ArticlePDF Available

Improving the Efficiency of Speech-In-Noise Hearing Screening Tests

May 2017
Ear and Hearing 38(6)

License
CC BY-NC-ND 4.0

Authors:

Amsterdam University Medical Center

Objective: Speech-in-noise hearing screening tests have become increasingly popular. These tests follow an adaptive procedure with a fixed number of presentations to estimate the speech reception threshold. The speech reception threshold is compared with an established cutoff signal to noise ratio (SNR) for a pass result or refer result. A fixed SNR procedure was developed to improve the efficiency of speech-in-noise hearing screening tests. Design: The cutoff SNR is used for all presentations in the fixed-SNR procedure. After each response a reliable test result is given (pass/refer) or an extra stimulus is presented. The efficiency and pass/refer rates between the adaptive procedure and the fixed-SNR procedure were compared. Results: An average reduction of 67% in the number of presentations can be achieved (from 25 to an average of 8.3 presentations per test). Conclusions: The fixed-SNR procedure is superior in efficiency to the adaptive procedure while having nearly equal refer and pass rates.This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Results of Monte Carlo simulations. A, The average number of presentations in the fixed-SNR procedure (filled circles) and the number of presentations in the adaptive procedure (solid line). B, Pass and refer rates for both procedures. For reference, the pass and refer rates for a fixed-SNR procedure with 25 presentations per test are shown as solid lines. SNR, signal to noise ratio.

…

Figures - uploaded by Cas Smits

Content may be subject to copyright.

Content uploaded by Cas Smits

Content may be subject to copyright.

<zdoi; 10.1097/AUD.0000000000000446>

the American Auditory Society, by Wolters Kluwer Health, Inc.• Printed in the U.S.A.

Objective: Speech-in-noise hearing screening tests have become

increasingly popular. These tests follow an adaptive procedure with a

ﬁxed number of presentations to estimate the speech reception thresh-

old. The speech reception threshold is compared with an established

cutoff signal to noise ratio (SNR) for a pass result or refer result. A ﬁxed

SNR procedure was developed to improve the efﬁciency of speech-in-

noise hearing screening tests.

Design: The cutoff SNR is used for all presentations in the ﬁxed-SNR

procedure. After each response a reliable test result is given (pass/refer)

or an extra stimulus is presented. The efﬁciency and pass/refer rates

between the adaptive procedure and the ﬁxed-SNR procedure were

compared.

Results: An average reduction of 67% in the number of presentations

can be achieved (from 25 to an average of 8.3 presentations per test).

Conclusions: The ﬁxed-SNR procedure is superior in efﬁciency to the

adaptive procedure while having nearly equal refer and pass rates.

Key words: Adaptive, Efﬁciency, Hearing screening test, Speech-in-noise

test, SRT.

(Ear & Hearing 2017;XX;00–00)

INTRODUCTION

Hearing loss is common in older adults and increases in

prevalence and severity as a function of age. The majority of

older adults do not seek professional help for hearing loss.

Screening could identify individuals with hearing loss and pro-

mote help-seeking. Smits et al. (2004) developed a low-cost

functional telephone-based hearing screening test. The test con-

sists of a ﬁxed number of spoken three-digit sequences (digit

triplets) presented in a noise background. The test follows an

adaptive procedure to determine the speech to noise ratio (SNR)

required to achieve 50% correct recognition, called the speech

reception threshold (SRT). This type of hearing screening tests

has become increasingly popular over the past decade for use

over telephone (Jansen et al. 2010; Watson et al. 2012; Zokoll

et al. 2013), internet (Smits et al. 2006), or smartphone (Potgi-

eter et al. 2015). The hearing screening test estimates the SRT

of the individual and compares the SRT with an established

cutoff SNR for a test result. If the SRT is lower (better) than

the cutoff SNR the test result is “pass;” if the SRT is higher

(worse) than the cutoff SNR then the test results is “refer.” The

aim of a screening test is to identify affected individuals. The

screening test is very accurate for individuals with true SRTs

that are much lower or much higher than the cutoff SNR. These

individuals will be classiﬁed correctly (pass or refer) despite

the measurement error associated with each SRT estimate. For

individuals with SRTs near the cutoff SNR, the accuracy will

decrease and will be lowest (i.e., 50% correct) for those with

an SRT corresponding to the cutoff SNR. The dependence of

test accuracy on the SRT suggests that using a ﬁxed number of

presentations is not efﬁcient. This is illustrated by an example

where the individual reaches an SNR well below the cutoff SNR

very quickly during the adaptive test procedure such that the test

result will be pass irrespective the correctness of the responses

to the presentations that will follow.

The aim of this study is to present a method that improves

the efﬁciency of hearing screening tests without decreasing the

sensitivity or speciﬁcity of the test. Essential for the approach is

the use of a ﬁxed SNR and a variable number of presentations

in the test.

TEST PROCEDURES

Adaptive Procedure

Most speech-in-noise hearing screening tests use a standard

1-up 1-down adaptive procedure to estimate the SRT. The SNR

of a presentation is based on the correctness of the previous

response. That is, if the stimulus is recognized correctly, the

next stimulus is presented at a lower SNR, and if the response

is incorrect, the next stimulus is presented at a higher SNR. The

difference in SNR between two consecutive presentations is the

step size of the procedure. Often the step size is constant dur-

ing the task. A ﬁxed number of n presentations is used and the

SRT is determined by averaging the SNRs of the presentations

including the virtual n + 1st presentation while omitting the ﬁrst

few presentations. A typical step size in speech-in-noise tests is

2 dB. The SRT is compared with the cutoff SNR to classify the

test result as a pass or refer.

Fixed-SNR Procedure

The optimized test procedure uses a ﬁxed SNR. That is, the

same SNR is used for all the presentations. The result is expressed

as the proportion of correctly recognized presentations. When k

correct responses occur among n presentations, the proportion

correct is k/n. The ratio k/n approaches p for large n, with p the

true probability for a correct response of the individual at the ﬁxed

SNR. The ratio k/n is compared with a predeﬁned proportion to

classify the test result as a pass or refer. The optimal choice for a

screening test is to use 0.5 for this predeﬁned proportion, which

corresponds to a ﬁxed SNR equal to the cutoff SNR. This means

Improving the Efﬁciency of Speech-In-Noise Hearing

Screening Tests

Cas Smits

Department of Otolaryngology-Head and Neck Surgery, Section Ear &

Hearing, and Amsterdam Public Health Research Institute, VU University

Medical Center, Amsterdam, The Netherlands.

Auditory Society, by Wolters Kluwer Health, Inc. This is an open-access

article distributed under the terms of the Creative Commons Attribution-

Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is

permissible to download and share the work provided it is properly cited.

The work cannot be changed in any way or used commercially without per-

mission from the journal.

American Auditory Society, by Wolters Kluwer Health, Inc.

Supplemental digital content is available for this article. Direct URL cita-

tions appear in the printed text and are provided in the HTML and text of

this article on the journal’s Web site (www.ear-hearing.com).

2 SMITS / EAR & HEARING, VOL. XX, NO. X, XXX–XXX

that at a ﬁxed SNR corresponding to the cutoff SNR is used for

all the presentations, and the test gives a pass result when the

proportion correct is >50% and the test gives a refer result when

the proportion correct is <50%. When using the same number

of presentations for the ﬁxed-SNR procedure as for the adaptive

procedure, the test characteristics (pass rate, refer rate etc.) are

better for the ﬁxed-SNR procedure but the difference is extremely

small (Smits, Reference Note 1).

The efﬁciency of the ﬁxed-SNR procedure can be improved

largely by using the following procedure:

1. present a stimulus and judge the response

2. calculate: (a) k/n; (b) the probability that p > 0.5, that is,

the cumulative probability P(p > 0.5); (c) the probability

that p < 0.5, that is, P(p < 0.5)

3. repeat steps (1) and (2) until P(p > 0.5) or P(p < 0.5)

reach a high value (typically a cumulative probability

of more than 95% would satisfy for good test character-

istics) or when the maximum number of presentations,

nmax, is reached

4. determine the test result: the test result is pass when k/n

> 0.5; the test result is refer when k/n < 0.5.

The cumulative probabilities P(p > 0.5) and P(p < 0.5) cannot

be calculated directly, but the probability distribution of p can be

estimated by using Bayes’ theorem (Koch 2007). The estimated

probability distribution of p (i.e., the posterior distribution),

f(p|n, k), depends on the number of correct responses k, the total

number of presentations n and the prior distribution. Here, a

uniform prior distribution is used that gives equal weight to all

possible values of p. Then the posterior distribution is

fpnk n

kn k

|, ()!

!( )!

()

−−−

11 (1)

Note that for p = k/n, the posterior distribution f(p|n, k) becomes

maximal. The cumulative probability that p > 0.5, P(p > 0.5) is

Pp fpnkdp>

()

∫

| (2)

and P(p < 0.5) = 1 − P(p > 0.5).

To illustrate the use of these equations in a screening test,

assume that 6 stimuli have been presented during the test and

nmax > 6. Then, in step (2) of the procedure, the posterior dis-

tribution, f(p|6, k), is calculated with Eq. (1) for k = 0, 1…,6

correct responses. Figure 1A shows the different distributions.

The cumulative probabilities P(p > 0.5) and P(p < 0.5) can be

calculated with Eq. (2) and represent the probabilities that the

true SRT is higher or lower than the cutoff SNR, respectively.

Figure 1B shows the results. When using a cumulative probabil-

ity of at least 0.925 to terminate the test, the test may be ended

with a pass result after ﬁve or six correct responses out of six,

or after only zero or one correct responses out of six with a refer

result. Step (1) and step (2) will be repeated and another stimu-

lus will be presented if the number of correct responses is 2, 3,

or 4. Thus, very good or very poor performers may need only six

presentations for a reliable screening test result, whereas more

presentations are used to get a reliable result for other listeners.

The calculations were performed for series up to 25 presen-

tations. For each number of presentations, n, the minimal num-

ber of correct responses, k, for a pass result and the maximum

number of correct responses for a refer result was determined.

The k values where P(p > 0.5) and P(p < 0.5) are larger than

0.925 were chosen because they yield an average P(p > 0.5) and

P(p < 0.5) of approximately 0.95. Thus, the test should classify

95% of the results correctly. Table 1 shows the results.

A COMPARISON BETWEEN THE ADAPTIVE

PROCEDURE AND THE FIXED-SNR PROCEDURE

The efﬁciency and test characteristics (i.e., the pass rate and

refer rate) of a typical adaptive speech-in-noise hearing screen-

ing test were compared with a ﬁxed-SNR hearing screening test.

Monte Carlo simulations with 10,000 runs per data point modeled

an adaptive speech-in-noise test with a 2-dB step size and 25 pre-

sentations per test, or a ﬁxed-SNR speech-in-noise test with a vari-

able number of presentations per test and nmax = 25. The procedures

as described above under “Adaptive Procedure” and “Fixed-SNR

Procedure” were used. Each run of the simulated adaptive pro-

cedure started at the true SRT; the estimated SRT was calculated

from the 25 SNRs in the run and the virtual 26th presentation. The

estimated SRT was compared with the cutoff SNR to classify the

Fig. 1. A, The estimated probability distribution (posterior distribution) for k correct responses out of six presentations. B, The cumulative probabilities

P(p > 0.5) and P(p < 0.5) as a function of the number of correct responses. The dotted line represents a cumulative probability of 0.925. The test may be ended

when the cumulative probability is higher than this value. For lower cumulative probabilities another stimulus needs to be presented.

SMITS / EAR & HEARING, VOL. XX, NO. X, XXX–XXX 3

test result. Each run of the simulated ﬁxed-SNR procedure yielded

a classiﬁed test result and the number of presentations used in the

simulation. The Monte Carlo simulations modeled SRTs from

−10 dB SNR to + 5 dB SNR in 0.1-dB steps, with a cutoff SNR of

−7 dB SNR. The speech recognition function associated with the

SRT of −10 dB SNR was a cumulative normal distribution with

a maximum slope of 20%/dB. The maximum slope of the speech

recognition function for higher SRTs was shallower and based

on the SRT value (Smits & Festen 2011). Figure 2A shows the

number of presentations, averaged over 10,000 runs, in the ﬁxed-

SNR procedure (represented by ﬁlled circles) and the number of

presentations in the adaptive procedure (i.e., 25 presentations per

test) represented by the solid line. The number of presentations in

the ﬁxed-SNR procedure, averaged over the range of SRTs, equals

8.3 which compares very favorably to the 25 presentations in the

adaptive procedure. Histograms showing the distribution of the

number of presentations for all integer SRTs from −10 to 5 dB

SNR are available in Supplemental Digital Content 1 (http://links.

lww.com/EANDH/A349). Figure 2B shows the pass and refer

rate as a function of SRT. These rates are very similar for both

procedures as demonstrated by the nearly identical functions. For

reference, the pass and refer rates for a reference ﬁxed-SNR pro-

cedure with 25 presentations per test were calculated and shown

as solid lines in Figure 2B. The percentage of wrongly classiﬁed

test results is 4.2% for the adaptive procedure, 4.8% for the ﬁxed-

SNR procedure, and 4.3% for the reference ﬁxed-SNR procedure.

Because a priori information about the true SRT was used in the

Monte Carlo simulations of the adaptive procedure (i.e., each run

started at the true SRT), the pass and refer rates are slightly better

than the reference ﬁxed-SNR procedure with an equal number of

presentations as in the adaptive procedure. Note that the average

number of presentations in the ﬁxed-SNR procedure and the per-

centages of wrongly classiﬁed test results depend on the distribu-

tion of SRTs in the test population.

DISCUSSION AND CONCLUSION

The ﬁxed-SNR procedure is superior in efﬁciency to the adap-

tive procedure while having nearly equal refer and pass rates. An

average reduction of 67% in the number of presentations was

achieved (from 25 to an average of 8.3 presentations per test)

for the modeled speech-in-noise tests and SRT population. The

reduction depends on the distribution of SRTs in the population

(Fig. 2A), but for most populations a reduction of at least 50%

in the average number of presentations can be achieved. These

numbers should be interpreted with caution because they are

only based on calculations and simulations. Listener-related fac-

tors could lead to differences in refer and pass rates between the

procedures. It may be hypothesized that, for example, an adap-

tive procedure aiming at 50% correct is more encouraging than

a ﬁxed-SNR procedure with very easy- or very difﬁcult-to-rec-

ognize stimuli for some listeners, which may have an effect on

performance. The adaptive speech-in-noise screening tests use a

few presentations (typically 4) to reach an SNR near the SRT.

The responses to these presentations are omitted from the SRT

calculation, which means that the total number of presentations

TABLE 1. The minimal and maximal number of correct responses, k, out of n presentations for a pass or refer result

n 4 5 6 7 8 9 10 11 12 13 14

Minimal k for pass 4 5 5 6 7 7 8 8 9 10 10

Maximal k for refer 0 0 1 1 1 2 2 3 3 3 4

P(p > 0.5); P(p < 0.5) 0.97 0.98 0.94 0.96 0.98 0.95 0.97 0.93 0.95 0.97 0.94

n 15 16 17 18 19 20 21 22 23 24 25

Minimal k for pass 11 11 12 13 13 14 14 15 16 16 17

Maximal k for refer 4 5 5 5 6 6 7 7 7 8 8

P(p > 0.5); P(p < 0.5) 0.96 0.93 0.95 0.97 0.94 0.96 0.93 0.95 0.97 0.95 0.96

The last row shows the corresponding cumulative probability that p is larger or smaller than 0.5.

Fig. 2. Results of Monte Carlo simulations. A, The average number of presentations in the ﬁxed-SNR procedure (ﬁlled circles) and the number of presentations

in the adaptive procedure (solid line). B, Pass and refer rates for both procedures. For reference, the pass and refer rates for a ﬁxed-SNR procedure with 25

presentations per test are shown as solid lines. SNR, signal to noise ratio.

4 SMITS / EAR & HEARING, VOL. XX, NO. X, XXX–XXX

in the adaptive procedure modeled in this article would be higher

than 25. For the ﬁxed-SNR procedure the test could start imme-

diately, which makes the test even more efﬁcient compared with

the adaptive test. However, a few trial presentations might be ben-

eﬁcial in the ﬁxed-SNR procedure to make the listener familiar

with the test and reduce a possible training effect. Experimental

data are needed to verify whether these dummy presentations are

necessary or not, and at which SNR they should be presented.

An advantage of the adaptive procedure is that it provides more

information than the ﬁxed-SNR procedure. The SRT, for exam-

ple, gives an estimate of the amount of hearing loss.

The applicability of the procedure is not limited to speech-

in-noise screening tests but it can be applied to essentially all

psychophysical screening tests. When the cutoff value corre-

sponds to a proportion correct that is different from 50%, then

the minimal number of correct responses needs to be deter-

mined for that speciﬁc percentage. Otherwise the values from

Table 1 can be used.

In conclusion, this study shows a highly efﬁcient procedure

for speech-in-noise hearing screening tests which can be easily

implemented in existing tests. The average number of presenta-

tions needed is only 30–50% of the number of presentations

used in an adaptive speech-in-noise screening test.

ACKNOWLEDGMENTS

The authors have no conﬂicts of interest to disclose.

Address for correspondence: Cas Smits, Department of Otolaryngology-

Head and Neck Surgery, Section Ear & Hearing, and Amsterdam Public

Health Research Institute, VU University Medical Center, P.O. Box 7057,

1007 MB Amsterdam, The Netherlands. E-mail: c.smits@vumc.nl

Received October 18, 2016; accepted March 20, 2017.

REFERENCES

Jansen, S., Luts, H., Wagener, K. C., et al. (2010). The French digit triplet

test: A hearing screening tool for speech intelligibility in noise. Int J

Audiol, 49, 378–387.

Koch, K. R. (2007). Introduction to Bayesian Statistics (2nd ed.). Springer

Science & Business Media, Heidelberg, Germany.

Potgieter, J. M., Swanepoel, d. e. W., Myburgh, H. C., et al. (2015). Devel-

opment and validation of a smartphone-based digits-in-noise hearing test

in South African English. Int J Audiol, 55, 405–411.

Smits, C., & Festen, J. M. (2011). The interpretation of speech reception

threshold data in normal-hearing and hearing-impaired listeners: Steady-

state noise. J Acoust Soc Am, 130, 2987–2998.

Smits, C., Kapteyn, T. S., Houtgast, T. (2004). Development and validation

of an automatic speech-in-noise screening test by telephone. Int J Audiol,

43, 15–28.

Smits, C., Merkus, P., Houtgast, T. (2006). How we do it: The Dutch func-

tional hearing-screening tests by telephone and internet. Clin Otolaryn-

gol, 31, 436–440.

Watson, C. S., Kidd, G. R., Miller, J. D., et al. (2012). Telephone screening

tests for functionally impaired hearing: Current use in seven countries

and development of a US version. J Am Acad Audiol, 23, 757–767.

Zokoll, M. A., Hochmuth, S., Warzybok, A., et al. (2013). Speech-in-noise

tests for multilingual hearing screening and diagnostics1. Am J Audiol,

22, 175–178.

REFERENCE NOTE

Smits, C. (2006). Hearing screening by telephone: Fundamentals and appli-

cations. Ph.D. Thesis. VU University Medical Center, Amsterdam.

Language-Independent Hearing Screening – Increasing the Feasibility of a Hearing Screening Self-Test at School-Entry

Article

Full-text available

Sep 2022

A tablet-based language-independent self-test involving the recognition of ecological sounds in background noise, the Sound Ear Check (SEC), was adapted to make it feasible for young children. Two experiments were conducted. The first experiment investigated the SEC‘s feasibility, as well as its sensitivity and specificity for detecting childhood hearing loss with a monaural adaptive test procedure. In the second experiment, the SEC sounds, noise, and test format were adapted based on the findings of the first experiment. The adaptations were combined with three test procedures, one similar to the one used in Experiment 1, one presenting the sounds dichotically in diotic noise, and one presenting all the sounds with a fixed signal-to-noise ratio and a stopping rule. Results in young children show high sensitivity and specificity to detect different grades of conductive and sensorineural hearing loss (70–90%). When using an adaptive, monaural procedure, the test duration was approximately 6 min, and 17% of the results obtained were unreliable. Adaptive staircase analyses showed that the unreliable results probably occur due to attention/motivation loss. The test duration could be reduced to 3-4 min with adapted test formats without decreasing the test-retest reliability. The unreliable test results could be reduced from 17% to as low as 5%. However, dichotic presentation requires longer training, reducing the dichotic test format‘s feasibility.

Improving the Efficiency of the Digits-in-Noise Hearing Screening Test: A Comparison Between Four Different Test Procedures

Article

Full-text available

Dec 2021
J SPEECH LANG HEAR R

Purpose This study compared the test characteristics, test–retest reliability, and test efficiency of three novel digits-in-noise (DIN) test procedures to a conventional antiphasic 23-trial adaptive DIN (D23). Method One hundred twenty participants with an average age of 42 years (SD = 19) were included. Participants were tested and retested with four different DIN procedures. Three new DIN procedures were compared to the reference D23 version: (a) a self-selected DIN (DSS) to allow participants to indicate a subjective speech recognition threshold (SRT), (b) a combination of self-selected and adaptive eight-trial DIN (DC8) that utilized a self-selected signal-to-noise ratio (SNR) followed by an eight-trial adaptive DIN procedure, and (c) a fixed SNR DIN (DF) approach using a fixed SNR value for all presentations to produce a pass/fail test result. Results Test–retest reliability of the D23 procedure was better than that of the DSS and DC8 procedures. SRTs from DSS and DC8 were significantly higher than SRTs from D23. DSS was not accurate to discriminate between normal-hearing and hard of hearing listeners. The DF and DC8 procedures with an adapted cutoff showed good hearing screening test characteristics. All three novel DIN procedure durations were significantly shorter (< 70 s) than that of D23. DF showed a reduction of 46% in the number of presentations compared to D23 (from 23 presentations to an average of 12.5). Conclusions The DF and DC8 procedures had significantly lower test durations than the reference D23 and show potential to be more time-efficient screening tools to determine normal hearing or potential hearing loss. Further studies are needed to optimize the DC8 procedure. The reference D23 remains the most reliable and accurate DIN hearing screening test, but studies in which the potentially efficient new DIN procedures are compared to pure-tone thresholds are needed to validate these procedures.

Musician-Advantage on Listening Effort for Speech in Noise Perception: A Dual-Task Paradigm Measure

Article

Full-text available

Aug 2023

Background and Objectives Speech in noise (SIN) perception is essential for effective day-to-day communication, as everyday conversations seldom transpire in silent environments. Numerous studies have documented how musical training can aid in SIN discrimination through various neural-pathways, such as experience-dependent plasticity and overlapping processes between music and speech perception. However, empirical evidence regarding the impact of musical training on SIN perception remains inconclusive. This study aimed to investigate whether musicians trained in South Indian classical “Carnatic” style of music exhibited a distinct advantage over their non-musician counterparts in SIN perception. The study also attempted to explore whether the listening effort (LE) associated in this process was different across musicians and non-musicians, an area that has received limited attention. Subjects and Methods A quasi-experimental design was employed, involving two groups comprising 25 musicians and 35 non-musicians, aged 18-35 years, with normal hearing. In phase 1, participants’ musical abilities were assessed using the Mini-Profile of Music Perception Skills (Mini-PROMS). In phase 2, SIN abilities were tested using the Tamil phonemically balanced words and Tamil Matrix Sentence Test at -5 dB, 0 dB, and +5 dB SNR. Phase 3 tested LE using a dual-task paradigm including auditory and visual stimuli as primary and secondary tasks. Results Fractional logit and linear regression models demonstrated that musicians outperformed non-musicians in the Mini-PROMS assessment. Musicians also fared better than non-musicians in SIN and LE at 0 dB SNR for words and +5 dB SNR for sentences. Conclusions The findings of this study provided limited evidence to support the claim that musical training improves speech perception in noisy environments or reduces the associated listening effort.

Listening Effort, An Overview of App Validation and Testing by the Audiology 4 all Project

Article

Full-text available

Jan 2023

A language-independent hearing screening self-test at school-entry

Article

Full-text available

Jan 2024

The usage of a tablet-based language-independent self-test involving the recognition of ecological sounds in background noise, the Sound Ear Check, was investigated. The results of 692 children, aged between 5 and 9 years and 4 months, recruited in seven different countries, were used to analyze the validity and the cultural independence of test. Three different test procedures, namely a monaural adaptive procedure, a procedure presenting the sounds dichotically in diotic noise, and a procedure presenting all the sounds with a fixed signal-to-noise ratio and a stopping rule were studied. Results showed high sensitivity and specificity of all three procedures to detect conductive, sensorineural and mixed hearing loss > 30 dB HL. Additionally, the data collected from different countries were consistent, and there were no clinically relevant differences observed between countries. Therefore, the Sound Ear Check can offer an international hearing screening test for young children at school entry, solving the current lack of hearing screening services on a global scale.

Hearing Assistive Technology Facilitates Sentence-in-Noise Recognition in Chinese Children With Autism Spectrum Disorder

Article

Full-text available

Jul 2023
J SPEECH LANG HEAR R

Purpose Hearing assistive technology (HAT) has been shown to be a viable solution to the speech-in-noise perception (SPIN) issue in children with autism spectrum disorder (ASD); however, little is known about its efficacy in tonal language speakers. This study compared sentence-level SPIN performance between Chinese children with ASD and neurotypical (NT) children and evaluated HAT use in improving SPIN performance and easing SPIN difficulty. Method Children with ASD (n = 26) and NT children (n = 19) aged 6–12 years performed two adaptive tests in steady-state noise and three fixed-level tests in quiet and steady-state noise with and without using HAT. Speech recognition thresholds (SRTs) and accuracy rates were assessed using adaptive and fixed-level tests, respectively. Parents or teachers of the ASD group completed a questionnaire regarding children's listening difficulty under six circumstances before and after a 10-day trial period of HAT use. Results Although the two groups of children had comparable SRTs, the ASD group showed a significantly lower SPIN accuracy rate than the NT group. Also, a significant impact of noise was found in the ASD group's accuracy rate but not in that of the NT group. There was a general improvement in the ASD group's SPIN performance with HAT and a decrease in their listening difficulty ratings across all conditions after the device trial. Conclusions The findings indicated inadequate SPIN in the ASD group using a relatively sensitive measure to gauge SPIN performance among children. The markedly increased accuracy rate in noise during HAT-on sessions for the ASD group confirmed the feasibility of HAT for improving SPIN performance in controlled laboratory settings, and the reduced post-use ratings of listening difficulty further confirmed the benefits of HAT use in daily scenarios.

Hearing Assistive Technology Facilitates Sentence-in-Noise Recognition in Children with Autism Spectrum Disorder

Preprint

Full-text available

Mar 2023

Purpose: Hearing assistive technology (HAT) has been shown to be a viable solution to the speech-in-noise perception (SPIN) issue in children with autism spectrum disorder (ASD); however, little is known about its efficacy in tonal language speakers. This study compared sentence-level SPIN performance between Chinese children with ASD and neurotypical (NT) children and evaluated HAT use in improving SPIN performance and easing SPIN difficulty. Methods: Children with ASD (n=26) and NT children (n=19) aged 6-12 performed two adaptive tests in steady-state noise and three fixed-level tests in quiet and steady-state noise with and without using HAT. Speech recognition thresholds (SRT) and accuracy rates were assessed using adaptive and fixed-level tests, respectively. Parents or teachers of the ASD group completed a questionnaire regarding children’s listening difficulty under six circumstances before and after a ten-day trial period of HAT use. Results: Although the two groups of children had comparable SRTs, the ASD group showed a significantly lower SPIN accuracy rate than the NT group. Also, a significant impact of noise was found in the ASD group’s accuracy rate, but not in the NT group’s. There was a general improvement in the ASD group’s SPIN performance with HAT and a decrease in their listening difficulty ratings across all conditions after the device trial. Conclusion: The findings indicated inadequate SPIN in the ASD group using a relatively sensitive measure to gauge SPIN performance among children. The markedly increased accuracy rate in noise during HAT-on sessions for the ASD group confirmed the feasibility of HAT for improving SPIN performance in controlled laboratory settings, and the reduced post-use ratings of listening difficulty further confirmed the benefits of HAT use in daily scenarios.

Factors Influencing the Korean Version of the Digit-in-Noise Test

Article

Full-text available

Mar 2023

Background and objectives: : The digits-in-noise (DIN) test was developed as a simple and time-efficient hearing-in-noise test worldwide. The Korean version of the DIN (K-DIN) test was previously validated for both normal-hearing and hearing-impaired listeners. This study aimed to explore the factors influencing the outcomes of the K-DIN test further by analyzing the threshold (representing detection ability) and slope (representing test difficulty) parameters for the psychometric curve fit. Subjects and methods: : In total, 35 young adults with normal hearing participated in the K-DIN test under the following four experimental conditions: 1) background noise (digit-shaped vs. pink noise); 2) gender of the speaker (male vs. female); 3) ear side (right vs. left); and 4) digit presentation levels (55, 65, 75, and 85 dB). The digits were presented using the method of constant stimuli procedure. Participant responses to the stimulus trials were used to fit a psychometric function, and the threshold and slope parameters were estimated according to pre-determined criteria. The accuracy of fit performance was determined using the root-mean-square error calculation. Results: : The listener's digit detection ability (threshold) was slightly better with pink noise than with digit-shaped noise, with similar test difficulties (slopes) across the digits. Gender and the tested ear side influenced neither the detection ability nor the task difficulty. Additionally, lower presentation levels (55 and 65 dB) elicited better thresholds than the higher presentation levels (75 and 85 dB); however, the test difficulty varied slightly across the presentation levels. Conclusions: : The K-DIN test can be influenced by stimulus factors. Continued research is warranted to understand the accuracy and reliability of the test better, especially for its use as a promising clinical measure.

Sensitivity of the antiphasic digits-in-noise test to simulated unilateral and bilateral conductive hearing loss

Article

Full-text available

Sep 2022

Objectives The objective of this study is (1) to assess whether the presentation level of the antiphasic digits-in-noise (DIN) test affects the speech recognition threshold (SRT), (2) to evaluate how accurately simulated unilateral and bilateral conductive hearing loss is detected (CHL) and (3) to determine whether increasing the presentation level normalises the antiphasic DIN SRT. Design Participants performed antiphasic and diotic DINs at different presentation levels with unilateral, bilateral or no earplugs. Study sample Twenty-four and twelve normal hearing adults. Results Without earplugs, antiphasic DIN SRTs did not differ between 60 and 80 dB SPL. At 60 dB SPL, the antiphasic DIN correctly classified 92% of the unilateral earplug cases; the diotic DIN 25%. The binaural intelligibility level difference did not differ between the no-earplug condition and the condition with bilateral earplugs when the presentation was increased with the attenuation level. Conclusions In normal hearing participants, diotic and antiphasic DIN SRTs are independent of presentation level above a minimum level of 60 dB SPL. The antiphasic DIN is more sensitive than the diotic DIN for detecting unilateral CHL; not for bilateral CHL. The effect of CHL on DIN SRTs can be largely compensated by increasing the presentation level. Audibility plays an important role in the antiphasic and diotic DIN.

The digit triplet test for school-age hearing screening: comparing languages, school grades and scoring methods in a large sample

Article

Jan 2025
Int J Audiol

Objective: To investigate the reliability of a bilingual school-age hearing screening in four school grades based on the Digit Triplet Test (DTT) in two languages and to investigate three calculation methods for referral values in their ability to detect hearing losses and avoid false-positive results. Design and study sample: 3255 children, aged between 10 and 17 years old, were tested during a systematic hearing screening program in a bilingual, French-German area in Belgium. French speaking children were tested with a French DTT, German children were tested with a German DTT. The SRT-values, their stability and measurement error were investigated per grade and language. The number of false-positive results was studied for three referral methods, using additional audiometric data of 71 children. Results: Our data showed that reliable results with high stability and a small measurement error can be obtained in only around two minutes per ear. Differences between languages are minimal and grade-specific referral values were necessary. A referral method considering the lowest SNRs in the adaptive staircase reduces the number of false-positive results substantially. Conclusion: The DTT versions of different languages can be used reliably in a bilingual school-age hearing screening program when grade-specific, alternative referral methods are implemented.

Development and validation of a smartphone-based digits-in-noise hearing test in South African English

Article

Full-text available

Apr 2016
Int J Audiol

Objective: The objective of this study was to develop and validate a smartphone-based digits-in-noise hearing test for South African English. Design: Single digits (0-9) were recorded and spoken by a first language English female speaker. Level corrections were applied to create a set of homogeneous digits with steep speech recognition functions. A smartphone application was created to utilize 120 digit-triplets in noise as test material. An adaptive test procedure determined the speech reception threshold (SRT). Experiments were performed to determine headphones effects on the SRT and to establish normative data. Study sample: Participants consisted of 40 normal-hearing subjects with thresholds ≤15 dB across the frequency spectrum (250-8000 Hz) and 186 subjects with normal-hearing in both ears, or normal-hearing in the better ear. Results: The results show steep speech recognition functions with a slope of 20%/dB for digit-triplets presented in noise using the smartphone application. The results of five headphone types indicate that the smartphone-based hearing test is reliable and can be conducted using standard Android smartphone headphones or clinical headphones. Conclusion: A digits-in-noise hearing test was developed and validated for South Africa. The mean SRT and speech recognition functions correspond to previous developed telephone-based digits-in-noise tests.

Speech-in-Noise Tests for Multilingual Hearing Screening and Diagnostics1

Article

Full-text available

Jun 2013
Am J Audiol

Purpose New complementary multilingual speech-in-noise tests in Russian, Turkish, and Spanish for hearing self-screening purposes and follow-up hearing diagnostics are compared to the speech tests of the European project, HearCom (Hearing in the Communication Society). Method The tests consist of spoken numbers (Digit Triplet Test; Smits, Kapteyn, & Houtgast, 2004) or sentences (Matrix Test; e.g., Hagerman, 1982) presented in a background noise and estimate the speech reception threshold, which is the signal-to-noise ratio that yields 50% speech intelligibility. All tests were developed according to the HearCom minimum quality standards for speech intelligibility tests. This report presents a cross-language comparison of reference speech intelligibility functions for monaural headphone measurements with normal-hearing listeners. The same model function was employed to describe the speech intelligibility functions for all of the tests. Results Reference speech intelligibility functions of the new versions of the Digit Triplet Test and Matrix Test show high comparability to the HearCom tests. In order to achieve the highest possible comparability across languages, language- and speaker-dependent factors in speech intelligibility should be compensated for. Conclusion To date, several complementary tests for screening and diagnostics have been developed in several languages. Adhering to the HearCom standards, the tests are highly comparable across languages. For the Matrix Test, equal syntax and linguistic complexity were maintained across languages due to common methodological standards.

The interpretation of speech reception threshold data in normal-hearing and hearing-impaired listeners: Steady-state noise

Article

Full-text available

Nov 2011
J ACOUST SOC AM

Speech-in-noise-measurements are important in clinical practice and have been the subject of research for a long time. The results of these measurements are often described in terms of the speech reception threshold (SRT) and SNR loss. Using the basic concepts that underlie several models of speech recognition in steady-state noise, the present study shows that these measures are ill-defined, most importantly because the slope of the speech recognition functions for hearing-impaired listeners always decreases with hearing loss. This slope can be determined from the slope of the normal-hearing speech recognition function when the SRT for the hearing-impaired listener is known. The SII-function (i.e., the speech intelligibility index (SII) against SNR) is important and provides insights into many potential pitfalls when interpreting SRT data. Standardized SNR loss, sSNR loss, is introduced as a universal measure of hearing loss for speech in steady-state noise. Experimental data demonstrates that, unlike the SRT or SNR loss, sSNR loss is invariant to the target point chosen, the scoring method or the type of speech material.

The French digit triplet test: A hearing screening tool for speech intelligibility in noise

Article

Full-text available

May 2010
Int J Audiol

A French speech intelligibility screening test in noise that applies digit triplets as stimuli has been developed and evaluated for both telephone and broadband headphone use. After optimizing the speech material based on the intelligibility of the individual digits, norms for normal-hearing subjects were established. speech reception thresholds (SRTs) of -6.4 +/- 0.4 and -10.5 +/- 0.3 dB SNR, and slopes of 17.1 and 27.1 %/dB were obtained for telephone and broadband headphone presentation, respectively. The French digit triplet test by telephone was then implemented as an automatic self-screening test by home telephone, and further evaluated in normal-hearing and hearing-impaired listeners. A test-retest variability of 0.7 dB was found and the correlation between SRT and pure-tone average (PTA(0.5,1,2,4)) was 0.77. One month after launching the test, 20,000 calls were registered. It can be concluded that both versions of the newly developed test have steep slopes and small SRT differences across normal-hearing listeners. The screening test by telephone is highly reliable and proves to fulfill the need for an easily accessible and objective hearing screening.

Hearing screening by telephone : fundamentals & applications /

Article

Full-text available

Jan 2005

Proefschrift Vrije Universiteit Amsterdam, 2006. Met lit. opg. - Samenvatting in het Nederlands. Auteursnaam op omslag: Cas Smits.

Development and validation of an automatic speech-in-noise screening test by telephone

Article

Full-text available

Feb 2004

To meet the need for an objective self-test for hearing screening. a new Dutch speech-in-noise test was developed. Digit triplets were used as speech material. The test was made fully automatic, was controlled by a computer, and can be done by telephone. It measures the speech reception threshold (triplet SRT(n)) using an adaptive procedure, in about 3 min. Our experiments showed no significant influence of telephone type or listening environment. Measurement errors were within 1 dB. which makes the test accurate. In additional experiments with hearing-impaired subjects (76 ears of 38 listeners), the new test was compared to the existing sentence SRT(n) test of Plomp and Mimpen, which is considered to be the standard. The correlation between both SRT(n)s was 0.866. As expected, correlations between the triplet SRT(n) test by telephone and average pure-tone thresholds are somewhat lower: 0.732 for PTA(0,5,1,2), and 0.770 for PTA(0,5,2,4). When proper SRT(n) values were chosen for distinguishing between normal-hearing and hearing-impaired subjects, the triplet SRT(n) test was found to have a sensitivity of 0.91 and a specificity of 0.93.

How we do it: The Dutch functional hearing-screening tests by telephone and internet

Article

Full-text available

Nov 2006

• The Dutch National Hearing Test is a reliable and very successful functional hearing–screening test by telephone. An internet version of the National Hearing Test was also implemented. • The National Hearing Test is a fully automatic adaptive speech-in-noise test that uses digit-triplets as speech material. The result of the test is given as ‘good,’‘insufficient,’ or ‘poor.’ • The test by telephone performs better in reaching older subjects, who are more likely to suffer from hearing loss, than the test by internet. • More than 50% of the participants who scored ‘insufficient’ or ‘poor’ followed the recommendation to visit a GP, hearing-aid dispenser, ENT specialist or Audiological Center. • The tests contribute to increase the identification and treatment of older hearing-impaired subjects.

Introduction to Bayesian Statistics

Book

Jan 2007

Karl-Rudolf Koch

The Introduction to Bayesian Statistics (2nd edition) presents Bayes' theorem, the estimation of unknown parameters, the determination of confidence regions and the derivation of tests of hypotheses for the unknown parameters. It does so in a manner that is simple, intuitive and easy to comprehend. The methods are applied to linear models, in models for a robust estimation, for prediction and filtering and in models for estimating variance components and covariance components. Regularization of inverse problems and pattern recognition are also covered while Bayesian networks serve for reaching decisions in systems with uncertainties. If analytical solutions cannot be derived, numerical algorithms are presented, such as the Monte Carlo integration and Markov Chain Monte Carlo methods. © Springer-Verlag Berlin Heidelberg 2007. All rights are reserved.

Telephone Screening Tests for Functionally Impaired Hearing: Current Use in Seven Countries and Development of a US Version

Article

Nov 2012

Background: An estimated 36 million US citizens have impaired hearing, but nearly half of them have never had a hearing test. As noted by a recent National Institutes of Health/National Institute on Deafness and Other Communication Disorders (NIH/NIDCD) Working Group, "In the United States (in contrast to many other nations) there are no readily accessible low cost hearing screening programs…" (Donahue et al, 2010, p. 2). Since 2004, telephone administered screening tests utilizing three-digit sequences presented in noise have been developed, validated, and implemented in seven countries. Each of these tests has been based on a test protocol conceived by Smits and colleagues in The Netherlands. Purpose: Investigators from Communication Disorders Technology, Inc., Indiana University, and VU University Medical Center of Amsterdam agreed to collaborate in the development and validation of a screening test for hearing impairment suitable for delivery over the telephone, for use in the United States. This test, utilizing spoken three-digit sequences (triplets), was to be based on the design of Smits and his colleagues. Research design: A version of the digits-in-noise test was developed utilizing digit triplets spoken in Middle American dialect. The stimuli were individually adjusted to speech-to-noise ratio (SNR) values yielding 50% correct identification, on the basis of data collected from a group of 10 young adult listeners with normal hearing. A final set of 64 homogeneous stimuli were selected from an original 160 recorded triplets. Each test consisted of a series of 40 triplets drawn at random, presented in a noise background. The SNR threshold for 50% correct identification of the triplets was determined by a one-down, one-up adaptive procedure. The test was implemented by telephone, and administered to listeners with varying levels of hearing impairment. The listeners were then evaluated with pure-tone tests and other audiometric measures as clinically appropriate. Study sample: Ninety participants included 72 who were volunteers from the regular client population at the Indiana University Hearing Clinic, and 18 who were recruited with a newspaper ad offering a free hearing test. Of the 90 participants, 49 were later determined to have mean pure-tone thresholds greater than 20 dB hearing level (HL). Data collection and analysis: The primary data analyses were correlations between telephone test thresholds and other measures, including pure-tone thresholds and speech recognition tests, collected for the same participants. Results: The correlation between the telephone test and pure-tone thresholds (r = 0.74) was within the range of correlations observed with successful telephone screening tests in use in other countries. Thresholds based on the average of only 21 trials (trials five through 25 of the 40-trial tracking history) yielded sensitivity and specificity values of 0.80 and 0.83, respectively, using pure-tone average((0.5, 1.0, 2.0 kHz)) >20 dB HL as the criterion measure. Conclusions: This US version of the digits-in-noise telephone screening test is sufficiently valid to be implemented for use by the general public. Its properties are quite similar to those telephone screening tests currently in use in most European countries. Telephone tests provide efficient, easy to use, and valid screening for functional hearing impairment. The results of this test are a reasonable basis for advising those who fail to seek a comprehensive hearing evaluation by an audiologist.

Improving the Efficiency of Speech-In-Noise Hearing Screening Tests

Abstract and Figures

Recommended publications

Efficiency of incoherent reception of the FSK signal on the background of impulse FSK interference i...

ORBITALLY NONEXPANSIVE MAPPINGS

Spectral Efficiency Enhancement in the Multi-user MIMO Downlink With Partial Feedback Information

On the Efficiency of Generalized Antithetic Transformations for Monte Carlo Integration