Executive function
Object Physics Task
Also called: intuitive physics, trajectory prediction, physical reasoning
Measure how people predict a rolling ball’s destination, with accuracy for straight paths and paths that bounce off a wall.
4-6 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
- 142
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Watch a ball roll across a table, then predict which half of the far edge it will reach. Scores physical prediction accuracy for paths with and without a wall bounce.
Constructs
- Intuitive physics
- Trajectory prediction
- Spatial reasoning
Research uses
- Research on physical prediction and spatial reasoning
How the paradigm works
Participants watch a ball roll across a table, then predict which half of the far edge it will reach after it disappears. The original scenes require either a straight continuation or one reflection from a side wall.
More about the paradigm
Each trial has 500 ms of fixation and 650 ms of visible motion. The complete path takes 2400 ms at constant speed, with ideal wall reflections. Destinations lie 28 or 64 pixels from the center. The standard plan has 64 trials, balanced across bounce count, destination side and margin, plus six practice trials with feedback.
Scores report accuracy overall, by bounce count and by margin, plus answer time and frame quality. Bounce cost is straight-path accuracy minus bounce accuracy. The seeded scene plan stays fixed regardless of responses. The task measures predictions of simple trajectories, not general physics knowledge.
Scores
- Prediction Accuracy%
- Share of usable predictions with the correct destination.
- Higher means more correct predictions
- Straight-path Accuracy%
- Accuracy when no wall bounce is needed.
- Higher is better
- Bounce Accuracy%
- Accuracy when one wall bounce is needed.
- Higher is better
- Bounce Cost%
- Straight-path accuracy minus bounce accuracy.
- Positive values mean lower bounce accuracy
- Near-center Accuracy%
- Accuracy for destinations 28 px from the center.
- Higher is better
- Far-center Accuracy%
- Accuracy for destinations 64 px from the center.
- Higher is better
- Median Answer Timems
- Median time from motion offset to a usable answer.
- Informational
- Anticipationscount
- Answers under 100 ms, excluded from accuracy.
- Lower is better
- Missing Answerscount
- Rows without a recognized answer.
- Lower is better
- Early Key Pressescount
- Arrow keys pressed before the answer screen, ignored.
- Informational
- Frame Intervalms
- Median measured interval between frames.
- About 16.7 ms at 60 Hz
- Dropped Frames%
- Skipped frames divided by expected display frames.
- Above 0.05 is flagged
- Motion Durationms
- Median measured motion duration.
- About 650 ms
- Completed Trialstrials
- Answered test trials in the selected run.
- Fewer than planned is flagged
- Planned Trialstrials
- Number of test trials planned.
- Informational
Use and evidence
142 papers from 2021 to 2025 name this task in their title or abstract, up 118% compared with 2016 to 2020. That ranks it 196 of 582 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 40 |
| 2024 | 25 |
| 2023 | 18 |
| 2022 | 23 |
| 2021 | 36 |
| 2020 | 11 |
| 2019 | 14 |
| 2018 | 14 |
| 2017 | 14 |
| 2016 | 12 |
| 2015 | 9 |
| 2014 | 19 |
| 2013 | 15 |
| 2012 | 12 |
| 2011 | 12 |
| 2010 | 9 |
| 2009 | 4 |
| 2008 | 8 |
| 2007 | 9 |
| 2006 | 8 |
| 2005 | 9 |
| 2004 | 4 |
| 2003 | 8 |
| 2002 | 9 |
| 2001 | 5 |
| 2000 | 5 |
Use grade C
100 to 299 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 13
- 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.
Note on the counts: Original constant-speed ball paths; two-choice destination judgments; no general physics score..
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.
- Display
- An original ball and table drawn in a 320 by 260 px panel. The ball travels at constant speed and reflects from the side walls.
- Timing
- 500 ms fixation, then 650 ms of visible motion. The full path would take 2400 ms. Answer after the ball disappears; no response deadline. Actual frame intervals, display duration and dropped frames are saved.
- Trials
- 64 standard trials: 8 in each combination of 0 or 1 bounce, left or right destination, and a margin of 28 or 64 px from the center. The setting allows 4 to 12 repeats, giving 32 to 96 trials.
- Practice
- 6 trials with 1200 ms feedback. Breaks after each 32 test trials.
- Response
- Left or right arrow, or click Left or Right, to predict which half of the far edge the ball will reach.
- Scores
- Prediction accuracy overall, by bounce count and by margin, plus bounce cost, answer time and frame quality. Answers under 100 ms leave accuracy and answer-time estimates.
Our original scenes use two destination choices and fixed trial sets. They adapt the published hidden-motion prediction method without reproducing its paddle responses or experimental scenes. Norms are not enabled.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: intuitive_physics_test from the fullest run, with the latest tied run selected. Destination judgments have two choices. Incomplete runs and poor display timing are flagged. Responses under 100 ms follow the timed-test rule.
Why these rules
Original trajectories test physical prediction with zero or one wall bounce. Accuracy is descriptive; no clinical threshold is applied. Brief motion makes measured display timing part of interpretation.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Trials per bounce, destination and margin | 8 | 4 to 12 |
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
- 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 Object Physics Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Smith and Vul, 2013), with the standard settings (trials per bounce, destination and margin: 8). The main score was prediction accuracy (share of usable predictions with the correct destination).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Smith, K. A., & Vul, E. (2013). Sources of uncertainty in intuitive physics. Topics in Cognitive Science, 5(1), 185-199. https://doi.org/10.1111/tops.12009Predicting a moving ball’s path while its later motion is hidden.
Source papers
- Smith, K. A., & Vul, E. (2013). Sources of uncertainty in intuitive physics. Topics in Cognitive Science, 5(1), 185-199.Paradigm source, Adults
Questions
How long does the Object Physics Task take?
About 4 to 6 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?
Prediction Accuracy, Straight-path Accuracy, Bounce Accuracy, Bounce Cost, Near-center Accuracy, Far-center Accuracy, Median Answer Time, Anticipations, Missing Answers, Early Key Presses, Frame Interval, Dropped Frames, Motion Duration, Completed Trials and Planned Trials. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Object Physics Task used in research?
142 papers from 2021 to 2025 name it in their title or abstract, up 118% compared with 2016 to 2020. That ranks it 196 of 582 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.39.0, counted 8 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Object Physics 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?
Trials per bounce, destination and margin (4 to 12). 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.