Processing Speed
Temporal Bisection Task
Also called: interval bisection task, duration bisection, temporal bisection paradigm
Measure how people judge short durations as participants learn a short and a long time, then call each new duration short or long.
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. Preview any test free after you sign up.
- Papers, 2021 to 2025
- 125
- Evidence grade
- C
- Registered clinical studies
- 2
What it measures
Measures how people judge short durations. Participants learn a short and a long time for which a disk stays on screen, then say whether each new time was closer to the short or the long one. The main score is the bisection point, the time judged long half the time.
Constructs
- Time perception
- Interval timing
- Duration discrimination
- Temporal categorization
Research uses
- Time perception research
- Aging and development research
- Within-study group comparison
How the paradigm works
The temporal bisection task measures how people judge durations of a fraction of a second to a few seconds (Church and Deluty, 1977; with people, Allan and Gibbon, 1991). A blue disk first appears for a short and a long time, each named three times, and the participant practises telling the two apart with feedback. Then the disk appears for seven durations from the short to the long one, and after each the participant presses F if the duration was closer to the short one or J if it was closer to the long one. Participants are asked not to count (Rattat and Droit-Volet, 2012).
More about the paradigm
The standard short and long times are 400 and 1,600 ms, the 4:1 ratio most human studies used (Kopec and Brody, 2010); a study can use 200 and 800 ms or 1,000 and 4,000 ms instead. The seven durations run from the short to the long time in equal steps on a log scale. Each comes 10 times (6 to 16 as a setting), in runs that hold every duration once, in an order drawn from the session's seed: 70 trials at the standard settings. Each trial is a cross for 500 ms, the disk for the duration, then the question until a key, with no time limit and no feedback. The time the disk was on screen is measured from the screen's frames and saved.
The main score is the bisection point: the duration answered long half the time, from a cumulative Gaussian fitted to the share of long answers at each duration on a log scale. The difference limen is half the distance between the durations answered long 25% and 75% of the time, and the Weber ratio, the difference limen over the bisection point, shows how finely durations are told apart: lower is finer (Droit-Volet and Zélanti, 2013). The bisection point is also given as a ratio to the geometric and the arithmetic mean of the two times. Where it falls depends on the ratio of the two times and the spacing of the durations between them (Allan, 2002; Kopec and Brody, 2010), so compare it within studies that use the same settings. Scores support research and do not diagnose.
Scores
- Bisection Pointms
- Duration answered long half the time, from the fitted curve.
- Lower means durations were called long sooner; no better direction
- Weber Ratioscore
- Difference limen divided by the bisection point.
- Lower means finer timing
- Difference Limenms
- Half the distance between the durations answered long 25% and 75% of the time.
- Lower means finer timing
- Bisection Point to Geometric Meanscore
- Bisection point divided by the geometric mean of the short and long times.
- 1 is bisection at the geometric mean
- Bisection Point to Arithmetic Meanscore
- Bisection point divided by the arithmetic mean of the short and long times.
- 1 is bisection at the arithmetic mean
- Long Answers by Duration%
- Share of long answers at each duration, keyed by its planned ms.
- The psychometric function
- Accuracy on the Short and Long Times%
- Share of right answers when the short or the long time itself was shown.
- Higher means the two times were remembered
- Lapse Rate%
- Fitted share of answers that did not follow the duration.
- Lower is better
- Fit R2%
- Share of the variance in long answers across durations that the fitted curve explains, weighted by answers.
- Higher means the curve fits better
- Median Response Timems
- Median time from the question to the answer, over answers at or after 100 ms.
- Informational
- Anticipationscount
- Answers under 100 ms from the question, left out of the scores.
- More than max(2, 20%) of trials makes the session invalid
- Early Key Pressescount
- Keys pressed while the cross or the disk showed, not taken as answers.
- Informational
- Frame Intervalms
- Median time between screen frames.
- Set by the screen
- Durations on Time%
- Share of answered trials whose disk showed within one frame of plan; only these are scored.
- Under 0.9 is flagged for review
- Completed Trialscount
- Test trials answered.
- Informational
- Planned Trialscount
- Test trials planned by the settings.
- Informational
Use and evidence
125 papers from 2021 to 2025 name this task in their title or abstract, up 7% compared with 2016 to 2020. That ranks it 165 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 26 |
| 2024 | 33 |
| 2023 | 17 |
| 2022 | 21 |
| 2021 | 28 |
| 2020 | 22 |
| 2019 | 21 |
| 2018 | 30 |
| 2017 | 18 |
| 2016 | 26 |
| 2015 | 22 |
| 2014 | 24 |
| 2013 | 20 |
| 2012 | 13 |
| 2011 | 13 |
| 2010 | 12 |
| 2009 | 13 |
| 2008 | 11 |
| 2007 | 6 |
| 2006 | 5 |
| 2005 | 4 |
| 2004 | 3 |
| 2003 | 2 |
| 2002 | 5 |
| 2001 | 9 |
| 2000 | 1 |
Use grade C
100 to 299 papers from 2021 to 2025.
Evidence grade C
At least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies.
- Papers on test-retest reliability
- 1
- Papers on reliability or validity
- 34
- Registered clinical studies using it as an outcome
- 2
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 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Learning
- The short and the long time, each named and shown 3 times, alternating; then 10 practice trials of the two with feedback for 1,000 ms.
- Trial
- On a dark panel, a cross for 500 ms, then a blue disk 120 px across for the duration, then the question until F (short) or J (long), with no time limit. A blank of 500 to 1,500 ms follows.
- Durations
- A short and a long time with a 4:1 ratio: 400 and 1,600 ms (standard), 200 and 800 ms, or 1,000 and 4,000 ms, and seven durations from one to the other in equal steps on a log scale: 400, 504, 635, 800, 1,008, 1,270 and 1,600 ms as standard.
- Trials
- Each duration 10 times (standard, 6 to 16), in runs that hold every duration once, in an order drawn from the session's seed: 70 trials, with no feedback.
- Fixed times
- Every duration is set before the session and shown in whole screen frames, and the time on screen is measured. No answer changes the next trial.
- Score
- The bisection point and the Weber ratio, from a cumulative Gaussian fitted to the long answers on log duration by maximum likelihood. Trials whose disk ran more than one frame off plan, and answers under 100 ms, leave the score.
Where the bisection point falls depends on the ratio of the two times and the spacing of the durations between them, so compare it within studies that use the same settings. The disk is timed in screen frames, so the test runs on a desktop or laptop. 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: temporal_bisection_test from the scored run, F (short) or J (long) after the disk, with no time limit, graded from the saved key and duration. Only the two anchors have a right answer (chance 50%; each comes equally often): below chance invalidates and not above chance is flagged. One key on 90% of answers invalidates when the anchors are not above chance. Answer times run from the question; responses under 100 ms follow the timed-test rule. No bisection point from the fit, a run short of its planned trials, or fewer than 90% of durations within one frame of plan is flagged for review.
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). The durations between the anchors have no right answer, so only the anchors are judged against chance. An answer under 100 ms from the question was pressed before the question could be read. The durations are the design, so trials whose disk ran late leave the score.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Short and long times | 400 and 1,600 ms (standard) | 400 and 1,600 ms (standard), 200 and 800 ms, 1,000 and 4,000 ms |
| Trials per duration | 10 | 6 to 16 |
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 Temporal Bisection Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Church and Deluty, 1977; Allan and Gibbon, 1991), with the standard settings (short and long times: 400 and 1,600 ms; trials per duration: 10). The main score was bisection point (duration answered long half the time, from the fitted curve).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Church, R. M., & Deluty, M. Z. (1977). Bisection of temporal intervals. Journal of Experimental Psychology: Animal Behavior Processes, 3(3), 216-228. https://doi.org/10.1037/0097-7403.3.3.216The task: learn a short and a long duration, then classify the durations between them.
- Allan, L. G., & Gibbon, J. (1991). Human bisection at the geometric mean. Learning and Motivation, 22(1-2), 39-58. https://doi.org/10.1016/0023-9690(91)90016-2The task with people.
- Allan, L. G. (2002). The location and interpretation of the bisection point. The Quarterly Journal of Experimental Psychology Section B, 55(1b), 43-60. https://doi.org/10.1080/02724990143000162How the range, the ratio and the spacing of the durations move the bisection point.
- Kopec, C. D., & Brody, C. D. (2010). Human performance on the temporal bisection task. Brain and Cognition, 74(3), 262-272. https://doi.org/10.1016/j.bandc.2010.08.006The 4:1 ratio and the two spacings of human studies.
- Droit-Volet, S., & Zélanti, P. S. (2013). Development of time sensitivity and information processing speed. PLoS ONE, 8(8), e71424. https://doi.org/10.1371/journal.pone.0071424The difference limen and the Weber ratio, with a disk on a screen.
- Rattat, A.-C., & Droit-Volet, S. (2012). What is the best and easiest method of preventing counting in different temporal tasks? Behavior Research Methods, 44(1), 67-80. https://doi.org/10.3758/s13428-011-0135-3Asking participants not to count.
Source papers
- Church and Deluty (1977): the task. Allan and Gibbon (1991): the task with people. Kopec and Brody (2010): a meta-analysis of 148 human experiments on where the bisection point falls. Droit-Volet and Zélanti (2013): the difference limen and the Weber ratio.Paradigm source, Adults
Questions
How long does the Temporal Bisection 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?
Bisection Point, Weber Ratio, Difference Limen, Bisection Point to Geometric Mean, Bisection Point to Arithmetic Mean, Long Answers by Duration, Accuracy on the Short and Long Times, Lapse Rate, Fit R2, Median Response Time, Anticipations, Early Key Presses, Frame Interval, Durations on Time, Completed Trials and Planned Trials. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Temporal Bisection Task used in research?
125 papers from 2021 to 2025 name it in their title or abstract, up 7% compared with 2016 to 2020. That ranks it 165 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Temporal Bisection Task have?
In the published research, 1 paper reports its test-retest reliability and 2 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Short and long times (400 and 1,600 ms (standard), 200 and 800 ms, 1,000 and 4,000 ms) and trials per duration (6 to 16). 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.