). Boundaries extend and contract in scene memory depending on image properties. Current Biology, 30(3), 537-543.e3. https://doi.org/10.1016/j.cub.2019.12.004Extension for 93.2% of object-oriented pictures, and as much contraction as extension for wide scenes.
Source papers
- Intraub H, Richardson M (1989). Wide-angle memories of close-up scenes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15(2):179-187.Paradigm source
- Ménétrier E, Iralde L, Le Bohec L (2019). Spatial layout extrapolation in aging: underlying cognitive and executive mechanisms. Visual Cognition, 27(9-10):668-686.Paradigm source, Younger and older adults
Questions
How long does the Boundary Extension Task take?
About 8 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?
Rating of the Same View, Rating of the Same Close-Up, Rating of the Same Wide View, Rating of Close-Up to Wide View, Rating of Wide View to Close-Up, Changed Views Told Apart, Same Views Rated Closer Up, Same Views Rated the Same, Median Rating Time, Rated Scenes and Anticipations. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Boundary Extension Task used in research?
65 papers from 2021 to 2025 name it in their title or abstract, up 20% compared with 2016 to 2020. That ranks it 273 of 495 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Boundary Extension 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?
Time per study picture (seconds) (4 to 15). 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.