Attention
Symmetry Detection Task
Also called: mirror symmetry detection, reflection symmetry, symmetry perception, symmetric dot patterns
Measure mirror-symmetry detection in brief dot patterns. Compare accuracy and response time across axes and noise levels.
6-8 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
- 63
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measure how quickly and accurately people see mirror symmetry. Dot patterns flash for 200 ms. Participants judge whether a pattern has mirror symmetry or is random. Four axes and three fixed noise levels show how orientation and noise affect detection.
Constructs
- Mirror-symmetry perception
- Perceptual organization
- Visual form processing
Research uses
- Visual perception research
- Perceptual organization research
How the paradigm works
The Symmetry Detection Task measures mirror-symmetry perception. A pattern of 80 white dots flashes for 200 ms. Participants press the left arrow for symmetry or the right arrow for a random pattern. Symmetric patterns reflect across a vertical, horizontal or one of two diagonal axes. At fixed noise levels, 0%, 25% or 50% of the dots are random. Controls have 80 random dots with the same size, circular field and minimum spacing. No axis is cued.
More about the paradigm
The seeded plan crosses four axes, three noise levels and two answers. Six repetitions give 144 test trials, after eight practice trials with feedback. Each trial starts with a 500 ms cross. The 2000 ms answer window starts at pattern onset; the pattern disappears after 200 ms and the blank remains until the window ends, even after an answer. A break comes halfway. All conditions and original dot positions are set before the session. Nothing adapts to performance.
Accuracy and correct response time are reported overall and by axis and noise. Random controls are assigned to matching condition cells; they have no physical symmetry axis. Target displays more than one screen frame from 200 ms leave accuracy and speed scores. Omissions and responses under 100 ms count as accuracy errors and leave response-time medians.
Scores
- Accuracy%
- Share correct on trials shown within one screen frame of plan. Omissions and responses under 100 ms count as errors.
- Higher is better
- Accuracy by Axis%
- Accuracy for vertical, horizontal and both diagonal conditions, each including matched random controls.
- Compare the four orientations
- Accuracy by Noise%
- Accuracy at 0%, 25% and 50% noise, each including matched random controls.
- Compare resistance to noise
- Median Correct Response Timems
- Median correct response time from pattern onset, from 100 to 2000 ms on displays shown on time.
- Shorter means faster detection
- Response Time by Axisms
- Correct-response medians in each axis condition, including its random controls.
- Compare speed beside accuracy
- Response Time by Noisems
- Correct-response medians at each noise level, including its random controls.
- Compare speed beside accuracy
- Hit Rate%
- Timely symmetry answers divided by symmetric trials shown on time. Omissions remain in the denominator.
- Higher means more signal patterns detected
- False Alarm Rate%
- Timely symmetry answers divided by random trials shown on time. Omissions remain in the denominator.
- Lower means fewer random patterns called symmetric
- Anticipationscount
- Arrow-key answers under 100 ms after pattern onset.
- Excluded from speed scores
- Missed Responsescount
- Patterns without an arrow-key answer within 2000 ms.
- Fewer is better
- Early Key Pressescount
- Arrow presses before the pattern appeared.
- Not taken as answers
- Displays on Time%
- Share of targets shown within one screen frame of 200 ms.
- Below 0.9 is flagged
- Completed Trialstrials
- Test trials recorded in the selected run, including timed-out trials.
- Compare with planned trials
- Planned Trialstrials
- Number of test trials planned.
- Informational
Use and evidence
63 papers from 2021 to 2025 name this task in their title or abstract, down 3% compared with 2016 to 2020. That ranks it 331 of 582 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 10 |
| 2024 | 13 |
| 2023 | 14 |
| 2022 | 10 |
| 2021 | 16 |
| 2020 | 10 |
| 2019 | 13 |
| 2018 | 16 |
| 2017 | 15 |
| 2016 | 11 |
| 2015 | 9 |
| 2014 | 19 |
| 2013 | 11 |
| 2012 | 11 |
| 2011 | 13 |
| 2010 | 23 |
| 2009 | 4 |
| 2008 | 5 |
| 2007 | 12 |
| 2006 | 5 |
| 2005 | 10 |
| 2004 | 8 |
| 2003 | 12 |
| 2002 | 10 |
| 2001 | 8 |
| 2000 | 8 |
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
- 9
- 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: includes face symmetry studies.
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.
- Sequence
- 500 ms fixation, then 80 dots for 200 ms. The answer window is 2000 ms from pattern onset. The pattern is followed by a blank until the window ends; the first arrow-key answer is kept.
- Patterns
- Four axes: vertical, horizontal and both diagonals. Noise replaces 0%, 25% or 50% of signal dots with random dots. All controls are random, with the same 80 dots, 4 px dot diameter, 220 px circular field and 6 px minimum spacing. Axes are not cued.
- Trials
- 4 axes × 3 noise levels × 2 answers × 6 repeats = 144 test trials. Repeats range from 4 to 10, giving 96 to 240 trials. Each condition has balanced symmetric and random answers. A break comes halfway.
- Practice
- 8 trials with no noise in the symmetric patterns and 800 ms feedback. Main trials have no feedback.
- Scores
- Accuracy and median correct response time, overall and by axis and noise, plus hit and false-alarm rates. Each condition includes its assigned random controls. There is no adaptive threshold.
- Timing
- Frame-based displays save measured target duration. Targets more than one frame off 200 ms leave accuracy and speed. Omissions and responses under 100 ms are accuracy errors and leave speed scores. Fewer than 90% of displays on time is flagged.
- Interpretation
- Compare accuracy beside speed. Norms use only ConductCognition platform observations; the method papers are not percentile tables.
An original fixed-trial adaptation of dot-pattern symmetry methods. The 80 dots, noise levels, 200 ms exposure, 2000 ms response window and quality limits are ConductScience choices. Positions and sizes use screen pixels without visual-angle calibration.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scores test patterns from the run with the most trials, ties going to the latest run. Chance is 1 in 2. A short run or fewer than 90% of displays on time is flagged. Omissions and responses under 100 ms follow the timed-test rule.
Why these rules
Fixed exposure keeps comparisons across axes and noise levels consistent. Late displays leave accuracy and response-time scores. Timing and completion limits are ConductScience quality rules.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Trials per condition | 6 | 4 to 10 |
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.
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Symmetry Detection Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Barlow and Reeves, 1979; Wenderoth, 1994), with the standard settings (trials per condition: 6). The main score was accuracy (share correct on trials shown within one screen frame of plan).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Barlow, H. B., & Reeves, B. C. (1979). The versatility and absolute efficiency of detecting mirror symmetry in random dot displays. Vision Research, 19(7), 783-793. https://doi.org/10.1016/0042-6989(79)90154-8Detecting mirror structure in patterns that mix symmetric and random dots.
- Wenderoth, P. (1994). The Salience of Vertical Symmetry. Perception, 23(2), 221-236. https://doi.org/10.1068/p230221Comparison of symmetry detection across axes.
- Chen, C.-C., & Tyler, C. W. (2010). Symmetry: Modeling the Effects of Masking Noise, Axial Cueing and Salience. PLOS ONE, 5(4), e9840. https://doi.org/10.1371/journal.pone.0009840White-dot symmetry displays, matched random controls, noise and four axes.
Source papers
- Barlow, H. B., & Reeves, B. C. (1979). The versatility and absolute efficiency of detecting mirror symmetry in random dot displays. Vision Research, 19(7), 783-793.Paradigm source, Adults
- Wenderoth, P. (1994). The Salience of Vertical Symmetry. Perception, 23(2), 221-236.Paradigm source, Adults
- Chen, C.-C., & Tyler, C. W. (2010). Symmetry: Modeling the Effects of Masking Noise, Axial Cueing and Salience. PLOS ONE, 5(4), e9840.Paradigm source, Adults
Questions
How long does the Symmetry Detection Task take?
About 6 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?
Accuracy, Accuracy by Axis, Accuracy by Noise, Median Correct Response Time, Response Time by Axis, Response Time by Noise, Hit Rate, False Alarm Rate, Anticipations, Missed Responses, Early Key Presses, Displays on Time, Completed Trials and Planned Trials. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Symmetry Detection Task used in research?
63 papers from 2021 to 2025 name it in their title or abstract, down 3% compared with 2016 to 2020. That ranks it 331 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 Symmetry Detection 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).
Which settings can a researcher change?
Trials per condition (4 to 10). 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.