Attention
Tone-in-Noise Task
Also called: tone in noise detection, tone detection in noise, masked tone detection, two-interval tone detection, signal detection in noise
Measure how well people pick out a quiet tone in noise: two bursts of noise through headphones, the one with a 1 kHz tone chosen, at eight fixed levels.
6-9 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
- 110
- Evidence grade
- C
- Registered clinical studies
- 1
What it measures
Measures how well people pick out a quiet tone in noise. Participants hear two bursts of noise and choose the one that carried a 1 kHz tone, at eight tone levels from 36 to 15 dB below the noise, fixed before the session. The main score is the share of right answers; the share at each level and the level at which 75 percent are right come with it. A research measure, not a hearing test.
Constructs
- Auditory detection in noise
- Signal detection
- Auditory masking
Research uses
- Auditory perception research
- Listening and attention research
- Within-study group comparison
How the paradigm works
The tone-in-noise task measures how well people pick out a steady tone in noise, the two-interval detection task of signal detection theory (Green and Swets, 1966; Green, 1960). It needs headphones and normal hearing. 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.
More about the paradigm
On each trial the participant presses Play and hears two bursts of white noise, 400 ms each and 500 ms apart. One of them, the first or the second on half the trials at every level, also carries a 1 kHz tone of 300 ms at its center, the pitch of the volume-setting tone; the other is noise alone. The participant chooses the burst with the tone, with no time limit. The tone comes at eight levels, from 36 to 15 dB below the noise in 3 dB steps (the ratio of the tone's rms to the noise's), each 12 times (8 to 20 as a setting), in runs that hold every level once, in an order drawn from the session's seed: 96 trials at the standard settings. Six practice trials at the two loudest tones give feedback; the test gives none. Every level is set before the session and no answer changes the levels that follow (the method of constant stimuli), so there is no staircase. The browser makes the noise 10 dB below the volume-setting tone and starts every sound on the audio clock's sample grid; the output delay the device reports is saved.
The main score is the share of right answers over the eight levels, with the share at each level beside it. From those shares the test reads the tone's level at which 75 percent of answers are right, from a two-choice Weibull function of the tone's amplitude fitted by maximum likelihood. A tone in white noise is masked mostly by the noise near its own frequency (Fletcher, 1940; Hawkins and Stevens, 1950), so the tone's level is set relative to the noise. Levels are relative to the noise and to a volume the participant chose, never sound pressure levels, and headphones differ, so compare sessions run the same way.
Scores
- Accuracy%
- Share of usable answers that named the burst with the tone, over the eight levels.
- Higher means more tones were picked out; guessing gives 0.5
- Accuracy by Level%
- Share right at each tone level, keyed by the tone's level re the noise in dB.
- Rises with the tone's level
- Tone Level at 75 PercentdB
- The tone's level re the noise, in dB, at which the fitted Weibull function reaches 75 percent right.
- Lower means a quieter tone was picked out; not a hearing level
- Psychometric Slopescore
- Slope of the fitted Weibull function of the tone's amplitude.
- Higher means accuracy rose more steeply with the tone's level
- Lapse Rate%
- Fitted share of errors on tones loud enough to pick out.
- Lower means fewer lapses
- Accuracy, Tone in the First Burst%
- Share right when the tone was in the first burst.
- Informational
- Accuracy, Tone in the Second Burst%
- Share right when the tone was in the second burst.
- Informational
- First Burst Answers%
- Share of usable answers that named the first burst.
- 0.5 means no lean toward either burst
- Median Answer Timems
- Median time from when the answers opened, after both bursts, 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.
- 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
110 papers from 2021 to 2025 name this task in their title or abstract, up 11% compared with 2016 to 2020. That ranks it 204 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 28 |
| 2024 | 23 |
| 2023 | 22 |
| 2022 | 19 |
| 2021 | 18 |
| 2020 | 18 |
| 2019 | 16 |
| 2018 | 21 |
| 2017 | 21 |
| 2016 | 23 |
| 2015 | 19 |
| 2014 | 25 |
| 2013 | 24 |
| 2012 | 19 |
| 2011 | 15 |
| 2010 | 11 |
| 2009 | 14 |
| 2008 | 13 |
| 2007 | 18 |
| 2006 | 20 |
| 2005 | 13 |
| 2004 | 11 |
| 2003 | 11 |
| 2002 | 8 |
| 2001 | 12 |
| 2000 | 4 |
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
- 11
- Papers on reliability or validity
- 79
- Registered clinical studies using it as an outcome
- 1
Note on the counts: includes hearing science 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 0.22.0, 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.
- Trial
- Press Play to hear two bursts of white noise, 400 ms each and 500 ms apart. One also carries a 1 kHz tone of 300 ms at its center, with 20 ms ramps; the other is noise alone. Choose First or Second for the one with the tone, with no time limit.
- Levels
- Eight tone levels from 36 to 15 dB below the noise in 3 dB steps: -36, -33, -30, -27, -24, -21, -18 and -15 dB, as the ratio of the tone's rms to the noise's.
- Trials
- Each level 12 times (standard, 8 to 20), in runs that hold every level once, in an order drawn from the session's seed, with the tone in each burst on half the trials at every level: 96 trials, with no feedback.
- Practice
- 6 trials at the two loudest tones, with feedback for 1,000 ms.
- Fixed levels
- Every level is set before the session, and no answer changes the next one: there is no staircase. The browser makes the noise 10 dB below the volume-setting tone and starts every sound on the audio clock's sample grid; the output delay the device reports is saved.
- Score
- The share of right answers over the eight levels, and the share at each level. The tone's level at which 75 percent of answers are right comes from the random dot motion task's two-choice Weibull of the tone's amplitude, fitted by maximum likelihood. Answers under 100 ms leave the score.
Levels are relative to the noise and to the volume the participant chose, not sound pressure levels, and headphones differ, so compare results within studies run the same way. The test needs normal hearing and headphones, wired if possible. It is a research measure, not a hearing test: it gives no hearing level and no pass or fail, and its scores are not a sign of hearing loss. 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: tone_in_noise_test from the scored run, a button for the first or the second burst of noise after both have played, with no time limit (chance 50%; the tone is in each burst on half the trials at every level), graded from the saved button and the burst that carried the tone. Below chance invalidates and not above chance is flagged; one button on 90% of answers invalidates when accuracy is 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, no 75 percent level within the levels played, 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 both sounds have 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. A failed headphone check that stopped the test leaves no trials to score (Woods, Siegel, Traer and McDermott, 2017). A level outside the levels played cannot be estimated from them.
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 |
| Trials per level | 12 | 8 to 20 |
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 Tone-in-Noise Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Green and Swets, 1966; Hawkins and Stevens, 1950), with the standard settings (when the headphone check fails: Stop the test; trials per level: 12). The main score was accuracy (share of usable answers that named the burst with the tone, over the eight levels).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Green, D. M., & Swets, J. A. (1966). Signal detection theory and psychophysics. Wiley. Two-interval detection of a signal in noise, with the share right as the measure of detectability.
- Green, D. M. (1960). Psychoacoustics and detection theory. The Journal of the Acoustical Society of America, 32(10), 1189-1203. https://doi.org/10.1121/1.1907882Detection theory applied to hearing a signal in noise: a measure of detectability apart from the listener's criterion.
- Fletcher, H. (1940). Auditory patterns. Reviews of Modern Physics, 12(1), 47-65. https://doi.org/10.1103/RevModPhys.12.47A tone in noise is masked mostly by the noise near its own frequency.
- Hawkins, J. E., & Stevens, S. S. (1950). The masking of pure tones and of speech by white noise. The Journal of the Acoustical Society of America, 22(1), 6-13. https://doi.org/10.1121/1.1906581The masking of pure tones by white noise, as the ratio of a tone's level to the noise's level per cycle.
- 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
- Green and Swets (1966): two-interval detection of a tone in noise, with the share right as the measure of detectability. Hawkins and Stevens (1950): the masking of pure tones by white noise.Paradigm source, Adults
Questions
How long does the Tone-in-Noise Task take?
About 6 to 9 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?
Accuracy, Accuracy by Level, Tone Level at 75 Percent, Psychometric Slope, Lapse Rate, Accuracy, Tone in the First Burst, Accuracy, Tone in the Second Burst, First Burst Answers, 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 Tone-in-Noise Task used in research?
110 papers from 2021 to 2025 name it in their title or abstract, up 11% compared with 2016 to 2020. That ranks it 204 of 495 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Tone-in-Noise Task have?
In the published research, 11 papers report its test-retest reliability and 1 registered clinical study uses 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?
When the headphone check fails (Stop the test, Continue the test) and trials per level (8 to 20). 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.