Home Email this page Print this page Bookmark this page Decrease font size Default font size Increase font size
Noise & Health  
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Email Alert *
Add to My List *
* Registration required (free)  

   Article Figures
   Article Tables

 Article Access Statistics
    PDF Downloaded4    
    Comments [Add]    

Recommend this journal


  Table of Contents    
Year : 2022  |  Volume : 24  |  Issue : 113  |  Page : 49-60
Cochlear synaptopathy causes loudness perception impairment without hearing loss

1 Audiology Department Health Sciences Faculty, Hacettepe University, Ankara, Turkey
2 Audiology Department Health Sciences Faculty, Ankara University, Ankara, Turkey

Click here for correspondence address and email
Date of Submission17-Sep-2020
Date of Decision18-Apr-2021
Date of Acceptance13-May-2021
Date of Web Publication25-Jul-2022

Purpose: In this study, the development of a quantitative measurement method to predict long-term auditory adaptation through the stimuli that have been modulated according to different short-term modulation types was aimed to form a psychoacoustic test battery. It might be used in the evaluation process of individuals with hidden hearing loss. Methods: The individuals participating in our study were separated into two groups: high-risk group (n = 39) and low-risk group (n = 30) according to the noise-exposure score. To all participants, auditory brainstem response (ABR), dichotically digit test, Turkish matrix sentence test, otoacoustic emissions test, amplitude modulation detection test, and loudness adaptation test were applied. Stimuli, used in loudness adaptation tests, were provided in three different experiment pairs (experiment 1–2, experiment 3–4, and experiment 5–6). Results: The amplitude of wave I of ABR increased as the intensity level increased in the low-risk group, whereas the amplitude reduced as the intensity level increased in the high-risk group (P < 0.05). When different carrier frequency stimuli were used in amplitude modulation detection test, we found that loudness adaptation was highest at 1 kHz carrier frequency with background noise (P < 0.05). Conclusion: We observed that individuals assumed having hidden hearing loss had high adaptation scores. It was thought that this result might be related to auditory nerve fibers with low spontaneous rate and thus distortion in temporal coding skills might lead to abnormal loudness adaptation, especially with contralateral noise.

Keywords: Auditory brainstem responses, hidden hearing loss, loudness adaptation, matrix

How to cite this article:
Cildir B, Tokgoz-Yilmaz S, Türkyilmaz MD. Cochlear synaptopathy causes loudness perception impairment without hearing loss. Noise Health 2022;24:49-60

How to cite this URL:
Cildir B, Tokgoz-Yilmaz S, Türkyilmaz MD. Cochlear synaptopathy causes loudness perception impairment without hearing loss. Noise Health [serial online] 2022 [cited 2022 Aug 19];24:49-60. Available from: https://www.noiseandhealth.org/text.asp?2022/24/113/49/351968

  Introduction Top

Noise-induced hearing loss is one of the most frequently encountered health problems in the industrial field in worldwide.[1],[2] According to World Health Organization, approximately 1.1 million adolescents and young adults aged 12 to 35 years are at risk of hearing loss due to exposure to various sounds such as night clubs, some sports, firearms, and high-volume sirens (fire trucks, ambulance, etc.).[3]

After noise exposure (even in accompanying single high-dose noise), losses in synaptic connections (especially ribbon synapse connections) between auditory nerve (AN) fibers (afferent type I) and cochlear hair cells may occur without affecting hearing thresholds.[4],[5],[6],[7] Kujawa and Liberman, according to their findings from auditory brainstem responses (ABRs) and distortion product otoacoustic emission (DPOAE) on animals, indicated that permanent synapse loss could occur up to 50% within the frequency field in which temporal hearing threshold change was maximal.[6] Researchers who investigated the relationship between behavioral threshold and inner hair cells pointed out that up to 80% inner hair cell loss causes <5 dB threshold change.[8] In most biologic systems, it is underlined that the minuscule amount of cell loss does not result in any functional distortion.[9] As only 20% of inner hair cells are sufficient to maintain hearing sensitivity in a quiet environment, it is stated that this amount of loss is more negatively effective in noisy environments.[10] It is pointed out that noise-induced damage on suprathreshold responses is correlated with the damage on synaptic connections of AN fibers with high firing threshold and low spontaneous rates.[10] However, it is relatively less related to a loss on the synapses with low firing threshold and high spontaneous rates AN fibers.[11] As it is not always possible to detect such a noise-induced hearing loss audiologically, this disorder is defined by different expressions such as neuropathy, cochlear synaptopathy, or most popularly, hidden hearing loss (HHL).[8] The permanent loss of AN fibers can be identified by the amplitude of wave I of ABR, which reflects auditory neural function.[11],[12] This reduction in the amplitude is observed as a response to mild and high-intensity levels, but not as a response to low-intensity levels.[13],[14] Low threshold fibers should not be suppressed to obtain information from the neural responses about the medium intensity sounds. In solving this problem, it had been reported that the auditory cortex performs an adaptive encoding, and these neurons adapt to the incoming responses by changing the dynamic range to reflect the sound intensity in any environment. The threshold adaptation in neural responses changes depending on the deterioration in intensity sounds where neural responses increase or decrease.[15]

Low spontaneous rate AN fibers show more adaptation than moderate or high spontaneous rate AN fibers.[16] It has been suggested that the reason why loudness adaptation affects the high frequencies most may be related to the auditory responses in the basal region of the cochlea, where nerve fibers with low spontaneous rates are adapted.[17]

We aimed the development of a quantitative measurement method to predict long-term auditory adaptation through the stimuli that were been modulated by different short-term modulation types (modulation frequency, modulation depth, and carrier frequency). The enabling of conducting a rehabilitation plan and follow-up for individuals with HHL through evaluation of long-term auditory adaptation level and producing a psychoacoustic test battery are among purposes of this study.

  Methods Top


In Hacettepe University Health Sciences Faculty, Audiology Department, 69 individuals with normal hearing threshold were evaluated for 2 years, between the years 2016 and 2018. This study was approved by Ethics Committee: Hacettepe University Non-Interventional Clinical Researches Ethics Committee on July 26, 2016 (GO 16/510). All individuals who signed informed consent forms and have been exposed to varying noise types and intensities were tested with “1-minute noise screen.”[18] They were separated into two groups: high- and low-risk groups with respect to this questionnaire. The high-risk group included 39 participants, of whom 16 were men and 23 were women with a noise exposure score of 5 or above (mean age = 21.3 ± 3.46 and range = 18–32), whereas the low-risk group consisted of 30 participants, of whom 8 were men and 22 were women with a noise exposure score of 4 and below (mean age = 20.9 ± 2.66 and range = 18–32). The inclusion criteria were as follows: having ≤25 dB HL in the frequency range of 0.125 to 12 kHz; having normal acoustic reflex threshold in the range of 0.5 to 4 kHz; having 6 dB and above the signal to noise ratio at 4 points in each octave in the range of 1 to 8 kHz according to DPOAE test[19]; having 21 and above score on Montreal Cognitive Assessment scale[20]; and having 96% and above dichotic digit test score.[21]

Ethical Statement

Ethical approval for this study (Ethical Committee N° GO 16/510) was provided by the Ethical Committee NAC of Hacettepe University, Ankara, on 26 June 2016.

Assessment of noise exposure

A 1-minute screening questionnaire consisting of three questions was used to identify people at higher risk for noise-induced hearing loss. This survey questions how often respondents have been exposed to high-intensity explosion, workplace noise such as piercing, or loud music over the past year. The participants choosed one of the four answers (“None” (0), “Every few months” (1), “Once a month” (2), “Once a week” (3), and “Once a day” (4)) to each question. A survey score of 5 and above defines the high-risk noise exposure group, whereas 4 and below indicates the low-risk noise exposure group.[18]

Pure tone audiometry and immittance meter measurements

The pure tone air conduction thresholds in the normal range at 0.125 to 8 kHz octave frequencies (including the intemediate frequencies of 3 and 6 kHz) with Sennheiser TDH 49 P supra-aural headset, and bone conduction thresholds at 500 to 4000 Hz octave frequencies were measured through the use of standard audiometric procedures recommended by the British Society of Audiology (2011). High-frequency air conduction hearing thresholds were evaluated with Sennheiser HDA 200 circumaural headphones at the octave frequency range from 8 to 12 kHz.[22] About 226 Hz probe tone tympanometry was used to evaluate the middle ear function (GSI TympStar, Grason-Stadler, Minnesota, America). Acoustic reflex thresholds were obtained through the use of pure tone stimulus between 500 and 4000 Hz.

Auditory brainstem response measurements

When the participants were at rested position and natural sleep on a stretcher in the Faraday cage room, ABR measurements were conducted using the Vivasonic Integrity™ V500 (Vivasonic Inc., Toronto, Canada). After the electrode areas were cleaned; the active electrode (positive) was placed on the upper part of forehead (FZ), the ground electrode was placed on the lower forehead (Fpz), one of the references (negative) electrodes was on the left mastoid area (M1), and the other was placed on the right mastoid area (M2). The electrode-skin impedance was ensured to be below 3 kΩ. The stimulus was transmitted by using alternative polarity and a 30-Hz high pass filter. The acoustic stimuli were at the limited bandwidth of 1500 Hz and transmitted through ER-3A insert earphones. ABR was recorded by a transmitted stimulus with 55 dB nHL broadband noise, a rate of 9.1 Hz and had 100 μs click at 70, 80, 90, 99 nHL levels, to the opposite ear. In addition, 2048 sweeps at 70, 80 nHL levels and 1024 sweeps at 90, 99 nHL levels were repeated and averaged as much as the permitted artifact rejects value in the software.

Turkish matrix sentence test

The Turkish matrix sentence test is a new sentence recognition test that randomly selects 50 words from the most frequently used Turkish words and establishes syntactic fixed sentences and evaluates speech intelligibility under noise.[23]

The words in this test are arranged as noun–adjective–object–verb in accordance with the Turkish sentence structure and consist of randomly selected sentences of 50 words in the same sentence structure. Each sentence contains 10 words that are independent from each other in terms of meaning. There is no suffix in sentence structure, only past tense and third person singular suffixes are used in verbs. This structure eliminates the learning effect, as 100,000 different sentences with correct sentence structures can be obtained.

In our study, both adaptive and nonadaptive methods were used in noise and each was evaluated with a 20-sentence test procedure. When making an adaptive measurement in the matrix test, the noise level was kept constant at an average intensity of 65 dB Sound Pressure Level (SPL), which the patient can easily hear. At the end of the test, the adaptive procedure approached the patient’s threshold of understanding. The signal-to-noise ratio of the participants was determined as dB at 50% threshold.

Before the test, the individuals were informed about the test and were told to repeat whatever they heard. The Turkish matrix sentence test was conducted in the Industrial Acoustics Company quiet room by using Otometrics Madsen Astera-2 brand 2-channel audiometer device and Sennheiser HDA 200 circumaural headphones.[23]

Psychophysical tests

Amplitude modulation detection test

In the presence of contralateral narrowband noise, each participant was subjected to Amplitude modulation detection test (AMDT). A sound with the level of 75 dB SPL, which was produced through sinusoidally modulating of 5 kHz carrier frequency to 19 Hz, was used in this test. The carrier frequency of 5 kHz was choosed as it is the most affected area in those with noise exposure.[24],[25] During the experiments, the modulation depth was adaptively changed between 0 and 1, and the amplitude modulation difference thresholds of individuals were determined as dB in the formula: 20 log10 m.[26]

A stimulus with 40 dB narrowband noise level and 5 kHz central frequency was selected to reduce the contribution of high spontaneous rate AN fibers and to suppress the coding envelope connected to low spontaneous rate AN fibers.[27] The bandwidth of narrowband noise was set to one-third of an octave.

Parameter estimation by sequential testing method was utilized to perform the AMDT.[28] According to this method, three interval–three alternative selection method and P target value were determined as 0.75. Three sounds were embedded in narrowband noise and were transmitted to the participants. As one of these sounds was modulated to stimulation amplitude, the other two sounds were generated from nonmodulated sounds. The target tone’s position was given randomly in each trial; the AM depth of the target was initially set at 6 dB (50%). The test was adaptively carried on until the final modulation change size reached at 0.45 dB. The threshold value was determined by taking the average of the two last steps.[28] The modification of the stimulus used in the AMDT was created with the Matlab 2017.b program (The MathWorks, Inc., United State).

Loudness adaptation experiments

The bandwidth of the stimulus was calculated using the center frequency (F0), Q factor and cutoff frequencies (f1 and f2 = −3 dB). In our study, the geometric mean of f1 and f2 frequencies were taken to determine the center frequency, and the bandwidth was one-third octave for all stimuli. Immediately after delivery of the stimuli for training (involving two 500-millisecond bursts), the participants were asked the 185-second test stimulus generated for each experiment.

The amount of loudness adaptation was measured for a total of seven times: in the first 5 seconds after the onset of the stimulus, and every 30 seconds (5, 35, 65, 95, 125, 155, and 185 seconds).[29] The participants were asked to indicate the magnitude of the loudness adaptation value using a scale of 0 to 10 (0: not hear anything, 10: loud and clear hearing), whereas a visual signal (turning on and off the lamp of the silent cabin) was given by another audiologist to prevent confusion and direct them at the appropriate time intervals. All stimuli were randomly listened to reduce residual adaptation between experiments. The stimuli were listened to by participants with a mean interval of 5 minutes. The magnitude of adaptation was calculated using the following equation (equations 1 and 2) according to a scale between 0 and 10. In regards to this calculation, 0% adaptation indicates that there is no change in loudness and 100% means that the sound is inaudible. Equations that we used in calculating loudness adaptation are as follows:

where Lt indicates the loudness value at t time and L0 indicates the initial loudness value. Positive numbers indicate that the sound is loud over time.

The time period of the loudness adaptation can also be evaluated via an exponential equation:

where y(t) illustrates the amount of adaptation at time t and s is the asymptotic saturation value that represents the amount of adaptation in an infinite length of the stimulus. τ is the constant representing the amount of time that reaches 63% of the asymptotic saturation value. The stability value is between −100% and 0%, the time constant is between 0 and 180 seconds.

For t = 0, y = 0.

For t → ∞, y(∞) = s

Our study was composed of three experiment pairs (experiment 1–2, experiment 3–4, and experiment 5–6) in which the transmitted stimuli were different. The effect of modulation frequency change on adaptation was evaluated in experiment 1–2 pairs (experiment 1: without noise; experiment 2: with contralateral noise); the effect of carrier frequency change on the loudness adaptation was measured in experiment 3–4 pairs (experiment 3: without noise; experiment 4: with contralateral noise); the effect of modulation amplitude change on loudness adaptability was evaluated in experiment 5–6 pairs (experiment 5: without noise; experiment 6: with contralateral noise).

The effect of modulation frequency change on the loudness adaptation (experiment 1–2 pair): In experiment 1, to determine the amount of loudness adaptation at 12 kHz and 15 dB sensation level (SL) level, five stimuli were modulated at five different modulations frequency (0, 4, 20, 50, and 100 Hz) and 100% modulation depth was transmitted sinusoidally and in a random sequence.[30] In experiment 2, the procedure used in experiment 1 was repeated at 40 dB SL, with a contralateral “Gaussian” noise with the center frequency of 12 kHz and one-third octave bandwidth. The stimulus generated through the change in the modulation frequency was given through the right ear, and the noise from the contralateral side (usually left ear) was presented for 180 seconds, 5 seconds after the stimulus started.

Carrier frequency effect of the loudness adaptation (experiment 3–4 pair): In experiment 3, each of the stimuli with 1, 6, and 12 kHz carrier frequencies and modulated at 100% modulation depth were listened to separately at 15 dB SL intensity level. In addition to experiment 3, three separate stimuli were listened to by participants again in the presence of three contralateral noises with the same carrier frequency in experiment 4.

The effect of modulation depth change on adaptability (experiment 5–6 pair): In experiment 5; stimuli with 12 kHz carrier frequency, 4 Hz modulation frequency, and 4 different depths of modulation (0%, 25%, 50%, and 100%) were presented to the participants at 15 dB SL level. In experiment 6, the stimuli used in experiment 5 were masked with narrow-band noise with the same center frequency, and the change in adaptation was investigated.

Statistical analysis

Data analysis was performed using Statistical Package for Social Sciences (SPSS) v. 23.0 (SPSS Inc., Chicago, Illinois, USA). Loudness values were calculated with Matlab 2017.b software; curve fitting exponential function was used for fitted parameters (time constant, saturation, and R2 values), and the curve obtained from this function was used to analyze the change in the loudness adaptability during the decrease or increase in the loudness with respect to time. Gender and age differences between low- and high-risk groups were evaluated using the Chi-squared test. Adaptive and nonadaptive Turkish matrix sentence test means, ABR, and AMDT score between groups were compared using Student t test. The paired-samples t test was used for intragroup comparison. For all experiments, the nonparametric Friedman test was used to compare the loudness adaptation values obtained from individuals against all variable signals. When necessary, pairwise comparisons were made using the Wilcoxon test and Bonferroni correction was made for comparisons. In all experiments, loudness adaptation, saturation, and time constant values of unmodulated and modulated sounds between low- and high-risk groups were compared using the Mann–Whitney U test. Cohen d was used as a measure of effect size, in which values of 0.20, 0.50, and 0.80 designate small, medium, and large effect sizes, respectively.

  Results Top

Auditory brainstem response measurement results

The amplitudes of ABR waves reduced as the stimuli levels increased in the high-risk group. The amplitude of the wave I of ABR at 99 dB nHL was found to be about 20% lower than at 90 dB nHL and about 50% lower than at 80 and 70 dB nHL intensity levels [Figure 1]. The amplitude of wave I of ABR increased in proportion to the increased intensity (especially at 80–90–99 dB nHL intensity levels) in the low-risk group [Figure 1].
Figure 1 Average the amplitude of waves I, III, and V of auditory brainstem response (ABR) represented by microvolts at different intensity levels for low- and high-risk groups. The box plot on the right indicates the high-risk group, whereas the box plot on the left shows the low-risk group. The symbols used in both the groups are similar. The amplitude of wave I of ABR with a white error bar at all intensity levels and the amplitude of wave III are shown with stripe patterned error bar, and the amplitude of wave V with white and black dot patterned error bar.

Click here to view

When the wave I amplitudes in both the groups were cross-compared, the difference between the groups was statistically significant (P < 0.05). A statistically significant difference was observed between the wave I amplitudes at 80 dB nHL (P = 0.21) and 70 dB nHL intensity levels (P = 0.001) of two groups. However, there was an inverse relationship between the mean values to the intensity change in the high- and the low-risk groups [Table 1]. In the high-risk group, a statistically higher difference was found between V/I amplitude ratios at 70 dB nHL intensity level (P < 0.05). However, wave V/I amplitude ratio in the high-risk group at 99 dB nHL was found higher than the average value of the amplitude ratios obtained from the other levels (P < 0.05). At the same time, the V/I amplitude ratio obtained in the high-risk group was found approximately 60% higher than the low-risk group [Table 1].
Table 1 Auditory brainstem responses of low- and high-risk groups

Click here to view

Turkish matrix test results

A statistical difference was found only in the right ear in terms of the means of the adaptive Turkish matrix sentence test between the groups (P < 0.05). There was no statistical difference between the groups in terms of nonadaptive Turkish matrix test averages (P > 0.05).

Amplitude modulation detection test results

Amplitude modulation detection performance was lower in the low-risk group than the high-risk group, and this difference was statistically significant (P < 0.05).

Effect of modulation frequency change on loudness adaptation (experiments 1 and 2)

In the first 120 seconds after the transmission of the stimulus, which was not modulated in experiment 1, 100% loudness adaptation was observed. A statistically significant difference between the nonmodulated stimulus (0 Hz) and the modulated stimuli [4 Hz (Z = −3.233, P = 0.001) and 20 Hz (Z = −3.126, P = 0.002)] was observed 180 seconds after the stimuli transmitted to the participants in the high-risk group. However, this statistically significant difference was not observed in the low-risk group. In the low-risk group, the highest increase was found in the stimulus modulated to 50 Hz, and the increase is approximately 32% [Figure 2].
Figure 2 Loudness adaptation for high- and low-risk groups of the 12 kHz stimulus modulated in amplitude at 100% depth as a function of time in experiments 1 and 2. Each symbol indicates different modulation frequencies in each figure. Error bars represent 95% of confidence intervals.

Click here to view

In experiment 1, a statistically significant relationship was found between the high- and low-risk groups at the s (saturation constant) and t (time constant) values of the nonmodulated stimulus and the modulated stimulus to 20 Hz (P < 0.05). When it was examined explicitly to the time constant, it was concluded that the loudness adaptation in the nonmodulated stimulus developed relatively earlier than in the modulated stimuli in both the groups. A statistically significant difference was found between the groups in the stimuli with 0, 4, and 50 Hz modulation 120 seconds after stimulation (P < 0.05). The time–loudness adaptation relation for experiments 1 and 2 is given in [Figure 2].

When experiment 2 is analyzed, a statistically significant difference was found between the nonmodulated stimulus (0 Hz) and the modulated stimulus to 4 Hz in the high-risk group 180 seconds after transmission (Z = −1.654, P = 0.001). In the low-risk group, no statistically significant difference was found in loudness adaptation values between the nonmodulated stimulus and the stimuli which were modulated in found different frequencies, 180 seconds after transmission. More adaptation was observed in the high-risk group at 0, 4, 20, and 50 Hz modulation frequencies 180 seconds after the stimulation was transmitted and it was statistically important (P = 0.01).

In experiment 1, when the time (τ) and saturation constants (s) of the variables were compared, there was no statistically significant difference between the high- and low-risk groups (P = 0.0025). According to equation 2, although there was excellent compatibility between the adaptation values in all stimuli in the high-risk group, the highest compliance level was observed in the stimulus modulated to 100 Hz (adjust R2 > 0.9). In the low-risk group, goodness-of-fit value was low in the nonmodulated stimuli and the modulated stimuli at 4 Hz (adjust R2 < 0.9) [Table 2].
Table 2 Fitted parameters and goodness-of-fit according to equation 2 of low- and high-risk groups for experiments 1 and 2

Click here to view

In experiment 2, in the high-risk group, the compliance between good compatibility adaptation values was found as good according to equation 2 (adjust R2 > ∼0. 9). At 4 and 20 Hz modulation frequencies, time and asymptotic saturation constants were less than other frequency-modulated (0, 50, and 100 Hz) stimuli. In the low-risk group, good compatibility adaptation values were low in all nonmodulated and the modulated stimuli according to equation 2 (adjust R2 < 0.9). In all cases, the time and asymptotic saturation constants obtained were in similar outcome values. In experiment 2, there was a statistically significant difference between the time constant “tau (τ)” and saturation values in both the groups (P < 0.05) [Table 2].

Effect of carrier frequency change on loudness adaptation (experiments 3 and 4)

For the high-risk group, no statistically significant correlation was found between the loudness adaptation values 180 seconds after the stimulus was given at the carrier frequencies (1, 6, and 12 kHz; P > 0.05). Depending on carrier frequency change, the mean and median values of the adaptation obtained were in similar to each other. In the low-risk group, there was no statistically significant difference between the loudness adaptation values of the stimuli with three different carrier frequencies over time, 180 seconds after the stimulus was listened (P > 0.05).

In experiment 3, there was a more adaptation obtained from the stimuli with 1 kHz (Z = −3802, P = 0.001), 6 kHz (Z = −3.327, P = 0.001), and 12 kHz (Z = −4.420, P = 0.001) carrier frequencies of the high-risk group than the low-risk group 180 seconds after the stimuli. The time–adaptation relation of the experiments 3 and 4 is given in [Figure 3].
Figure 3 Loudness adaptation to 1, 6, and 12 kHz tone carrier frequency modulated at 100% depth as a function of time in experiment 3 without mask and experiment 4 with the mask in experiment 4. Each symbol indicates different modulation frequencies in each figure. Error bars represent 95% of confidence intervals.

Click here to view

In experiment 4, there was no statistically significant difference in the loudness adaptation values of both the groups with the stimuli obtained from three different carrier frequencies 180 seconds after the stimuli (P > 0.05). In the high-risk group, the mean and median values of the loudness adaptation were in similar to each other. In contrast, the highest increase in the loudness level was observed at 1 kHz in the low-risk group.

For the high- and low-risk groups in experiments 3 and 4, no statistically significant difference was found between carrier frequencies 180 seconds after the stimulus (P > 0.05). Goodness-of-fit value and fitted parameters for equation 2 of experiments 3 and 4 are summarized in [Table 3].
Table 3 Fitted parameters and goodness-of-fit according to equation 2 of low- and high-risk groups for experiments 3 and 4

Click here to view

There was no difference between the saturation and time constant of the high- and low-risk groups for experiment 3–4 pairs (P > 0.05). In experiment 3–4 pairs, according to Equation 2, there was a moderate agreement between the good fit adaptation values (adjust R2 > 0.9) with the stimuli modulated in three different carrier frequencies for the high-risk group; however, the same amount was found to be low for the low-risk group (adjust R2 < 0.9). In experiment 4, the time and asymptotic saturation constants were similar in the high-risk group; the time and asymptotic saturation constant values of the stimulus with 1 kHz carrier frequency were higher than the time and asymptotic saturation constant values of the other carrier frequencies in the low-risk group [Table 3].

Effect of modulation depth change on loudness adaptation (experiments 5 and 6)

In experiment 5, there was a significant relationship between the adaptation values obtained from the modulated stimuli [modulation depth of 0% and 50% (Z = −3.548, P = 0.001); 0% and 100% (Z = −3.411, P = 0.001); 25% and 100% (Z = −4.079, P = 0.001); and 25% and 50% (Z = −2.737, P = 0.006)] of the high-risk group. The modulation depth increased as the loudness adaptation decreased in the high-risk group. There was a statistically significant difference between modulation depths 0% and 25% (Z = −3.318, P = 0.001), 25% and 50% (Z = −4.025, P = 0.001), and 0% and 50% (Z = −3.304, P = 0.001) in the low-risk group. When the time and saturation constants of the high- and low-risk groups were compared in experiment 5, a statistically significant difference was found between the groups only in the case with 25% modulation depth in terms of saturation constant (Z = −3.190, P = 0.001). The time–loudness adaptation relation of experiments 5 and 6 and is given in [Figure 4].
Figure 4 Loudness adaptation to a 12-kHz tone, amplitude modulated at three different modulation depths as a function of time in experiment 5 without mask and experiment 6 with the mask. Each symbol indicates different modulation frequencies in each figure. Error bars represent 95% of confidence intervals.

Click here to view

In experiment 6, there were a significant differences in terms of loudness adaptation between modulation depths [0% and 50% (Z = −3.548, P = 0.000), 0% and 100% (Z = −3.411, P = 0.001), 25% and 100% (Z = −4.079, P = 0.001), and 25% and 50% (Z = −2.737, P = 0.006)] in the high-risk group (adaptation increased as modulation depth increased). With respect to the initial state, the maximum adaptation increase was at modulation depth of 50% in the background noise. The magnitude of loudness adaptation was found statistically different between 25% and 100% modulation depth under background noise in the low-risk group [Figure 4].

In the comparison of the high- and low-risk groups in experiments 5 and 6, no statistically significant difference was found between the different modulation depths for both the groups (P > 0.05).

In experiment 5, no statistically significant difference was found in the binary comparison between asymptotic saturation and time constants for the high- and low-risk groups (P > 0.05). For the high-risk group, to equation 2, a moderate agreement was found between the goodness-of-fit values in the stimuli obtained depending on the modulation depth (0, 25, 50, and 100%) (adjust R2 < 0.9); however, the agreement was low for the low-risk group (adjust R2 < 0.9). In all participants but for the case of 100% modulation depth, the time and asymptotic saturation constants were in similar. In experiment 5, there was a statistically more adaptation at the modulation depths of 0% (Z = −3.027, P = 0.002) and 50% (Z = −2.657, P = 0.008) in the high-risk group with respect to the low-risk group [Table 4].
Table 4 Fitted parameters and goodness-of-fit according to equation 2 of low- and high-risk groups for experiments 5 and 6

Click here to view

In experiment 6, for the low-risk group, there was a significant difference between the 0% and 100% modulation depth (Z = −3.472, P = 0.01) in terms of asymptotic saturation values. To equation 2, goodness-of-fit value for the stimuli obtained depending on the modulation depth was very good (adjust R2 > 0.9) for the high-risk group, and goodness-of-fit value e was low (adjust R2 < 0.9) for the low-risk group. The time and asymptotic saturation constants were similar between 0% and 25% for the high-risk group, and all-time and asymptotic saturation constants were similar for the low-risk group [Table 4].

  Discussion Top

There are still many unacknowledged points about HHL which impairs human hearing. Findings such as being seen in individuals with normal hearing who have been exposed to noise and deterioration in speech perception and temporal processing skills are the criteria which help for the diagnosis of HHL.[31] Huss and Moore investigated the loudness adaptability of individuals with known inner hair cell loss. They determined that the adaptation was higher near the dead area of the cochlea.[32] In our study, we aimed to identify the change in amplitude modulation in individuals with noise exposure and to generate a behaviorally different test battery to evaluate HHL.

Auditory brainstem responses

It was reported that AN function, which is sensitive to noise exposure effects, reflects the wave I of ABR in the rodent model.[6] In these models, when the amplitude of wave I decrease at moderate and high-intensity levels in relation to low spontaneous AN fiber loss, it does not demonstrate any change at low-intensity levels.[6],[33] Even in cases in which increase in pure-tone hearing thresholds as a result of noise exposure is not present, a decrease in high-level click ABR I wave amplitude such as 90 and 100 dB nHL has been demonstrated to be relatively decreased.[5],[6],[18],[25],[34] In a study on animals, it was indicated that after 2 hours of exposure to 100 dB SPL (8–16 kHz) noise level, absolute hearing sensitivity was not affected, but the permanent decrease in ABR I wave amplitude reflected the reduction in AN activities and was reported in moderate/high-intensity stimulus responses.[4] In our study, the low amplitude of the wave I was compared to other intensity levels support the possibility of peripheral AN damage in individuals in the high-risk group. The relationship between intensity levels and amplitude values of wave I was determined to be different in the high- and low-risk groups (increase in amplitude as intensity increases in the low-risk group and decrease in amplitude as intensity increases in the high-risk group). Although the nature of the changes due to noise exposure is not fully understood, the decrease in amplitude of wave I in the high-risk group in our measurements is significant in terms of being similar to other studies that investigate low spontaneous fiber loss.[4],[6],[33],[35]

Waves I and V amplitude ratios are used to evaluate the central auditory gain.[36],[37] In the literature, in studies on individuals with cochlear synaptopathy, it has been reported that wave V amplitude did not decrease.[33] Besides this, wave I amplitude decreased; the difference between V/I amplitude ratios may be useful in revealing HHL since V wave amplitude does not change.[37] According to the noise questionnaire score, in our study, it was found that there was a higher difference in the V/I amplitude ratio in the high-risk group. Therefore, this finding of our research supports that there may be an interruption in auditory activity in the central auditory pathways after exposure to noise. These results support that noise exposure does not only affect individuals’ peripheral hearing, but it may also cause disturbances in central auditory pathways. We think that not only wave I amplitude value but also ABR V/I wave amplitude ratio can be utilized to diagnose individuals with HHL.

Adaptation effect of modulation frequency change

The effect of change in amplitude on the loudness perception of fixed alert and whether this change is to be regarded as a separate phenomenon has always been a subject of research from past to present.[38] High frequency and amplitude modulated sounds were used in our study, since the mechanism of the loudness adaptation which includes synchronized cortical activities can be explained through experiments on loudness adaptation in which modulation affects high frequencies.[39] In a study which was conducted on individuals with normal hearing, 180 seconds after a stimulus with a carrier frequency of 12 kHz which was modulated at different modulation frequencies at 15 dB SL suprathreshold level was transmitted to the individual, a decrease in loudness was observed and it was found that the adaptation decreased with increasing modulation frequency.[39] The decrease in the mean sound level accompanying the increase of the modulation frequency was found in our study, and this result is consistent with the study of Wynne et al.[39] The finding that there is a significant difference between the nonmodulated stimuli and the stimuli which was modulated to 4 and 20 Hz 180 seconds after the warning was transmitted suggests that neural synchronization in individuals who have been exposed to noise may be changed with the use of the stimuli which are modulated to this frequency. It has been reported that the stimuli which are modulated to frequencies between 20 and 50 Hz are generally associated with speech impediments[40] and that modulation frequencies of 40 Hz and above are the frequency region in which the broadest neural responses are produced in the auditory cortex.[41] In our study group, the adaptation decreased especially in the 4 and 20 Hz region with the increase of the modulation frequency and the fact that the nonmodulated stimuli showed more adaptation in the participants in the control group compared to the stimuli which are modulated to 50 and 100 Hz. This finding also points out the possibility that loudness adaptation might not only have peripheral roots but also is related to the cortical regions. The central mechanism of loudness adaptation continues to be a research topic of interest. We found a statistically significant relationship between nonmodulated stimuli and stimuli modulated to different frequencies (4 and 20 Hz), which has been regarded as in accordance with studies that illustrate although the peak in the stimulus that is modulated to high modulation amplitude at low modulation frequencies is clearly noticeable, these stimuli are not perceived as intermittent stimuli.[36],[39],[42]

It is believed that in low-spontaneous fiber loss for listening in noisy environments, the auditory sensitivity of important sounds in temporal modulation is impaired.[43] According to our study, the fact that the modulation frequency change in the study group in the presence of background noise is much higher than the change in the loudness adaptation which was obtained without noise. This finding suggests that the noise exposure both decreases the auditory perception in the noisy environment and the neural fatigue against the stimuli other than speech is relatively developed in these individuals with respect to the low-risk group.

The fact that there was a statistically significant difference between the individuals in the high- and low-risk group under contralateral noise in our study supports that there is lower spontaneous fiber loss, rather than high spontaneous auditory fibers, after noise exposure. These findings also point out the fact that background noise should be used in the loudness adaptation change and diagnosis of hearing impairment, especially with individuals who are considered to have HHL after noise exposure. It is speculated noise could disrupt loudness adaptation since the average time which is required for a complete adaptation (300–400 seconds) in the group with noise exposure was shorter than the time of complete adaptation (350–560 seconds) without noise.

Strength adaptation effect of carrier frequency change

In our study, the finding that there is no difference between the stimuli at low modulation rate in the high-risk group at different carrier frequencies and in the absence of contralateral noise showed that carrier frequency change had no effect on the adaptation of the intensity and this finding was different in the study by Wynne et al.[42] It is thought that this difference may be related to hearing distinction performance (based on matrix sentence test scores), which is thought to be impaired after noise exposure. The fact that the adaptation values (especially at 1 and 6 kHz carrier frequencies) which were obtained from the individuals in the high group showed more adaptation than the control group 180 seconds after the stimuli were transmitted and it was in an accordance with other studies which demonstrate that the threshold for noticing amplitude modulation may be impaired.[39],[42],[44] Although there was no significant difference between the saturation values which were obtained in the high group, it was found that there was more and faster adaptation at high carrier frequencies compared to low carrier frequencies (approximately 90 seconds). In our study, the low compatibility values of the individuals in the low-risk group in the adaptability assessment suggest that many of the participants might not be able to adapt to the carrier frequency change or show little adaptation.

It was found that in the presence of background noise, there was a higher loudness perception than the level before the noise transmission through the opposite ear and the adaptation value was higher than the initial value which was obtained before the noise was transmitted. This finding supports the theory which was stated by Oberfeld that the intensity of the weak sound can be perceived at a different loudness level in the presence of a different noise and the perceptual loudness of this noise decreases approximately 100 milliseconds after the mask is provided. In the loudness adaptation findings which were obtained from the stimuli at different carrier frequencies in the presence of background noise, the low carrier frequency stimuli demonstrated more adaptation than the high carrier frequency stimuli. This finding shows fundamental similarities with the findings which are obtained in studies on ribbon synapse disorder and auditory neuropathy.[42] We believe that the change in the loudness which is due to the change in the carrier frequency in the presence of background noise may cause impairment in the temporal processing ability of individuals with HHL. We think that this situation may cause impaired speech perception in noisy environments, as all of the information in speech sounds are carried over time with changes in signal amplitude and frequency transitions[45] and these changes can be determined by adaptation parameter.[46] The fact that the maximum change in the adaptation of the individuals in the presence of background noise is at 1 kHz is in accordance with the information that the binaural beat detection is more at low frequencies.[39] The fact that loudness adaptation in the presence of background noise occurs in a shorter span than the loudness adaptation without the noise is similar to the previous studies and supports that the adaptation is more affected in the individuals with HHL in comparison to individuals with normal hearing. However, these experiments fail to fully understand whether noise affects only the peripheral or only the central region.

Effect of modulation depth change on loudness adaptation

It is indicated that amplitude modulation gives important information in speech and other sounds.[47] The change in amplitude modulation generally may impair speech intelligibility in noisy environments, in cases in which hearing loss does not accompany.[48],[49] Generally, the amplitude modulation model in speech has amplitude modulation depth above the threshold at which amplitude modulation is noticed. The processing of information which is carried by amplitude modulation depends on the ability to distinguish variances in amplitude modulation depth.[50] In a study on amplitude modulation, it was stated that in the stimulus with 50% modulation depth, a decrease in loudness modulation is greater than the modulation with the stimulus with 100% modulation depth.[39] In this part of our study, the finding that loudness adaptation which is modulated to the amplitude decreases as the modulation depth increases in individuals in the high-risk group is in accordance with the study of Wynne et al..[42] Together with the increase in the modulation depth, the stimuli are perceived as more intermittent rather than fixed, which may result in a lesser amount of development in loudness adaptation. Especially, the fact that adaptation started earlier than the low-risk group after the stimuli were transmitted is similar to the findings obtained from the study on the ribbon synapse.[42] According to the results which were obtained in the comparison with normal individuals, we thought that adaptation might develop when the modulation depth of the stimulus changes in individuals with HHL.

  Conclusion Top

It was assumed that in individuals with HHL, who have a history of noise exposure, especially in the presence of environmental noise, there might be a change in amplitude modulation and distortion in auditory sensitivity over time. The use of stimuli which are modulated under contralateral noise in the differential diagnosis of HHL gives useful information in the evaluation of the hearing system. In our study, the adaptation duration of the high-risk group to stimuli in the presence of noise was shorter than the low-risk group. In many experiments we have done, most of the individuals in the low-risk group developed an adaptation approximately 120 seconds after the stimulation was transmitted, and depending on the stimuli changes, the adaptation occurred approximately 40 seconds after the stimulus transmitted in the high-risk group. When the literature was analyzed, no study of loudness adaptation with similar sample groups and stimuli with our study has been encountered. In light of the findings obtained in our study, we thought that loudness adaptation assessment could be used to identify individuals with suspected hearing loss after exposure to noise. At the same time, although we found a significant difference in the Turkish matrix sentence test in the right ear in our study, we do not think that this result showed a significant result among our study groups. However, we think that the use of noise discrimination tests in unusual cases such as auditory neuropathy or hidden hearing loss is important in evaluating the auditory function of individuals.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Concha-Barrientos M, Steenland K, Prüss-Üstün A, Campbell-Lendrum DH, Corvalán CF, Woodward A, World Health Organization. Occupational noise: Assessing the burden of disease from work-related hearing impairment at national and local levels. Geneva: World Health Organization; 2004.  Back to cited text no. 1
Themann CL, Masterson EA. Occupational noise exposure: a review of its effects, epidemiology, and impact with recommendations for reducing its burden. J Acoust Soc Am 2019;146:3879-905.  Back to cited text no. 2
WHO. 1.1 Billion People at Risk of Hearing Loss. Geneva: WHO; 2015.  Back to cited text no. 3
Prendergast G, Millman RE, Guest H et al. Effects of noise exposure on young adults with normal audiograms II: behavioral measures. Hear Res 2017;356:74-86.  Back to cited text no. 4
Kujawa SG, Liberman MC. Synaptopathy in the noise-exposed and aging cochlea: primary neural degeneration in acquired sensorineural hearing loss. Hear Res 2015;330:191-9.  Back to cited text no. 5
Kujawa SG, Liberman MC. Adding insult to injury: cochlear nerve degeneration after “temporary” noise-induced hearing loss. J Neurosci 2009;29:14077-85.  Back to cited text no. 6
Song Q, Shen P, Li X et al. Coding deficits in hidden hearing loss induced by noise: the nature and impacts. Sci Rep 2016;6:25200.  Back to cited text no. 7
Lobarinas E, Salvi R, Ding D. Insensitivity of the audiogram to carboplatin induced inner hair cell loss in chinchillas. Hear Res 2013;302:113-20.  Back to cited text no. 8
Lobarinas E, Salvi R, Ding D. Selective inner hair cell dysfunction in chinchillas impairs hearing-in-noise in the absence of outer hair cell loss. J Assoc Res Otolaryngol 2016;17:89-101.  Back to cited text no. 9
Qiu C, Salvi R, Ding D, Burkard R. Inner hair cell loss leads to enhanced response amplitudes in auditory cortex of unanesthetized chinchillas: evidence for increased system gain. Hear Res 2000;139:153-71.  Back to cited text no. 10
Furman AC, Kujawa SG, Liberman MC. Noise-induced cochlear neuropathy is selective for fibers with low spontaneous rates. J Neurophysiol 2013;110:577-86.  Back to cited text no. 11
Schaette R, McAlpine D. Tinnitus with a normal audiogram: physiological evidence for hidden hearing loss and computational model. J Neurosci 2011;31:13452-7.  Back to cited text no. 12
Lobarinas E, Spankovich C, Le Prell CG. Evidence of “hidden hearing loss” following noise exposures that produce robust TTS and ABR wave-I amplitude reductions. Hear Res 2017;349:155-63.  Back to cited text no. 13
Bramhall NF, Konrad-Martin D, McMillan GP, Griest SE. Auditory brainstem response altered in humans with noise exposure despite normal outer hair cell function. Ear Hear 2017;38:e1.  Back to cited text no. 14
Bakay WMH, Anderson LA, Garcia-Lazaro JA, McAlpine D, Schaette R. Hidden hearing loss selectively impairs neural adaptation to loud sound environments. Nat Commun 2018;9:4298.  Back to cited text no. 15
Javel E. Long‐term adaptation in cat auditory‐nerve fiber responses. J Acoust Soc Am 1996;99:1040-52.  Back to cited text no. 16
Hellman R, Miśkiewicz A, Scharf B. Loudness adaptation and excitation patterns: effects of frequency and level. J Acoust Soc Am 1997;101:2176-85.  Back to cited text no. 17
Johnson TA, Cooper S, Stamper GC, Chertoff M. Noise exposure questionnaire: a tool for quantifying annual noise exposure. J Am Acad Audiol 2017;28:14-35.  Back to cited text no. 18
Ramos JA, Kristensen S, Beck DL. An overview of OAEs and normative data for DPOAEs. Hear Rev 2013;20:30-3.  Back to cited text no. 19
Avcı Gülen HK, Yıldırım A, Emre U, Karagöz Y, Akbaş EA. Hafif Bilişsel Bozuklukta MOBİD Ölçek Puanları ile Bilişsel Düzey Korelasyonunun Değerlendirilmesi. Istanbul Med J 2017;18.  Back to cited text no. 20
Turkyilmaz M, Budak B (Eds). Normalisation of the “time compressed” and “dichotic digit” speech tests. Int J Psychophysiol 2006 Elsevier Science BV PO Box 211, 1000 AE Amsterdam, The Netherlands.  Back to cited text no. 21
Audiology B. Pure Tone Air and Bone Conduction Threshold Audiometry with and Without Masking and Determination of Uncomfortable Loudness Levels. UK: British Society of Audiology Reading; 2004.  Back to cited text no. 22
Zokoll MA, Fidan D, Türkyılmaz D et al. Development and evaluation of the Turkish matrix sentence test. Int J Audiol 2015;54(Suppl 2):51-61.  Back to cited text no. 23
Paul BT, Bruce IC, Bosnyak DJ, Thompson DC, Roberts LE. Modulation of electrocortical brain activity by attention in individuals with and without tinnitus. Neural Plast 2014;2014:127824.  Back to cited text no. 24
Roberts LE, Bosnyak DJ, Bruce IC, Gander PE, Paul BT. Evidence for differential modulation of primary and nonprimary auditory cortex by forward masking in tinnitus. Hear Res 2015;327:9-27.  Back to cited text no. 25
Bharadwaj HM, Verhulst S, Shaheen L, Liberman MC, Shinn-Cunningham BG. Cochlear neuropathy and the coding of supra-threshold sound. Front Syst Neurosci 2014;8:26.  Back to cited text no. 26
Zilany MS, Bruce IC, Carney LH. Updated parameters and expanded simulation options for a model of the auditory periphery. J Acoust Soc Am 2014;135:283-6.  Back to cited text no. 27
Taylor M, Creelman CD. PEST: Efficient estimates on probability functions. J Acoust Soc Am 1967;41:782-7.  Back to cited text no. 28
Scharf B, Botte M-C., Canévet G. Loudness Adaptation Induced Interaurally and Monaurally. Hearing—Physiological Bases and Psychophysics. Springer; 1983 p. 257-63.  Back to cited text no. 29
Canévet G, Scharf B. The loudness of sounds that increase and decrease continuously in level. J Acoust Soc Am 1990;88:2136-42.  Back to cited text no. 30
Kumar UA, Ameenudin S, Sangamanatha A. Temporal and speech processing skills in normal hearing individuals exposed to occupational noise. Noise Health 2012;14:100.  Back to cited text no. 31
  [Full text]  
Huss M, Moore BC. Tone decay for hearing-impaired listeners with and without dead regions in the cochlea. J Acoust Soc Am 2003;114:3283-94.  Back to cited text no. 32
Sergeyenko Y, Lall K, Liberman MC, Kujawa SG. Age-related cochlear synaptopathy: an early-onset contributor to auditory functional decline. J Neurosci 2013;33:13686-94.  Back to cited text no. 33
Ruggles D, Bharadwaj H, Shinn-Cunningham BG. Normal hearing is not enough to guarantee robust encoding of suprathreshold features important in everyday communication. Proc Nat Acad Sci U S A 2011;108:15516-21.  Back to cited text no. 34
Prendergast G, Guest H, Munro KJ et al. Effects of noise exposure on young adults with normal audiograms I: electrophysiology. Hear Res 2017;344:68-81.  Back to cited text no. 35
Jerger J. Bekesy audiometry in analysis of auditory disorders. J Speech Hear Res 1960;3:275-87.  Back to cited text no. 36
Holt LL, Lotto AJ. Behavioral examinations of the level of auditory processing of speech context effects. Hear Res 2002;167:156-69.  Back to cited text no. 37
Steinschneider M, Reser DH, Fishman YI, Schroeder CE, Arezzo JC. Click train encoding in primary auditory cortex of the awake monkey: evidence for two mechanisms subserving pitch perception. J Acoust Soc Am 1998;104:2935-55.  Back to cited text no. 38
Wynne DP, George SE, Zeng F-G. Amplitude modulation reduces loudness adaptation to high-frequency tones. J Acoust Soc Am 2015;138:279-83.  Back to cited text no. 39
McKay CM, Henshall KR. Amplitude modulation and loudness in cochlear implantees. J Assoc Res Otolaryngol 2010;11:101-11.  Back to cited text no. 40
Picton TW, John MS, Dimitrijevic A, Purcell D. Human auditory steady-state responses: respuestas auditivas de estado estable en humanos. Int J Audiol 2003;42:177-219.  Back to cited text no. 41
Wynne DP, Zeng F-G., Bhatt S, Michalewski HJ, Dimitrijevic A, Starr A. Loudness adaptation accompanying ribbon synapse and auditory nerve disorders. Brain 2013;136:1626-38.  Back to cited text no. 42
Bharadwaj HM, Shinn-Cunningham BG. Rapid acquisition of auditory subcortical steady state responses using multichannel recordings. Clin Neurophysiol 2014;125:1878-88.  Back to cited text no. 43
Wynne DP. Measuring and Modeling Loudness Adaptation in Normal and Impaired Hearing. Irvine: University of California 2013.  Back to cited text no. 44
Licklider JC, Miller GA. The perception of speech. 1951.  Back to cited text no. 45
Kay R, Matthews D. On the existence in human auditory pathways of channels selectively tuned to the modulation present in frequency‐modulated tones. J Physiol 1972;225:657-77.  Back to cited text no. 46
Moore BC. Auditory Processing of Temporal Fine Structure: Effects of Age and Hearing Loss. World Scientific;2014.  Back to cited text no. 47
Moore BC. Speech processing for the hearing-impaired: successes, failures, and implications for speech mechanisms. Speech Commun 2003;41:81-91.  Back to cited text no. 48
Plomp R. Auditory handicap of hearing impairment and the limited benefit of hearing aids. J Acoust Soc Am 1978;63:533-49.  Back to cited text no. 49
Moore BC, Wojtczak M, Vickers DA. Effect of loudness recruitment on the perception of amplitude modulation. J Acoust Soc Am 1996;100:481-9.  Back to cited text no. 50

Correspondence Address:
Bünyamin Cildir
Language and Speech Therapy Department Health Sciences Faculty, Ankara Yildirim Beyazit Üniversity, Ankara
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/nah.NAH_67_20

Rights and Permissions


  [Figure 1], [Figure 2], [Figure 3], [Figure 4]

  [Table 1], [Table 2], [Table 3], [Table 4]