|Year : 2016 | Volume
| Issue : 2 | Page : 43-50
Implementation of objective audiometery among Suez Canal Authority workers
Mostafa K Madi MSc , Mohammed Shabana, Lobna Hamed, Noha A Hosny, Hedayat El Fouly
Audiovestibular Unit, Department of Otolaryngology, Al-Kasr Al-Aini School of Medicine, Cairo University, Cairo, Egypt
|Date of Submission||30-Nov-2016|
|Date of Acceptance||15-Dec-2016|
|Date of Web Publication||20-Mar-2017|
Mostafa K Madi
Audiovestibular Unit, Suez Canal Authority Hospital, Ismailia
Source of Support: None, Conflict of Interest: None
Verification of the hearing level in the malinger workers is a long-standing problem. Otolaryngologists and audiologists are often called upon to evaluate the auditory thresholds of workers who file claims for compensation as a result of noise-induced hearing loss. Although objective diagnostic methods tend to dominate modern medical science, behavioral pure-tone audiometry (PTA) remains the golden standard for identifying hearing threshold levels.
A number of auditory-evoked potential techniques have been implemented for this purpose over the past three decades. The most widely used of these techniques has been the auditory brainstem response (ABR) and more recently another auditory-evoked potential, the auditory steady-state response (ASSR). We also used old techniques such as postauricular myogenic potential and late cortical-evoked potential P100 as an alternative technique for objective audiometry.
Integration of different objective hearing tests is deficient in the literature on high-risk adult population.
To implement an objective protocol for assessing hearing in adult patients and for those difficult to test by routine PTA in Suez Canal Authority.
Materials and methods
This study was designed as a case–control study to collect and analyze data from September 2012 to be finished on June 2014. Sixty adult patients divided into two groups: those suffering from normal hearing and those suffering from sensorineural hearing loss. All patients in this research were submitted to the following: full history taking and otologic examination; basic audiological evaluations (PTA, speech audiometry and immittacemetry); tone burst ABR recorded using 500, 1000, 2000, and 4000 Hz stimulus; and ASSR stimulus using carrier frequencies 500, 1000, 2000, and 4000 Hz; postauricular myogenic potential response using 1000 and 4000 Hz; and late cortical-evoked potential P100 using 1000, 2000, and 4000 Hz.
In the normal hearing group, ASSR and ABR thresholds are closer to PTA thresholds than posterior auricular muscle (PAM) thresholds, the difference decrease with increasing frequency being closer at 4000 Hz than 500 Hz results. In the study group ASSR and ABR thresholds are approximated to PTA thresholds but still the ASSR thresholds are closer to PTA thresholds than ABR thresholds and PAM threshold but much higher in the case of P100. In the study group, ABR and ASSR thresholds show the best level of prediction of PTA thresholds. We found that the mean difference between all test and pure-tone thresholds had a tendency to be smaller with increasing frequency in both groups. However, the mean difference in the study group was statistically significantly lower than the control group. There are statistically significant positive correlation between PTA threshold and both ABR and ASSR threshold at all frequencies. A statistically significant correlation was found only at 1000 Hz in PAM test and a statistically significant correlation was found only at 1000 and 2000 Hz in P100 test.
ASSR is more accurate at higher frequencies, making ASSR more suitable in accessing auditory thresholds in patients with noise-induced hearing loss.
Keywords: auditory brainstem response, auditory steady-state response, P100, posterior auricular muscle, sensorineural hearing loss
|How to cite this article:|
Madi MK, Shabana M, Hamed L, Hosny NA, El Fouly H. Implementation of objective audiometery among Suez Canal Authority workers. Adv Arab Acad Audio-Vestibul J 2016;3:43-50
|How to cite this URL:|
Madi MK, Shabana M, Hamed L, Hosny NA, El Fouly H. Implementation of objective audiometery among Suez Canal Authority workers. Adv Arab Acad Audio-Vestibul J [serial online] 2016 [cited 2017 Oct 21];3:43-50. Available from: http://www.aaj.eg.net/text.asp?2016/3/2/43/202554
| Introduction|| |
Noise-induced sensorineural hearing loss (SNHL) is a permanent hearing impairment resulting from prolonged exposure to high levels of noise. Excessive noise is present in many situations. Some of the more common ones include occupational noise (machinery, etc.). Habitual exposure to noise above 85 dB will cause a gradual hearing loss in a significant number of individuals, and louder noises will accelerate this damage. For unprotected ears, the allowed exposure time decreases by one-half for each 5 dB increase in the average noise level .
Verification of the hearing level in malinger workers is a long-standing problem. Otolaryngologists and audiologists are often called upon to evaluate the auditory thresholds of workers who file claims for compensation as a result of noise-induced hearing loss. A principal concern in hearing loss compensation cases is exaggeration of the loss .
Although behavioral pure-tone audiometry (PTA) remains the golden standard for identifying hearing threshold levels, alternative methods become necessary to assess patients who are either too young to respond reliably, or who are uncooperative. Objective diagnostic methods tend to dominate modern medical science as many of the medicolegal patients who claimed compensation may exaggerate hearing loss that varies in degree, nature, and laterality .
A number of auditory-evoked potential (AEP) techniques have been implemented for this purpose over the past three decades. The most widely used of these techniques has been the auditory brainstem response (ABR). Its widespread popularity was as a result of the robust and highly replicable characteristics of the response irrespective of the patient’s state of consciousness. More recently another AEP, the auditory steady-state response (ASSR), has also gained popularity as an alternative technique for objective audiometry .
Although click ABRs have been used for many years to identify hearing loss in infants and young children, its specificity is limited to high frequency. It lacks of low tone information, such as 500–1000 Hz. ABR elicited by tone burst stimuli has reasonable accuracy and specific frequency. However, the clinical use of tone-evoked ABR tests is fairly limited for the unfamiliar waveform morphology and is time consuming. Besides, greater variability in the lower frequencies limits its clinical application in predicting behavioral thresholds from ABR thresholds .
P1–N1, however, has more frequency-specific stimuli compared with the ABR; the response is more resilient to electrophysiologic noise arising from small movements than is the ABR; and P1–N1 represents a more complete picture of the auditory system. It is unfortunate, especially in the USA, that the P1–N1 slow cortical response is underused, having been replaced by the ABR .
Though not popular, postauricular myogenic potential (PAMR) is an objective method for hearing assessment. During the 1970s and 1980s, before ABR becoming a popular clinical tool, PAMR was recommended as an objective hearing test for young children. The fact that it has not been used for this purpose is largely due to its perceived variability within and between patients . This variability appears to result from factors such as muscle tone, head and eye position, patient’s state, and recording filter pass band. Patuzzi and Thomson  found that PAMR can be reliably recorded in most adults when recording and stimulus parameters are optimal. Thornton in 1975 found that the mean difference between the click-evoked PAMR threshold and the 2 kHz audiometric threshold was 9 dB .
ASSR testing using modulated tones offers significant advantages over techniques that require short-duration acoustic stimuli. As the tones are continuous, they do not suffer from spectral distortion problems associated with brief tone bursts or clicks. Therefore, modulated tones are comparatively frequency specific. This specificity allows testing across the audiometric range and the generation of evoked potential audiograms, which, in patients with hearing impairments, can reflect the configuration of the loss accurately. Although ASSRs evoked by continuous modulated tones might estimate the hearing level more efficiently than ABRs, there are less clinical data available for these responses than for ABRs .
| Objectives|| |
- Pure-tone average.
- ABR threshold click and tone burst at frequencies 0.5, 1, 2, 4 kHz.
- ASSR threshold.
- Posterior auricular muscle (PAM) thresholds at frequencies 1, 4 kHz.
- P100 thresholds at frequencies 1, 2, 4 kHz.
- PTA configuration.
- Speech discrimination score.
- Sensitivity prediction by acoustic reflex threshold test of acoustic reflex (AR).
| Aim|| |
The aim of this study was to implement an objective protocol for assessing hearing in adult patients and for those difficult to test by routine PTA in Suez Canal Authority.
| Materials and methods|| |
Study design and sample size
This study was designed as a case–control study to collect and analyze data from September 2012 to June 2014. In this study, a total number of 60 patients (120 ears) were enrolled. Ears were grouped into two groups: (a) control group consisting of 40 patients (80 ears) and (b) a diseased group of 20 patients (40 ears). Their age range was 18–54 years. No statistically significant difference between the two groups as regards age (P=0.135), tested at the Suez Canal Authority Hospital in Ismailia. Only patients with type A tympanogram were included in this research. The emphases of the contrast of our result to sex as all workers in Suez Canal Authority are men.
Eligibility criteria included: (a) age range was 18–54 years; (b) presentation with noise-induced SNHL; and (c) referred from an otology outpatient to an audiovestibular clinic.
Data collection tools and methods
All patients in this research were submitted to the following:
- Full history taking.
- Full ENT examination.
- PTA and speech audiometry.
- Immitancemetery ‘tympanometry and acoustic reflexes’.
- Auditory brainstem-evoked potential response.
- Postauricular myogenic potential.
- Late cortical-evoked potential P100.
- Clinical Audiometer Interacoustics model AC40 (Copenhagen, Denmark).
- Immittancemeter Interacoustics model AT 235 (Copenhagen, Denmark) with a probe tone of 226 Hz.
- Two-channel AEP system, model Biologic Navigator PRO Natus (Copenhagen, Denmark).
Population of study and disease condition
Suez Canal Authority adult workers diagnosed as having bilateral SNHL, ranging from mild to moderately severe, with different audiometric configurations.
Adult patients having SNHL ranging from mild to moderately severe, with different audiometric configurations.
Patients included should be devoid of
- conductive hearing loss;
- mixed hearing loss;
- neurological disease;
- otological disease;
- vascular insufficiency.
Sample size (number of participants included)
- 20 patients as the study group.
- 40 normal subjects as a control group.
A written consent was signed by all patients showing their acceptance to participation in this study. Each patient was informed about all steps and of any possible side effects. The study has met institutional review board approval.
Statistical analyses of results were carried out using the SPSS system. The data collected were saved on Excel software (Microsoft Office Excel 2007) in a PC. Data were analyzed using statistical program for the social sciences (SPSS, version 18.0; SPSS Inc., Chicago, Illinois, USA). Data were statistically analyzed to evaluate the difference between groups and between each objective test in the groups under study as regards the various parameters.
The statistical analysis included: the arithmetic mean and SD for quantitative variables. Comparisons were made among different groups using the following tests:
- Independent samples t-test of significance was used when comparing between two mean.
- A one-way analysis of variance when comparing between more than two mean.
- χ2-Test of significance was used to compare proportions between two qualitative parameters.
- Pearson’s correlation coefficient (r) test was used for correlating data.
- P-value less than 0.05 was considered significant.
- P-value greater than 0.05 was considered insignificant.
| Results|| |
In this research we measured mean, SD, and range of hearing threshold level (in dBHL) of PTA, ABR, PAM, P100, and ASSR at different frequencies. We also measured the relation (correlation) between PTA and different tests at different frequencies. Hearing difference value between ABR, ASSR, PAM, P100 and pure-tone threshold was compared at different frequencies using comparison t-test or analysis of variance.
It was assumed that responses from each ear of a patient could be treated independently, so the results from right and left ears are pooled in [Table 1] and [Table 2].
|Table 1 Mean and SD of hearing threshold of pure-tone audiometry, auditory brainstem response, auditory steady-state response, posterior auricular muscle, and P100 in the normal hearing group|
Click here to view
|Table 2 Mean and SD of hearing threshold of pure-tone audiometry, auditory brainstem response, auditory steady-state response, posterior auricular muscle, and P100 in the study group|
Click here to view
Mean and SD of hearing thresholds of PTA, ABR, ASSR, PAM, and P100 in the normal and control group [Table 1] and [Table 2]. They revealed that ASSR thresholds are closer to PTA thresholds than ABR, PAM, and P100 thresholds; 4000 Hz results are better than 1000 and 2000 Hz results, while the ABR mean threshold is closer to PTA mean threshold only at 500 Hz. The highest difference being in the case of P100.
Mean and SD of hearing thresholds of PTA, ABR, ASSR, PAM, and P100 in the study group are shown in [Table 2]. They revealed that ASSR thresholds are closer to ABR thresholds than ABR, PAM, and P100 thresholds; 4000 Hz results are better than 500, 1000,and 2000 Hz results. Even the ASSR thresholds became better than PTA thresholds at 500 Hz. The highest difference being in the case of P100.
Regarding correlation between (ABR and ASSR) and PTA thresholds there was statistically significant positive correlation at all frequencies (P<0.05). The correlation coefficient increased with increasing frequency. While PAM thresholds showed positive correlation only at frequency 1000 Hz. Which concludes that ABR and ASSR closely reflected the configuration of the audiogram.
The difference between EP and PTA thresholds values were obtained by subtracting the pure-tone thresholds from the ASSR, ABR, P100, and PAM thresholds, respectively.
In the normal control group the mean differences between the threshold of pure tone and that of ABR, ASSR, PAM, and P100 for different frequencies are given in [Table 3]. We found that the mean differences between ASSR thresholds and PTA were lower than that of ABR, PAM, and P100 thresholds, the difference decrease with increasing frequencies; while the difference value at ABR mean threshold was lesser than the ASSR difference value only at 500 Hz. The highest difference was in the case of P100.
|Table 3 Mean and SD of difference value between hearing threshold of pure-tone audiometry and (auditory brainstem response, auditory steady-state response, posterior auricular muscle and P100) in the normal hearing group|
Click here to view
In the study group the mean differences between pure-tone threshold and that of ABR, ASSR, PAM, and P100 for different frequencies are given in [Table 4]. We found that the mean differences between ASSR thresholds and PTA were lower than that of ABR, PAM, and P100 thresholds at all frequencies, the difference decreases with increasing frequencies. The highest difference was the case of P100. The difference value had a tendency to be smaller with increasing frequency in the study group.
|Table 4 Mean and SD of difference value between hearing threshold of pure-tone audiometry and (auditory brainstem response, auditory steady-state response, posterior auricular muscle and P100) in the study group|
Click here to view
The mean threshold difference was statistically lower in ABR than ASSR at 0.5 kHz. The mean threshold differences were statistically lower in ASSR at 1 kHz, there was no statistical differences between ABR and PAM, while the mean threshold difference in P100 was statistically higher than other tests (ABR, ASSR, and PAM). This means that P100 the lesser test can be used as an objective test to detect hearing threshold.
The mean threshold differences were statistically lower in ASSR and ABR at 2 kHz; there was no statistical differences in ABR and ASSR, whereas the mean threshold difference in P100 was statistically higher than other tests (ABR and ASSR) and this also means that P100 the lesser test that can be used as an objective test to detect hearing threshold at 2 kHz. The mean threshold differences was statistically lower in ABR and ASSR at 4 kHz; there was no statistical differences in ABR and PAM, whereas the mean threshold difference in P100 was statistically higher than other tests (ABR, ASSR, and PAM), and this also means that p100 is the lesser sensitive test in objective assessment of hearing among other test.
| Discussion|| |
In this study the mean, SD, and range of pure-tone threshold ‘PTA’ in both groups are shown in [Table 1] and [Table 2], the increase in the mean of PTA threshold in the study group was more than the control group, this threshold increases with increasing frequency. Regarding this high-frequency sloping configuration would be attributed to their noise exposure.
In our results, the mean, SD, and range of auditory brainstem response ABR and ASSR thresholds are represented in [Table 1] and [Table 2]. The mean threshold in the study group was higher than the control group, and this threshold increases with increasing frequencies. There was a statistically significant difference at all tested frequencies 0.5–4 kHz. The increase in the mean threshold with increasing frequencies in the study group emphasizing high-frequency slop SNHL induced by noise exposure.
In the present study, there was a statistically significant positive correlation in the study group between mean ABR thresholds and mean PTA thresholds at all frequencies (500, 1000, 2000, and 4000 Hz). And this correlation increases with increasing frequencies. This significance correlation increases with increased severity of hearing loss. Our result agreed with Matson , who has declared that significant positive correlations between ABR and PTA thresholds were found in different studies. This emphasizes the good reliability of ABR for predicting hearing threshold in SNHL patients .
There was a statically significant difference between both groups regarding the difference value at all frequencies in both ears as represented in [Table 3] and [Table 4]. The difference value being lower in the study group and decrease with increasing frequencies; this could be explained by the fact that the study group had SNHL and the threshold increased especially at higher frequencies with less difference value, this would be attributed to the recruitment phenomenon.
Our results agreed with Stapells , who showed that the difference value would be 10–20 dB across the frequency range of 500–4000 Hz with the evoked potential thresholds being higher than the behavioral thresholds. He also declared better agreement between behavioral and ABR thresholds for higher frequencies (than for lower frequencies) and in persons with sensory hearing loss (in contrast to normal hearing). This result is consistent with the observation of Rance et al. (2006) , who found differences of about 10 dB in the case of moderate and severe hearing loss and about 20 dB in the case of normal hearing or mild hearing loss .
Mean ABR and ASSR thresholds demonstrate a significant statistical correlation with pure-tone thresholds at all frequencies (0.5, 1, 2, and 4 kHz). The strength of the relationship increases with increasing frequency and the best correlation is obtained at 4000 Hz. Several studies have demonstrated that ASSR thresholds have a strong relationship with pure-tone thresholds in adults with well-defined hearing losses. Dimitrijevic et al.  tested 59 ears of 31 adults with primarily sensory–neural hearing impairments ranging in severity from mild to severe, with nearly equal representation among mild, moderate, and severe degrees of loss. The ASSR thresholds showed a high correlation with the pure-tone thresholds, with r=0.92 for carriers in the range of 500–4000 Hz .
Correlation studies indicate that ASSR thresholds resemble more closely than ABR, PAM, and P100 in matching pure-tone thresholds when the degree of hearing loss increased. These results were agrees with Lee et al. (2008)  where they declared that ASSR thresholds demonstrate a positive correlation with pure-tone thresholds at all frequencies. The strength of the relationship increases with increasing frequency and the best correlation is obtained at 4000 Hz. This may be partly explained by the possible presence of recruitment, resulting in a more pronounced transition in ears with sensorineural losses .
This in agreement with the results of Ghannoum et al. , who found a significant difference between PTA and ASSR thresholds in groups of adult patients and that the difference decreased with increased severity of hearing loss. Moreover, Stapells and Herdman (2001)  found that ASSR thresholds in SNHLs were within 5–16 dB of behavioral thresholds for 500, 1000, 2000, and 4000 Hz .
In many studies using multiple frequency ASSR , it has been declared that errors in estimating audiometric threshold using the ASSR tend to be greater in normal-hearing individuals compared with those with hearing loss. Similar results have also been obtained by Swanepoel and Erasmus  in adult patients with moderate SNHL, confirming that the ASSR can reliably estimate behavioral thresholds of 60 dB HL and higher. Nevertheless, they recommend caution when estimating thresholds of less than 60 dB, due to increased variability . There are significant differences between studies reporting ranges of ASSR and behavioral threshold difference from 13 to 19 dB at 0.5–4 kHz . Our results agreed with Yin et al. (2008)  as they found that the mean difference between ASSR and pure-tone thresholds showed a tendency to be smaller with increasing frequency. These results support the results published by Stueve and Rourke . They explained the increased sensitivity of ASSR to the more severe degrees of hearing losses to the recruitment phenomenon which is associated with hearing impairment .
Another hypothesis that may explain the higher ASSR thresholds at low frequencies than that at high frequencies is that high-frequency stimuli elicit relatively larger ASSR amplitudes in sedated patients . This may lead to desynchronization of the neurons generating the responses because of jitter in the transmission time between the cochlear receptors and the neural generators. On the other hand, Kariya et al.  observed that in patients with SNHL, PTA thresholds and ASSR thresholds did not differ significantly at any frequencies, although they were significantly correlated .
PAMR was criticized as a useful test of hearing sensitivity; however, because of its variability within and between patients . This variability appears to be due to the effects of muscle tone and head position, the ‘habituation’ of the response with repeated stimulation, and the distorting effects of filter pass band on the response. Moreover, some patients with normal hearing do not have a measureable PAMR.
In our results, we measured the mean, SD, and range of the P1–N1 cortical ERP, threshold by using frequency-specific stimulus (frequencies 1, 2, and 4 kHz). We showed statistically significant differences between both groups regarding all tested frequencies (at 1, 2, and 4 kHz) (0.001). Regarding P100 threshold results there is a remarkable higher mean value of hearing threshold, declared poor sensitivity of this test, which may explain the less usage of this test in estimating hearing threshold. The results by Stapells and Van Maanen  were in contrast with our results, as he declared that P1–N1–P2 threshold estimates can be obtained in normal-hearing individuals as well as in individuals with hearing loss, and cortical auditory-evoked potential thresholds typically fall within 10 dB of behavioral thresholds as they used speech sounds to evoke cortical auditory-evoked potentials when the patients could easily behaviorally discriminate speech sounds such as/da/-/ga/ . Unlike ABRs, there are no published norms for clinicians to use for P1–N1–P2 responses because the peak latencies and amplitude vary depending on the stimuli used to evoke it (e.g. rise time, duration, speech, tones) .
| Conclusion|| |
ASSR is more accurate at higher frequencies, making it more suitable in accessing auditory thresholds in patients with noise-induced hearing loss.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Cai Q, Patel M, Coling D, Hu BH. Transcriptional changes in adhesion-related genes are site-specific during noise-induced cochlear pathogenesis. Neurobiol Dis 2012; 45:723–732.
Lins G, Picton W, Boucher L, Durieux A, Champagne C, Moran M et al.
Frequency specific audiometry using steady state responses. Ear Hear 2009; 17:81–96.
Komazec Z, Lemajic-Komazec S, Jovic R, Nadj C, Jovancevic L, Savovic S. Comparison between auditory steady-state responses and pure-tone audiometry. Vojnosanit Pregl 2010; 67:761–765.
Picton W, Dimitrijevic A, Perez Abolo M, Von Roon P. Estimating audiometric thresholds using auditory steady state responses. J Am Acad Audiol 2005; 16:140–156.
Stapells R, Van Maanen A. Comparison of multiple auditory steady-state responses (80 vs. 40 Hz) and slow cortical potentials for threshold estimation in hearing-impaired adults. Int J Audiol 2005; 44:613–624.
Patuzzi RB, Thomson SM. Auditory evoked response test strategies to reduce cost and increase efficiency: the postauricular muscle response revisited. Audiol Neurootol 2000; 5:322–332.
Sany ElBhaa HT, Kabel AH, Said N, Khaliel LH. Post auricular muscles response (PAMR) in auditory neuropathy (AN). Menoufia Med J 2009; 21:97–108.
Ahn H, Lee S, Kim J, Yoon H, Chung W. Comparing pure tone audiometry and auditory steady state response for the measurement of hearing loss. Otolaryngol Head Neck Surg 2007; 136:966–971.
Rance G, Tomlin D. Maturation of auditory steady-state responses in normal babies. Ear and Hear 2006; 27:20–29.
Dimitrijevic A, John S, Van Roon P, Purcell W, Adamonis J, Ostroff J et al.
Estimating the audiogram using multiple auditory steady state responses. J Am Acad Audiol 2002; 13:205–224.
Lee H, Ahn J, Chung J, Yoon T, Lee K. Clinical comparison of auditory steady state response with the click auditory steady state response. Clin Exp Otorhinolaryngol 2008; 12:184–188.
Ghannoum T, El-Khousht M, El-Abd S, Dabbous A, Soliman R. Comparison of auditory steady state response among normal hearers and patients with different degrees of sensorineural hearing loss. Med J Cairo Univ 2008; 76(Suppl II):349–358.
Stapells R, Herdman A. Current status of the auditory steady-state responses for estimating an infant’s audiogram. In Seewald RC, Bamford JM, editors. A sound foundation through early amplification 2004. Basel: Phonak AG; 2001. 43–59.
Swanepoel D, Erasmus H. Auditory steady state responses for estimating moderate hearing loss. Eur Arch Otorhinolaryngol 2007; 264:755–759.
Stueve M, Rourke C. Estimation of hearing loss in children: comparison of auditory steady state response, auditory brainstem response and behavioral test methods. Am J Audiol 2003; 12:125–136.
Yin SK, Feng YM, Chen ZN, Wang J. The effect of noise-induced sloping high-frequency hearing loss on the gap-response in the inferior colliculus and auditory cortex of guinea pigs. Hear Res 2008; 239:126–140.
Luts H, Desloovere C, Kumar A, Vandermersch E, Wouters J. Objective assessment of frequency specific hearing thresholds in babies. Int J Pediatr Otorhinolaryngol 2004; 68:915–926.
Kariya S, Fukushima K, Kawasaki A, Katoaoka J, Nishizaki K. Auditory steady-state responses to multiple simultaneous stimuli in children with functional or sensorineural hearing loss. Eur Arch Otorhinolaryngol 2008; 265:769–773.
Picton W, Skinner R, Champagne C, Kellett J, Maiste C. Potentials evoked by the sinusoidal modulation of the amplitude or the frequency of a tone. J Acoust Soc Am 1987; 82:165–178.
[Table 1], [Table 2], [Table 3], [Table 4]