GB2562599A - Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment - Google Patents
Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment Download PDFInfo
- Publication number
- GB2562599A GB2562599A GB1804604.5A GB201804604A GB2562599A GB 2562599 A GB2562599 A GB 2562599A GB 201804604 A GB201804604 A GB 201804604A GB 2562599 A GB2562599 A GB 2562599A
- Authority
- GB
- United Kingdom
- Prior art keywords
- user
- stimulus
- test
- parameter
- response curve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/12—Audiometering
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/486—Bio-feedback
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Heart & Thoracic Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Veterinary Medicine (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Psychiatry (AREA)
- Hospice & Palliative Care (AREA)
- Data Mining & Analysis (AREA)
- Child & Adolescent Psychology (AREA)
- Psychology (AREA)
- Educational Technology (AREA)
- Databases & Information Systems (AREA)
- Biodiversity & Conservation Biology (AREA)
- Developmental Disabilities (AREA)
- Social Psychology (AREA)
- Acoustics & Sound (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Using a method of continuous adjustment to obtain a response curve indicative of the user's perception of the variations in an adaptive parameter stimulus, the methods being directed to maximize regularity of user interactions around the response curve. The methods find particular application in audiometric tests, including when carried out using unreferenced or non-calibrated audio-systems, in embodiments this is psychometric tuning curve testing, but also have more general application and may also be used for similar test applications in other fields. An adaptive parameter is continuously adjusted and the rate of change of the adaptive parameter is adjusted depending on the regularity of user interactions. The adjustments are directed towards neutral user interaction around the response curve and providing more uniform intervals between successive user interactions. The stimulus may be an audio stimulus with the adaptive parameter being amplitude and the second variable parameter, frequency. Further embodiments may utilize a masking signal with a variable frequency sweep.
Description
(71) Applicant(s):
Mimi Hearing Technologies GmbH
Neue Schonhauser StraBe 19, Berlin 10178, Germany (72) Inventor(s):
Nicholas R Clark
Vinzenz H Schonfelder
Tina Baumann (74) Agent and/or Address for Service:
Stratagem Intellectual Property Management Limited Meridian Court, Comberton Road, Toft, Cambridge, CB23 2RY, United Kingdom (58) Field of Search:
INT CL A61B, A61N, G06F, G16H
Other: EPODOC, WPI, INSPEC, Patent Fulltext, MEDLINE, XPI3E, ΧΡΙΕΕ (54) Title of the Invention: Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment Abstract Title: Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment (57) Using a method of continuous adjustment to obtain a response curve indicative of the user's perception of the variations in an adaptive parameter stimulus, the methods being directed to maximize regularity of user interactions around the response curve. The methods find particular application in audiometric tests, including when carried out using unreferenced or non-calibrated audio-systems, in embodiments this is psychometric tuning curve testing, but also have more general application and may also be used for similar test applications in other fields. An adaptive parameter is continuously adjusted and the rate of change of the adaptive parameter is adjusted depending on the regularity of user interactions. The adjustments are directed towards neutral user interaction around the response curve and providing more uniform intervals between successive user interactions. The stimulus may be an audio stimulus with the adaptive parameter being amplitude and the second variable parameter, frequency. Further embodiments may utilize a masking signal with a variable frequency sweep.
Fig. 5
1/3
Fig. 3
2/3
Fig. 6
3/3
Fig. 7
Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment
The invention relates to methods employing user-dependent conditioning of stimuli in tests to elicit user responses to variations of an adaptive parameter of a stimulus using a method of continuous adjustment to obtain a response curve indicative of the user’s perception of the variations in the adaptive parameter of the stimulus, the methods being directed to maximize regularity of user interactions around the response curve. The methods find particular application in audiometric tests, including when carried out using unreferenced or non-calibrated audio-systems, and this description focuses on such applications, but the methods have more general application and may also be used for similar test applications in other fields of sensory behavioural science where a user interaction is measured in response to stimulus adaptations. The methods may conveniently be implemented by means of a computer program or mobile device application (“App”) allowing use of personal mobile devices as non-calibrated audio systems for users to perform selfadministered tests.
The methods of the present invention address the problems of improving the efficiency and utility of tests, for example hearing tests, performed under uncertain conditions. The methods particularly involve making predictions of an expected test response curve, contour or profile for a user, based on prior data and knowledge about the user. The prior data may comprise information on the user’s responses in previous tests, and/or comparative information on test results from a database of other users with similar socio-demographic profiles and pathology.
The methods particularly involve continuous real-time adjustments of one or more test parameters during a test, to steer user responses towards the predicted or expected response contour of the user, to maximise user response symmetry around an adaptive parameter of a stimulus.
Various behavioural methods have been developed in psychophysics to obtain psychometric data from observers, e.g., to measure a person's hearing ability. Typical examples are the method of limits, method of constant stimuli, method of adjustment and forced choice methods.
In the context of measuring an observer's hearing threshold, Bekesy developed a method of continuous adjustment called Bekesy tracking [ Bekesy, G. v., A new audiometer, Acta oto-laryngol., 35, 411-422 (1947)]. By way of a simple binary interaction of the user (pressing or releasing a single button), a parameter, i.e. the amplitude, of a sound stimulus is constantly increased or decreased resulting in an oscillation around a threshold. The threshold level can then be estimated from the points of user interaction occurring above and below the threshold.
A sweeping Bekesy tracking paradigm represents a variant of that general method, where a second parameter of the stimulus (e.g. frequency) is constantly changed so that the level of the perceptual threshold is traced along a range of values of that parameter of the stimulus.
Originally developed for estimating pure tone auditory thresholds, the general mechanics of the Bekesy method, i.e. the continuous adjustment of parameters of a stimulus based on user interaction, have also been applied in other contexts, e.g. for estimating psychophysical tuning curves (PTC). [ S$k, A., Alcantara, J., Moore, B. C. J., Kluk, K., & Wicher, A. (2005). Development of a fast method for determining psychophysical tuning curves. International Journal of Audiology, 44 (7), 408-420.] Bekesy audiometry has been recognised as a useful diagnostic tool in clinical audiology [see e.g. Granitz, D.W. “An evaluation of diagnostic parameters of Bekesy audiometry”, LSU Historical Dissertation Theses 2052 (1971)].
In a Bekesy tracking/continuous adjustment paradigm, for example, a user is tasked with pressing a button when he hears a sound, and releasing when he doesn't. As long as the button is pressed, the parameter, i.e. the amplitude, of the stimulus is continuously reduced at a fixed rate until the user releases the button. When the button is released the parameter, i.e. the amplitude, of the stimulus is increased at the same rate. As a result of this procedure, the parameter of the stimulus should continuously oscillate around the threshold level of a user at a given frequency.
Owing to its intuitive and engaging character, this method of continuous usercontrolled adjustment of the parameter of the stimulus lends itself particularly well for (but is not limited to) consumer (e.g. mobile device) implementations of psychometric tests, such as audiometric hearing tests. Users can quickly learn the task, they are not required to directly look at the device during a test and their continuous engagement means that a large body of data can be collected over a relatively short period.
The conventional implementation of such a continuous adjustment paradigm uses equal upward and downward rates of change of the user-controlled parameter of a stimulus. Consequently, a neutral user interaction, i.e. pressing and releasing regularly and at uniform intervals, results in a flat (constant level) threshold estimate.
Generally, accuracy and reliability of the data in such a continuous adjustment paradigm is dependent on regularly occurring user interaction, i.e. a threshold can only be reliably estimated in a parameter region where the user gives regular feedback.
If the actual threshold curve deviates significantly from the threshold levels that would be determined by neutral user interaction, users may only interact very rarely. In addition, data may exhibit an unpredictable bias when users become confused after not interacting for longer periods.
The stimulus signal levels that a user can follow in a sweeping continuous adjustment paradigm where a second parameter is continuously varied as the user’s response to the stimulus signal levels is measured, may be technically limited by the fixed rates of change of the stimulus level and the rate of change of the second, sweeping stimulus parameter. In particular, if the slope of the actual threshold curve is similar to, or steeper than, this technical limit, the measured trace may not reflect the true thresholds beyond this limit.
If a user can be steered to provide a neutral response trace following a shape that is closer to the actual threshold curve, the user will give more regular feedback, leading to more reliable data, and less bias in the data. Under some conditions, this will also allow for steeper threshold changes to be followed.
According to the present invention there is provided a method including a modification to the standard continuous adjustment paradigm, to facilitate neutral response behaviour from the user, and include the steps of:
estimating an initial expected response curve for a test of at least one parameter of a stimulus for a particular user, based on prior data and knowledge about that user;
adjusting the rate of change of the or each parameter of the stimulus in response to the regularity of user interactions during the test in order to steer the user responses towards the expected response curve to obtain more uniform intervals between successive user interactions (“neutral” user interaction);
modifying the initial expected response curve in accordance with the test results when substantially neutral user interactions are not obtained across the test range for a selected parameter of the stimulus.
Neutral user interaction can be achieved by employing dissimilar rates of change of a stimulus parameter when increasing and decreasing the parameter level to which the user responds during the test, by modifying the rate of change of this parameter depending on whether it is increasing or decreasing.
In a sweeping auditory test, for example, this may mean adjusting separately for upward and downward sections, i.e. having faster rates of change in stimulus levels during level increases and slower rates of change during level decreases, to steer the user response curve towards the expected response curve for the test parameter. As a result, any non-neutral behaviours of the user (test interaction at non-uniform intervals) correspond to deviations of the actual trace from the one predicted initially, and the predicted response contour can then be revised accordingly.
Importantly, the methods of the invention thereby allow a continuous analysis of how well the user responses match the initially expected response contour. By comparing the actual and predicted responses, it is possible to continuously adjust the expected response curve to maximise the probability of obtaining an optimum data set for the user. This may be achieved by a real-time windowed regression function calculation or LMS (least mean square) analysis of the user’s responses against a dictionary, or preferably a parametric model of actual responses from previous tests (of the same or other users), and dynamically adapting the expected response contour during the test.
The method may include the further step of modifying the expected response curve in accordance with detected test results when substantially neutral user interaction is not obtained across the range of variation of the second variable parameter. The second variable parameter of the stimulus may be continuously varied monotonically across the range of values for the test. The actual and expected user responses and the degree of neutral user interaction may be continuously analysed and monitored, and the expected response curve may therefore be continually adjusted to maximise the probability of obtaining neutral user interaction across the range of values for the test.
The rate of change of increases in the adaptive parameter of the stimulus to meet the specified condition may be increased when expected response curve predicts that the specified condition will not be met and neutral interaction will not be obtained without such increase.
Alternatively or additionally, the rate of change of decreases in the adaptive parameter of the stimulus after the specified condition has been met may be decreased when expected response curve predicts that the specified condition will not cease to be met and neutral interaction will not be obtained without such decrease.
The method according to the present invention may be used for an audiometric test, wherein the test comprises, or is, a supra-threshold test, a psychometric tuning curve test, a temporal fine structure test, or a temporal masking curve test. In particular, the supra-threshold test may be a psychometric tuning curve test. The psychometric tuning curve test may be measured for signal tones between frequencies of 500 Hz and 4 kHz, and at a sound level of between 20 dB SL and 40 dB SL, in the presence of a masking signal for the signal tone that sweeps from 50% of the signal tone frequency to 150% of the signal tone frequency.
The adaptive parameter of the stimulus may be an audio signal amplitude. The second variable parameter of the stimulus may be an audio tone frequency.
The rate of change of the second variable parameter of the stimulus may also be varied to steer the user responses towards neutral user interaction around the expected response curve with more uniform intervals between successive user interactions.
The method according to the present invention may be implemented under control of a software application for use on a non-calibrated or unreferenced audio system, and adapted for the user to self-administer the test.
Furthermore, according to the present invention there is provided a method for userdependent conditioning of stimuli in a test to elicit user responses to variations of an adaptive parameter of a stimulus utilising a method of continuous adjustment to obtain a response curve indicative of the user’s perception of the variations in the adaptive parameter of the stimulus over a range of values of a second variable parameter of the stimulus, the user responses comprising binary indications of whether or not the varying adaptive parameter of the stimulus meets a specified condition as the second variable parameter of the stimulus is also varied, each change of binary indication being identified as a user interaction, the method being directed to obtain at least a predetermined number of user interactions around the response curve, the method comprising the steps of:
- estimating the expected response curve for the user based on prior data and knowledge about the test and the user;
- adjusting the rate of change of the adaptive parameter of the stimulus depending on the regularity of user interactions during the test, the adjustments being directed to obtain at least a predetermined number of user interactions over the range of values of the second variable parameter of the stimulus.
Methods according to the invention will be explained and described in further detail and with reference to the accompanying Figures, in which:
Figure 1 illustrates user interactions in response to a Bekesy tracking method of continuous adjustment of a stimulus level at steady rates up and down over time;
Figure 2 illustrates change of a second stimulus parameter over time, consistent with a “sweeping” Bekesy tracking paradigm;
Figure 3 illustrates “neutral” user interactions in response to changes in stimulus levels and frequency over a test frequency range;
Figure 4 illustrates the problem whereby variations in a user’s sensitivity to changes in the adaptive parameters under a conventional sweeping Bekesy paradigm lead to a loss of interaction;
Figure 5 illustrates how a modified continuous adjustment method according to the invention results in extended user interaction;
Figure 6 illustrates differing rates of change for stimulus level vs. frequency and frequency vs. time for a sweeping auditory threshold test; and
Figure 7 illustrates Psychometric Tuning Curve (PTC) test results for changes in stimulus level and frequency with adaptive variation of a masker signal.
In Figure 1, user responses 10, 11 are shown for changes in stimulus level A of a single adaptive parameter (e.g. audio signal amplitude at a specific pure tone frequency) over time T. The user responses comprise pressing 10 and holding a button pressed when the tone reaches an audible level and releasing 11 the button when the tone is no longer audible. Between released and pressed button states, the tone amplitude is varied at constant rates in accordance with a conventional Bekesy tracking paradigm. Thus, when the button is released, the stimulus level is increased at a constant rate 12, and when the button is pressed, the stimulus level is decreased at a constant rate 13. The user responses 10, 11 thereby oscillate consistently around a threshold level 14.
Figure 2 illustrates the change of a second stimulus parameter over time, in a sweeping Bekesy paradigm, where a second independent stimulus parameter (in this case frequency F) is continuously varied over time T. The user then responds to changes in the first stimulus level as the second stimulus level is also changed.
Figure 3 illustrates an example of “neutral” user interaction, with user responses of pressing 10, and releasing 11, the button at regular intervals. In this case, the first stimulus level (i.e. audio signal amplitude A) is varied up and down as the second stimulus parameter (i.e. audio tone frequency F) thus continuously changing from lower to higher levels. The threshold level is thereby measured across a range of values of that second parameter, following a sweeping Bekesy tracking paradigm. As in the standard Bekesy tracking paradigm, the stimulus level of the first parameter is increased at a constant rate 32 when the button is released and decreased at a constant rate 33 when the button is pressed, as the second parameter (i.e. frequency
F) is monotonically increasing. This results in a threshold curve 34 over a range of frequencies (rather than the single threshold level for a single frequency tone as shown in Figure 1). The user’s interaction is “neutral” in this example, because the threshold curve is in fact at a constant level across this limited frequency range. This will not be so across the whole of an individual’s normal hearing range.
Figure 4 illustrates the situation which arises when a user’s response curve to changes in the first parameter of a stimulus is not constant with respect to changes in the second parameter of the stimulus - e.g. when a user’s audio threshold levels vary significantly at higher or lower frequencies. Thus, when the threshold curve 44 deviates significantly from a substantially constant level, the user interactions of pressing 10, and releasing 11, the button become less regular and less frequent. Eventually, after a final button release 45, the user no longer interacts. In these circumstances, whilst response data allows reliable validation of the expected response curve 44a below a cut-off frequency 46, no data is obtainable to validate the response curve 44b above the cut-off 46. The reason for this lack of data lies in the fixed rates of increase 42 and decrease 43 of the stimulus level. Beyond the last user interaction 45, the fixed rate of change of increase 42 in the stimulus level with change in the frequency is actually less than the increase in the user’s actual audio threshold level above the cut-off frequency 46. Thus it is no longer possible to get any further user interaction using the standard Bekesy paradigm.
Figure 5 illustrates how it is possible to extend the range of user interaction to obtain appropriate data by using a modified Bekesy paradigm according to the invention. In this example, the rates of change of increases 52a, 52b, 52c in stimulus level and of decreases 53a, 53b, 53c in stimulus level are not necessarily kept constant (as in the standard Bekesy paradigm of Figure 4) but are themselves adaptive and variable according to the predicted, or detected, variations in rate of change of a user’s threshold response curve 54 for the measured parameter. Thus, where the user’s threshold response level is only slowly increasing, the rates of change 52a, 53a up or down of the stimulus level are still relatively constant, as in the standard Bekesy paradigm. However, where the user’s threshold levels are predicted, or detected, to be increasing more rapidly (e.g. around the cut-off frequency 46 of Figure 4), the respective rates of change of increasing and decreasing stimulus levels diverge. Rates of change for increasing levels 52b, 52c (after button releases) are steeper, and rates of change for decreasing levels 53b, 53c are flatter or slower, and steered towards the expected response curve 54. Through this modification, it is possible to generate more regular and consistent data. In this particular example, when the actual threshold trace follows the expected one, the user is able to continue to interact and generate regular data points beyond the previous final data point 45, thereby allowing validation of the response curve 54 beyond the previous cut-off 46. Figure 6 illustrates how potential adaptive parameters (in this example, audio stimulus level 61 and audio tone frequency 62) may be varied at different rates. In the example of Figure 5, the rates of change of audio stimulus level were varied when increasing or decreasing the level. Alternatively, it would be possible to vary the rates of change of frequency for specific stimulus levels, or vary rates of change for both parameters.
Note also that, if the measured levels of an adaptive parameter are expected to drop off steeply with changes in a second parameter, then rather than increasing the rate of change of the parameter upwards and decreasing the rate of change downwards (c.f. Figure 5), it may be necessary to follow the opposite procedure - i.e. to increase the rate of change downwards and decrease the rate of change upwards.
It is also possible to adjust the rates of variation or sweep of any of the parameters during tests to obtain data meeting a desired quality criterion. Such a criterion could be a minimum total number of interactions (i.e. data points). Thus, for example, a test may be performed more rapidly for an individual who is fully able, but may need to be carried out more slowly for users who have some form of physical or cognitive impairment to meet a desired quality criterion. For example, the second parameter value may be changed (or swept) more slowly or more rapidly, depending on the rate and regularity of user interactions, to obtain at least a predetermined number (e.g. a minimum total number necessary to fully validate a response curve) of user interactions over the range of values of the second variable parameter of the stimulus.
Various audiometric tests may benefit from using modified Bekesy tracking in accordance with the methods of the invention. These include Psychometric Tuning Curve (PTC) tests.
A PTC test may typically be performed for signal tones between frequencies of 125 Hz and 16 kHz, preferably between frequencies of 250 Hz and 8 kHz, and even more preferably between frequencies of 500 Hz and 4 kHz, and may be performed at any level above the user’s hearing threshold (i.e. any level above 0 dB SL), preferably at signal levels between 10 dB SL and 60 dB SL, and even more preferably at signal levels between 20 dB SL and 40 dB SL, with a masking signal applied to sweep in a predefined range around each signal tone frequency, particularly from 50% of the signal tone frequency to 150% of the signal tone frequency. In an alternative embodiment of the present invention, the masking signal is applied to sweep around each signal tone frequency in a range of 60% of the signal tone frequency to 140% of the signal tone frequency, particularly between 80% of the signal tone frequency to
120% of the signal tone frequency.
The signal level of the masking signal is continually modulated according to a user’s responses to the tone. Particularly if the user responds that they can detect the signal tone, the masker signal level is increased, and if the user indicates they cannot detect the signal tone, the masker signal level is decreased.
In a calibrated system, more reliable data can be obtained by modulating the masker intensity (i.e. the signal level of the masking signal) around a curve that would provide a constant output in dB HL. This is so that the user does not experience any jumps in intensity due to discontinuities in the frequency response of the hardware setup, or due to differences in human sensitivity to tones of different frequency.
In an uncalibrated audio system however (assuming the output of the audio system is at least reasonably flat across frequency), and particularly given that the precise output level at each frequency is unknown in an un-calibrated system as explained above, International Patent Application PCT/EP2017/076679 filed October 2017 discloses that it is advantageous if the masker intensity is modulated relative to a standard weighting curve, such as for example A-weighting, such as defined in IEC 61672:2003, or an equal loudness contour, such as defined in ISO226. This provides a consistent feel of control of the masker intensity to the user across the frequency range of the test and yields more reliable results. It will be readily apparent that modified Bekesy tracking in accordance with the methods of the invention may also enhance such test results.
The invention may also be applied to other audiometric tests. Examples include supra-threshold tests, temporal fine structure tests, and temporal masking curve tests. Such tests are generally known in the art.
A temporal masking curve test involves a forward-masking task whereby the masker level required to mask a fixed, low-level pure-tone probe is measured as a function of the masker-probe interval to produce a temporal masking curve. Limiting the probe to a low level minimizes spread of excitation along the basilar membrane and effects of off-frequency listening. As the probe level is fixed, the required masker level increases with increasing masker-probe interval, resulting in temporal masking curves that have positive slopes. For an off-frequency masker, which is assumed to be processed linearly, the temporal masking curve is assumed to reflect decay of masking. As the masker-probe interval is increased, the masker level required at masked threshold increases to compensate for the time course of decay. For an onfrequency masker, the temporal masking curve is assumed to reflect the combined effects of the decay of masking and compression applied to the masker. Therefore, a larger change in masker level would be required to produce a given change in basilar membrane excitation when the response to the masker is compressive. This would be reflected as a steeper on-frequency temporal masking curve compared to an offfrequency temporal masking curve. Assuming the time course of decay of the masker is identical for all masker frequencies (and levels), the degree of basilar membrane compression can be estimated by comparing the slope of a temporal masking curve for an on-frequency masker against the slope of a temporal masking curve for a masker that is processed linearly by the basilar membrane (i.e. off-frequency masker or linear-reference temporal masking curve). Basilar membrane responses can be inferred by plotting the off-frequency masker level against the on-frequency masker level required for each masker-probe interval.
Where the invention is applied to a temporal masking curve test, the rate of change of the masker level and the masker-probe interval could be varied as the adaptive parameters of the stimulus.
Although the description with reference to Figure 6, above, is exemplified using potential adaptive parameters of audio stimulus level 61 and audio tone frequency 62, it will be understood that the test is applicable to any psychophysical test in which a threshold in sensory perception of a first factor is determined while second and third factors are changing over time.
For example, visual tests such as contrast sensitivity tests would be equally applicable. In a test of this nature, the ability to distinguish two lines as distinct objects with varying colour and/or intensity can be tested. In this case the adaptive parameters of the stimulus could be the hue (i.e. the frequency) of the line and the distance between the lines. As another option this test could use, as adaptive parameters, the intensity (typically measured in lumens) of the lines and the distance between the lines.
Alternatively, or additionally, the ability to read text of varying colour and intensity against a specific background. In this case the adaptive parameters of the stimulus could be the text hue and the background hue. As another option this test could use, as adaptive parameters, the text intensity and the background intensity In a similar manner to that described above with reference to the auditory example of Figure 6, the user response can be steered towards the predicted or expected response contour of the user. The user presses a button when they see the distinct objects and release the button when they can’t as colour and intensity are varied over time.
Further examples may include, but are not limited to visual acuity, where the object size and focus length are variables; peripheral vision, where the object size and angle from central vision are variables; colour vision, where the colour and intensity of two light sources are varied to match a third source; and taste or smell where the presence of a first smell or taste is tested using the variables of concentration and location.
Figure 7 shows an illustration of a PTC test measurement 100. A signal tone 102 at a first sound level 101 is masked by a masker signal 105 particularly sweeping 103 through different frequencies in the proximity of the signal tone 102. The test user indicates at which sound level he hears the signal tone over the masker signal. The signal tone and the masker signal are well within the user’s hearing range.
The diagram shows the audio level or intensity A in arbitrary units on the y-axis against frequency F on the x-axis.
While a signal tone 102 that of a constant frequency and intensity 101 is played to the user a masker signal 105 slowly sweeps 103 from a frequency lower to a frequency higher than the signal tone 102. The rate of sweeping 103 may be constant or may be varied in response to the consistency of the user’s interactions. The goal for the user is to hear the signal tone 102. When the user can no longer hear the signal tone 102 (e.g. as indicated by the user by releasing a pushbutton) the masker signal intensity is reduced 104 to a point where the user starts hearing the signal tone 102 (indicated by the user by pressing the pushbutton). While the masker signal tone 105 is still sweeping 103 upwards in frequency, the intensity of the masker signal 105 is increased 104 again, until the user no longer hears the signal tone 102 again. This way, the masker signal intensity oscillates 106 around the user’s expected or predicted hearing level response curve107 (as indicated by the solid line).
This hearing level response curve 107 can be compared with a well established and well known response curve for people having no hearing loss. Any deviations from this known curve would be indicative of a hearing loss.
In order to improve the efficiency and effectiveness of the methods of the invention, it is desirable that the initial expected user response curves or profiles for changes in one parameter over a range of variation of a second parameter should be predicted with sufficient accuracy to facilitate substantially neutral user interaction across a desired test range.
For audiometric purposes, or example, for threshold curves in Pure Tone Threshold tests, supra-threshold tests (e.g. Psychometric Tuning Curve tests), or equal loudness contour tests, such predictions may be based on information of various types, such as:
ISO7029: Acoustics - Statistical distribution of hearing thresholds as a function of age;
ISO226: Acoustics - Normal equal loudness contours;
prior test results (and particularly confirmed clinical-standard pure tone audiometry) from cohorts of other patients with similar characteristics;
other available meta-information about the data sets, including:
o Reaction times o Estimated accuracy of data set o Time of day o Geographic location o Test duration o Number of interruptions o Demographic user information
- Socio-demographic factors, such as o age of the user o sex of the user o the cognitive capacity of the user o genetic disposition of the user, possibly affecting hearing ability o genomic information for the user o working environment o leisure activities o music listening habits (types, loudness, frequency) o telephone setting preferences (loudness) o chronic or acute illnesses, pathologies affecting hearing ability o prescription/recreational drug use (including alcohol) performance details for the user’s audio equipment (e.g. mobile audio device and headphones);
A computer program, or mobile device application (“App”) comprising computer program code, may be provided to allow a user to self-administer tests using the methods according the invention, when the computer program is loaded, streamed or executed on a personal computer or mobile device.
It will further be appreciated by those skilled in the art that although the invention has been described by way of example with reference to several embodiments it is not limited to the disclosed embodiments and that alternative embodiments could be constructed without departing from the scope of the invention as defined in the appended claims.
Claims (13)
1. Method for user-dependent conditioning of stimuli in a test to elicit user responses to variations of an adaptive parameter of a stimulus utilising a method of continuous adjustment to obtain a response curve indicative of the user’s perception of the variations in the adaptive parameter of the stimulus over a range of values of a second variable parameter of the stimulus, the user responses comprising binary indications of whether or not the varying adaptive parameter of the stimulus meets a specified condition as the second variable parameter of the stimulus is also varied, each change of binary indication being identified as a user interaction, the method being directed to maximize regularity of user interactions around the response curve, the method comprising the steps of:
- estimating the expected response curve for the user based on prior data and knowledge about the test and the user;
- adjusting the rate of change of the adaptive parameter of the stimulus depending on the regularity of user interactions during the test, the adjustments being directed to steer the user responses towards neutral user interaction around the expected response curve with more uniform intervals between successive user interactions.
2. Method according to claim 1 including the further step of modifying the expected response curve in accordance with detected test results when substantially neutral user interaction is not obtained across the range of variation of the second variable parameter.
3. Method according to claim 1 wherein the second variable parameter of the stimulus is continuously varied monotonically across the range of values for the test.
4. Method according to any preceding claim wherein the rate of change of increases in the adaptive parameter of the stimulus to meet the specified condition is increased when expected response curve predicts that the specified condition will not be met and neutral interaction will not be obtained without such increase.
5. Method according to any preceding claim wherein the rate of change of decreases in the adaptive parameter of the stimulus after the specified condition has been met is decreased when expected response curve predicts that the specified condition will not cease to be met and neutral interaction will not be obtained without such decrease.
6. Method according to claim 2 wherein the actual and expected user responses and the degree of neutral user interaction are continuously analysed and monitored, and the expected response curve is continually adjusted to maximise the probability of obtaining neutral user interaction across the range of values for the test.
7. Method according to any preceding claim used for an audiometric test, wherein the test comprises, or is, a supra-threshold test, a psychometric tuning curve test, a temporal fine structure test, or a temporal masking curve test.
8. Method according to claim 7, wherein the supra-threshold test is a psychometric tuning curve test, the psychometric tuning curve test is measured for signal tones between frequencies of 500 Hz and 4 kHz, and at a sound level of between 20 dB SL and 40 dB SL, and wherein a masking signal for the signal tone sweeps from 50% of the signal tone frequency to 150% of the signal tone frequency.
9. Method according to any preceding claim, wherein the adaptive parameter of the stimulus is an audio signal amplitude.
10. Method according to any preceding claim, wherein the second variable parameter of the stimulus is an audio tone frequency.
11. Method according to any preceding claim, wherein the rate of change of the second variable parameter of the stimulus is also varied to steer the user responses towards neutral user interaction around the expected response curve with more uniform intervals between successive user interactions.
12. Method according to any preceding claim implemented under control of a software application for use on a non-calibrated or unreferenced audio system, and adapted for the user to self-administer the test.
13. Method for user-dependent conditioning of stimuli in a test to elicit user responses to variations of an adaptive parameter of a stimulus utilising a method of continuous adjustment to obtain a response curve indicative of the user’s perception of the variations in the adaptive parameter of the stimulus over a range of values of a second variable parameter of the stimulus, the user responses comprising binary indications of whether or not the varying adaptive parameter of the stimulus meets a specified condition as the second variable parameter of the stimulus is also varied, each change of binary indication being identified as a user interaction, the method being directed to obtain at least a predetermined number of user interactions around the response curve, the method comprising the steps of:
- estimating the expected response curve for the user based on prior data and knowledge about the test and the user;
- adjusting the rate of change of the adaptive parameter of the stimulus
5 depending on the regularity of user interactions during the test, the adjustments being directed to obtain at least a predetermined number of user interactions over the range of values of the second variable parameter of the stimulus.
Intellectual
Property Office
Application No: GB1804604.5
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/355,449 US11412958B2 (en) | 2017-03-22 | 2019-03-15 | Systems and methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment |
US17/328,990 US20210275060A1 (en) | 2017-03-22 | 2021-05-24 | Systems and methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP16024489 | 2017-03-22 | ||
EP17014132 | 2017-05-16 |
Publications (3)
Publication Number | Publication Date |
---|---|
GB201804604D0 GB201804604D0 (en) | 2018-05-09 |
GB2562599A true GB2562599A (en) | 2018-11-21 |
GB2562599B GB2562599B (en) | 2020-04-29 |
Family
ID=62068098
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB1804604.5A Active GB2562599B (en) | 2017-03-22 | 2018-03-22 | Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment |
Country Status (1)
Country | Link |
---|---|
GB (1) | GB2562599B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3673328A (en) * | 1971-02-26 | 1972-06-27 | Grason Stadler Co Inc | Rate of amplitude change control for audiometers of the von bekesy type |
EP2572640A1 (en) * | 2011-09-21 | 2013-03-27 | Coninx-Wittgens, Karin | Method and device for conducting a pure tone audiometry screening |
-
2018
- 2018-03-22 GB GB1804604.5A patent/GB2562599B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3673328A (en) * | 1971-02-26 | 1972-06-27 | Grason Stadler Co Inc | Rate of amplitude change control for audiometers of the von bekesy type |
EP2572640A1 (en) * | 2011-09-21 | 2013-03-27 | Coninx-Wittgens, Karin | Method and device for conducting a pure tone audiometry screening |
Also Published As
Publication number | Publication date |
---|---|
GB201804604D0 (en) | 2018-05-09 |
GB2562599B (en) | 2020-04-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8870786B2 (en) | Method and apparatus for tinnitus evaluation | |
Moore et al. | Effect of loudness recruitment on the perception of amplitude modulation | |
US10461711B2 (en) | Method and apparatus for outputting audio signal, method for controlling volume | |
JP5744745B2 (en) | Hearing measurement apparatus and method | |
US10341790B2 (en) | Self-fitting of a hearing device | |
CN109982637B (en) | Method for accurately estimating pure tone threshold using an unreferenced audio system | |
Busby et al. | Effects of threshold adjustment on speech perception in nucleus cochlear implant recipients | |
Rasetshwane et al. | Categorical loudness scaling and equal-loudness contours in listeners with normal hearing and hearing loss | |
US20190320946A1 (en) | Computer-implemented dynamically-adjustable audiometer | |
Eddins et al. | Developing clinically relevant scales of breathy and rough voice quality | |
Isarangura et al. | The effects of duration and level on spectral modulation perception | |
CN113302681B (en) | Noise masking apparatus and method for masking noise | |
Laback et al. | Temporal effects in interaural and sequential level difference perception | |
GB2562599A (en) | Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment | |
WO2018172472A1 (en) | Methods for user-dependent conditioning of stimuli in tests using a method of continuous adjustment | |
Jurado et al. | The effect of the helicotrema on low-frequency loudness perception | |
US11832936B2 (en) | Methods and systems for evaluating hearing using cross frequency simultaneous masking | |
Sottek | Sound quality evaluation of noises with spectro-temporal patterns | |
Florentine et al. | Maximum-likelihood yes-no procedure for gap detection: effect of track length | |
Julstrom et al. | Subjective assessment of cochlear implant users’ signal-to-noise ratio requirements for different levels of wireless device usability | |
Ward | Applications of loudness models in audio engineering | |
Rahmat et al. | The development of a fast method for recording Schroeder-phase masking functions | |
Toepken et al. | Differentiating between loudness and preference in the case of multi-tone stimuli | |
Villegas Rudolph et al. | Are existing speech-rating criteria adequate for characterizing speech in an unamplified soundscape? |