Attention
Pitch Illusions Task
Also called: tritone paradox, Shepard tones, Shepard tone illusion, octave illusion, pitch circularity, ambiguous pitch judgments
Measure how a listener hears ambiguous pitch: Shepard tone pairs judged up or down, including the tritone paradox, then the octave illusion through headphones.
8-11 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
- 51
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures how a listener organizes pitch. Through headphones, participants judge whether pairs of Shepard tones go up or down, including half-octave pairs that different people hear differently, then say where they hear the high and low tones of the octave illusion. The main score is how strongly the pitch class of the first tone sets the answer.
Constructs
- Pitch perception
- Pitch circularity
- Auditory illusions
- Auditory lateralization
Research uses
- Music cognition and pitch perception research
- Research on language and pitch perception
- Within-study group comparison
How the paradigm works
The Pitch Illusions Task measures how a listener organizes pitch, with sounds that different people hear differently. A Shepard tone is made of components an octave apart under a fixed envelope, so it has a pitch class but no clear height, and a tone a semitone or two above another is heard going up, the shorter way round the circle of pitch classes (Shepard, 1964). A pair half an octave apart has no shorter way. In the tritone paradox (Deutsch, 1986) listeners hear such pairs go down from some pitch classes and up from the opposite ones: the judgments of 22 of 29 listeners split the circle in two (Deutsch, Kuyper and Fisher, 1987), and where the highest region falls differs between listeners and with the language they grew up with (Deutsch, 1991). In the octave illusion (Deutsch, 1974) tones of 400 and 800 Hz alternate, the high tone in one ear while the low tone is in the other, and most right handers hear the high tone on the right and the low tone on the left.
More about the paradigm
The test starts with the sound setup: the participant turns the device's volume until a steady tone is comfortable, and a six-trial headphone check (Woods, Siegel, Traer and McDermott, 2017) confirms that headphones are in use. By default a failed check ends the test, and the check's result is saved either way. In the first part each trial plays two Shepard tones of 500 ms, the second right after the first, and the participant answers Up or Down with no time limit. Each tone has one component of its pitch class in each octave of a six-octave envelope, A(f) = 0.5 - 0.5 cos(2 pi log2(f / fmin) / 6), centered on C4, F#4, C5 or F#5, as in Deutsch, Kuyper and Fisher (1987). A round plays the half-octave pair of each of the 12 pitch classes under each envelope, 48 pairs, and 16 check pairs a semitone or two up or down, in one order drawn from the session's seed: 64 pairs at the standard 1 round (up to 3). Four check pairs with feedback come first. In the second part each trial plays 6 seconds of 400 and 800 Hz tones of 250 ms in turn, the high tone in one ear while the low tone is in the other, changing at zero crossings with no ramps as Deutsch (1983) made them, and the participant answers high on the right and low on the left, the other way round, or something else. Six such patterns, half starting with the high tone on each side, come mixed with two controls whose high tones really play in one ear only, after the two controls with feedback. Every sound is set before the session; nothing adapts to the answers. Desktop, laptop, tablet or phone, with headphones; about 8 to 11 minutes.
The main score is the strength of the tritone pattern: the share of half-octave pairs heard going down at each pitch class of the first tone, summarized by its first circular harmonic around the pitch-class circle and scaled so that a clean half circle, six neighbouring classes heard going down and the other six going up, is 1. Where that harmonic peaks gives the region heard as highest, in semitones above C, with the chance of a pattern as strong if the answers did not depend on the pitch class. The share of check pairs answered the shorter way and the shares of each octave answer come with it. No answer to a half-octave pair or an illusion pattern is scored as right; the check pairs and the controls are, and they flag a session whose answers did not follow the tones or whose earphones were on the wrong ears.
Scores
- Tritone Pattern Strengthscore
- The first circular harmonic of the share of half-octave pairs heard going down at each pitch class of the first tone, scaled so that a clean half circle is 1.
- 0 to 1; higher means the pitch class set the answer more; not a better or worse score
- Peak Pitch Classsemitones
- Where that harmonic peaks, the region heard as highest, in semitones above C (0 up to 12).
- Differs between listeners; informational
- Pattern Chance%
- The chance of a pattern at least this strong if the answers did not depend on the pitch class.
- 0.05 or above is flagged: the peak says little
- Heard Going Down%
- Share of half-octave pairs heard going down.
- Informational
- Going Down by Pitch Class%
- Share of half-octave pairs heard going down for each pitch class of the first tone, keyed by its name.
- The profile behind the pattern
- Check Pairs Right%
- Share of pairs a semitone or two apart answered in the direction of their shorter step.
- Under 0.75 is flagged for review
- High on the Right%
- Share of illusion patterns heard as high on the right and low on the left.
- Informational
- High on the Left%
- Share of illusion patterns heard as high on the left and low on the right.
- Informational
- Something Else%
- Share of illusion patterns heard as neither.
- Informational
- Controls Right%
- Share of control patterns, whose high tones play in one ear only, answered as they played.
- Under 1 is flagged: the earphones may be on the wrong ears
- Median Answer Timems
- Median time from when the answers opened, after the sound, to the answer.
- Informational; the answer has no time limit
- Anticipationscount
- Answers within 100 ms of the answers opening, which leave the score.
- Half or more of the answers makes the session invalid
- Completed Trialstrials
- Trials answered in the scored run, over both parts.
- Fewer than planned is flagged for review
- Planned Trialstrials
- Trials the run set out to play.
- Informational
- Headphone Check Passed0/1
- 1 when the headphone check passed (at least 5 of 6 right), 0 when it did not.
- 0 stops the test unless the study set it to continue
- Headphone Check Correctcount
- Headphone check trials answered right, of 6.
- 5 or 6 passes
- Reported Output Delayms
- Output delay the device reported to the browser; empty when the browser does not report it.
- Informational
Use and evidence
51 papers from 2021 to 2025 name this task in their title or abstract, up 42% compared with 2016 to 2020. That ranks it 304 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 15 |
| 2024 | 15 |
| 2023 | 6 |
| 2022 | 5 |
| 2021 | 10 |
| 2020 | 5 |
| 2019 | 6 |
| 2018 | 11 |
| 2017 | 8 |
| 2016 | 6 |
| 2015 | 3 |
| 2014 | 6 |
| 2013 | 6 |
| 2012 | 5 |
| 2011 | 8 |
| 2010 | 5 |
| 2009 | 3 |
| 2008 | 5 |
| 2007 | 7 |
| 2006 | 2 |
| 2005 | 3 |
| 2004 | 12 |
| 2003 | 2 |
| 2002 | 8 |
| 2001 | 5 |
| 2000 | 5 |
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
- 7
- Registered clinical studies using it as an outcome
- 0
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.
- Sound setup
- The participant turns the device's volume until a steady 1 kHz tone is comfortable, then takes a 6-trial headphone check (Woods and colleagues, 2017). By default a failed check ends the test; its result is saved either way.
- Shepard tones
- One sine component of the pitch class in each octave of a six-octave envelope, A(f) = 0.5 - 0.5 cos(2 pi log2(f / fmin) / 6), centered on C4, F#4, C5 or F#5 (Deutsch and colleagues, 1987). Components under 1% of the peak's amplitude are left out. Each tone lasts 500 ms with 10 ms ramps.
- Pairs
- Press Play to hear two Shepard tones, the second right after the first, and choose Up or Down, with no time limit.
- Trials
- Each round plays the half-octave pair of every pitch class under every envelope, 48 pairs, and 16 check pairs, 1 or 2 semitones up or down: 64 pairs in an order drawn from the session's seed, with a pause screen every 32. 1 round (standard; 1 to 3).
- Octave illusion
- 6 seconds of 400 and 800 Hz tones of 250 ms in turn with no ramps, the high tone in one ear while the low tone is in the other. Choose high on the right and low on the left, the other way round, or something else. 6 such patterns and 2 controls whose high tones play in one ear only.
- Practice
- 4 check pairs, then the 2 controls, each with feedback for 1,000 ms.
- Fixed sounds
- Every sound is set before the session, and no answer changes what follows: there is no staircase.
- Score
- Tritone pattern strength: the first circular harmonic of the share of half-octave pairs heard going down at each pitch class, scaled so that a clean half circle is 1, with its peak pitch class and its chance under guessing. Answers under 100 ms leave the score.
No answer to a half-octave pair or an illusion pattern is scored as right: listeners differ, and where the highest pitch classes fall varies with the language a listener grew up with. Check pairs and controls flag a session whose answers did not follow the tones or whose earphones were on the wrong ears. Levels are relative to the volume the participant chose, and headphones differ, so compare results within studies run the same way. The test needs normal hearing and headphones, wired if possible: wireless headphones can change the sound. It is a research measure, not a hearing test, and it claims no norm. No percentile is given.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: pitch_illusions_pair and pitch_illusions_octave from the scored run, buttons after the sound has played, with no time limit. Accuracy is judged against chance on the check pairs a semitone or two apart (chance 1 in 2) and the control patterns (chance 1 in 3); the half-octave pairs and the illusion patterns have no right answer, since listeners hear them differently. Below chance invalidates and not above chance is flagged; one button on 90% of all answers invalidates when the checks are not above chance; answers under 100 ms invalidate when they are at least half. A headphone check that stopped the test invalidates; a failed check with the test run anyway, check pairs right on fewer than 75%, a control pattern answered otherwise than it played, a tritone pattern no stronger than guessing, or a run short of its planned trials 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 answers open once the sound has played and have no time limit, so an answer under 100 ms was pressed before they opened, and only a session made mostly of such answers is invalid. Woods, Siegel, Traer and McDermott (2017) let only listeners who passed the headphone check go on, because a listener on loudspeakers hears the sounds differently; a session that ran after a failed check keeps its scores but not its percentile. Pairs a semitone or two apart are heard going the shorter way round by nearly every listener (Shepard, 1964), and a control pattern plays its high tones in one ear only, so wrong answers to them mean the tones, the ears or the task were not followed. The half-octave pairs and the octave illusion are heard differently by different listeners for reasons that include the language they grew up with and their handedness (Deutsch, 1991; Deutsch, 1974), so no answer to them is wrong.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| When the headphone check fails | Stop the test | Stop the test, Continue the test |
| Rounds of half-octave pairs | 1 | 1 to 3 |
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. Needs headphones. 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 Pitch Illusions Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Shepard, 1964; Deutsch, 1974, 1986), with the standard settings (when the headphone check fails: Stop the test; rounds of half-octave pairs: 1). The main score was tritone pattern strength (the first circular harmonic of the share of half-octave pairs heard going down at each pitch class of the first tone, scaled so that a clean half circle is 1).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Shepard, R. N. (1964). Circularity in judgments of relative pitch. The Journal of the Acoustical Society of America, 36(12), 2346-2353. https://doi.org/10.1121/1.1919362Shepard tones, heard going the shorter way round the circle of pitch classes.
- Deutsch, D. (1986). A musical paradox. Music Perception, 3(3), 275-280. https://doi.org/10.2307/40285337The tritone paradox: half-octave pairs of Shepard tones heard going up or down by pitch class.
- Deutsch, D., Kuyper, W. L., & Fisher, Y. (1987). The tritone paradox: Its presence and form of distribution in a general population. Music Perception, 5(1), 79-92. https://doi.org/10.2307/40285386The envelope and its four centers, the 48 pairs, and the share heard going down at each pitch class.
- Deutsch, D. (1991). The tritone paradox: An influence of language on music perception. Music Perception, 8(4), 335-347. https://doi.org/10.2307/40285517Where the highest pitch classes fall differs with the language a listener grew up with.
- Deutsch, D. (1974). An auditory illusion. Nature, 251(5473), 307-309. https://doi.org/10.1038/251307a0The octave illusion: 400 and 800 Hz tones of 250 ms in turn, a different one in each ear.
- Deutsch, D. (1983). The octave illusion in relation to handedness and familial handedness background. Neuropsychologia, 21(3), 289-293. https://doi.org/10.1016/0028-3932(83)90047-7Tones with no rise and fall times, and a forced choice of where the high and low tones were heard.
- Woods, K. J. P., Siegel, M. H., Traer, J., & McDermott, J. H. (2017). Headphone screening to facilitate web-based auditory experiments. Attention, Perception, & Psychophysics, 79(7), 2064-2072. https://doi.org/10.3758/s13414-017-1361-2The headphone check before the test.
Source papers
- Deutsch D, Kuyper WL, Fisher Y (1987). The tritone paradox: Its presence and form of distribution in a general population. Music Perception, 5(1):79-92.N 29, Adults
- Deutsch D (1974). An auditory illusion. Nature, 251(5473):307-309.N 86, Adults
Questions
How long does the Pitch Illusions Task take?
About 8 to 11 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops, tablets and phones. Needs headphones. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Tritone Pattern Strength, Peak Pitch Class, Pattern Chance, Heard Going Down, Going Down by Pitch Class, Check Pairs Right, High on the Right, High on the Left, Something Else, Controls Right, Median Answer Time, Anticipations, Completed Trials, Planned Trials, Headphone Check Passed, Headphone Check Correct and Reported Output Delay. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Pitch Illusions Task used in research?
51 papers from 2021 to 2025 name it in their title or abstract, up 42% compared with 2016 to 2020. That ranks it 304 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 Pitch Illusions 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?
When the headphone check fails (Stop the test, Continue the test) and rounds of half-octave pairs (1 to 3). 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.