This study employed descriptive and repeated measures design. Fifteen female and 14 male participants with age ranging from 18 to 28 years (mean of 23.4±2.65 years), right-handed, with normal hearing in both ears (thresholds ≤15dB HL from 500 to 4000Hz) and without any record of head injuries and neurological disorders or any active ear pathology took part in the study. All of them agreed to voluntary participation by signing the written informed consent. Prior to testing, an ethical approval from the respective institution was obtained, in accordance with the Declaration of Helsinki (USM/PPP/JEPeM [245.3(5)]).

To ensure normal hearing status, basic audiologic evaluations including otoscopy, admittance audiometry (Model AZ26, by Interacoustic) and pure tone audiometry (Model AC 40 two-channel audiometer) were performed on all participants in a sound-treated room within the Audiology Clinic, University Hospital.

ASSR thresholds were then recorded with a two-channel Eclipse system (Interacoustic Corporation, Denmark) and narrow-band CE-chirp® signals centered at 500, 1000, 2000, and 4000Hz were used as the test stimuli. The chirp stimuli were designed to compensate for the cochlear delay and to produce a bigger response.10 The amplitude and frequency modulation depths were 100% and 20% (±10%), respectively. For each test stimulus, the modulation frequencies were 40 and 90Hz. The stimulus level was calibrated in dB nHL. The common mode rejection (CMR) ratio of pre-amplifier was more than 115dB at any frequency. The number of epochs and time analysis were 16 and 1s, respectively. The recorded responses were amplified 100,000 times and filtered using a band-pass of 0.1–100Hz (12dB/octave).

Four scalp electrodes were placed on the participant's head: non-inverting on the forehead, inverting on each mastoid, and ground on the cheek. For both ipsilateral and contralateral recordings, the insert earphone was placed only in the right ear (i.e., ASSRs recorded in the right and left channels represented ipsilateral and contralateral conditions, respectively). The impedance of electrodes was maintained to be less than 3kΩ throughout the measurements.

After giving the proper instructions to the participants, the ASSR testing began. While lying comfortably on a test bed, the stimuli were presented to the right ear of the participants using the multiple auditory steady state response (MASTER) technique through the insert earphone. This technique offers a time-effective way for stimulus presentation, as four frequencies can be tested simultaneously. The measurement started with 40Hz MF and was followed by 90Hz MF recordings, as the participants were typically awake at the beginning of the testing. Nevertheless, during the 40Hz ASSR measurements, the participants’ state was closely monitored to ensure adequate wakefulness.

For detecting the response in a quick and accurate manner, the ASSR Eclipse system utilized the ‘Full Spectrum Detection Engine’ method. With this method, amplitude and phase coherence components were combined, and responses from higher harmonics were also included in the detection algorithm. In comparison to the use of information only from the first harmonic, significantly higher response detection rates and shorter detection times were observed when all available responses (amplitude and phase) from first and higher harmonics were utilized.11

With the ASSR Eclipse device, at a particular intensity level, the response is considered present if it reaches an amplitude level that is within 95% confidence within the default time of 6min. In the current study, the ASSR measurement started at 60dB nHL of intensity. If the response was clearly detected (reached 95% confidence) sooner than the default time, the trial was stopped and the stimulus intensity was decreased by 10dB. If the confidence of response was less than 50% during the first 3min, the trial was stopped and the test was repeated at a similar level. If unclear response was still observed (<50% confidence), the trial was stopped and the stimulus intensity was increased by 5dB. The measurements continued until ASSR threshold was obtained. The ASSR threshold was defined as the lowest intensity level that elicited response with 95% confidence in 6min. At the threshold level, the measurements were repeated twice to confirm the test reproducibility. The ASSR threshold was obtained at each test frequency at different MFs and for both recording conditions. To avoid fatigue during the recording, 10min of break was given to the participants between each trial.

For data analysis, both descriptive and inferential statistics were used. Mean and standard deviation (SD) were expressed as applicable. All the numerical data were found to be normally distributed as shown by Kolmogorov–Smirnov test (p>0.05). For each test frequency, two-way repeated measures ANOVA (with mode of recording and MF as the factors) was performed to compare mean ASSR thresholds between ipsilateral and contralateral recordings, as well as between 40Hz and 90Hz MFs. To analyze the simple main effects (in the conditions where significant interaction effects between factors were observed in the two-way ANOVA analysis), one-way repeated measures ANOVA was conducted to compare mean ASSR thresholds among the following conditions: ipsi-40Hz, contra-40Hz, ipsi-90Hz, and contra-90Hz. Prior to this, the Mauchly test was carried out to test the assumption of sphericity. For pairwise comparisons, Bonferroni correction was used. The statistical significance level was set at p<0.05. All data were analyzed using the SPSS software v. 20 (SPSS Inc., Chicago, IL, United States).

Recall that the present study aimed to compare the ASSR thresholds recorded at high and low modulation frequencies, as well as between ipsilateral and contralateral recordings in healthy adults. At the 500 and 1000Hz test frequencies, no significant differences in ASSR threshold were found between ipsi-40Hz and ipsi-90Hz conditions. However, contra-90Hz condition produced significantly higher ASSR thresholds than that of contra-40Hz condition. The weak efficiency of ASSR at low frequencies and at high MF found in the present study is in line with previous studies.12–14 The possible reason is that the ASSR at high MF is generated predominantly by the auditory brainstem. Consequently, it shows almost similar performance with ABR in the threshold testing.15,16 Elevated ABR thresholds at low frequencies have been well documented and are related to poorer neural synchrony.17,18

At the 500 and 1000Hz test frequencies, lower mean ASSR thresholds were observed at 40Hz MF than at 90Hz MF. In this situation, 40Hz ASSR seems to be superior to 90Hz ASSR, since it produced lower thresholds (perhaps closer to the behavioral hearing thresholds). The superiority of ASSR at low MF is consistent with the previous reports.6,14 van der Reijden et al.14 conducted a study to compare thresholds between tone burst ABR (t-ABR) and ASSR (at 40 and 90Hz MFs) at 500 and 2000Hz CFs. They then found that 40Hz ASSR yielded the lowest thresholds, particularly at 500Hz CF. ASSR at low MF is generated by the auditory midbrain, thalamus, and primary auditory cortex.19 If compared with high MF ASSR (i.e., predominantly generated by the auditory brainstem), lower ASSR thresholds at 40Hz MF is possibly due to increased neural connections and binaural activities within these upper regions of CANS.4 Furthermore, since female subjects were also included in the present study, the superiority of ASSR at 40Hz MF might also be influenced by a hormonal factor. Estrogen, the primary sex steroid for females, is known to affect GABAergic transmission that modulates ASSR amplitudes (see Zakaria et al.20 and Griskova-Bulanova et al.21 for detailed discussions).

At 2000 and 4000Hz test frequencies, the mean ASSR thresholds found in the present study were descriptively lower in ipsi-90Hz condition than in ipsi-40Hz condition. This difference was then found to be statistically significant only at the 4000Hz test frequency (Table 2). Both conditions (‘ipsi-40Hz vs. ipsi-90Hz’ and ‘contra-40Hz vs. contra-90Hz’) revealed insignificant statistical results at the 2000Hz test frequency. These findings are inconsistent with the results of the previous studies that found lower ASSR thresholds with low MF stimuli.6,14 The reason for this dissimilarity is unclear and possibly due to the methodological difference. The present study used narrow band CE-chirp® stimuli for determining ASSR thresholds, whereas pure tones were utilized for recording ASSR thresholds in the previous studies. Since the stimuli used are different, some differences in the study outcomes would be expected.

In the field of ASSR, literatures regarding the influence of ipsilateral and contralateral recordings are limited. In the present study, the majority of test frequencies (500, 1000, and 4000Hz) found no significant differences in ASSR threshold between ipsi-40Hz and contra-40Hz conditions. This suggests that the mode of recording has a subtle influence on ASSR thresholds evoked by low MF stimuli. This finding is in line with the study by Kaf and Danesh8 that found no significant differences in ASSR amplitudes and latencies between ipsilateral and contralateral recordings at 500, 2000, and 4000Hz CFs with 39Hz MF.

In contrast, for 90Hz MF stimuli, ipsi-90Hz condition produced statistically lower ASSR thresholds than that of contra-90Hz condition at 500, 1000, and 2000 test frequencies. In fact, at these test frequencies, contra-90Hz condition revealed the highest mean ASSR thresholds. Herein, a significant mode of recording effect on ASSR thresholds is noted for high MF stimuli. This result contradicts high MF findings in the study of Kaf and Danesh.8 That is, in their study, the ASSR amplitudes and latencies between ipsilateral and contralateral recordings were not found to be statistically different from each other at all tested CFs with 79Hz MF. This dissimilarity is possibly due to the methodological difference. While the present study employed ASSR threshold determination, Kaf and Danesh8 recorded ASSRs at a supra-threshold level (i.e., 65dB SPL). As stated earlier, ASSRs recorded at threshold and supra-threshold levels might yield different outcomes.9

The superiority of high MF ASSR in recording ipsilateral ASSR thresholds found in the present study is also inconsistent with the study by Small and Stapells.22 For adults, they found that the ASSR thresholds recorded at high MFs between ipsilateral and contralateral recordings were not statistically different at 500, 1000, 2000, and 4000Hz CFs. The reason for this disagreement is perhaps due to difference in sample size. While Small and Stapells22 recorded ASSR thresholds in 11 adults, the present study recruited a higher number of participants (n=29). It is known that a larger sample size would increase the statistical power and the likelihood for rejecting the null hypothesis.23 In other words, the significant ipsilateral ASSR results obtained for high MF in the present study seem valid due to the larger sample size. Nevertheless, the findings of the current study are consistent with previous studies in infants.7,22 Van Maanen and Stapells7 determined ASSR thresholds and amplitudes in two age groups of infants (>6 months and ≤6 months) at CFs of 500, 1000, 2000, and 4000Hz modulated between 81 and 101Hz. They then found that the contralateral ASSRs showed much smaller amplitudes and were often absent relative to the ipsilateral responses. In line with this, Small and Stapells22 found that for infants (mean age of 21 weeks), the ipsilateral ASSR thresholds were significantly lower than the contralateral ASSR thresholds at all tested CFs for both air- and bone-conduction stimulations.

The present study, nonetheless, has some limitations. Firstly, due to the significant interaction effects, the specific effect of test frequency on ASSR thresholds is not determined. Herein, the occurrence of more complex interaction effects was anticipated if frequency is included as one of the factors. Secondly, the present study only tested the right ear of the participants. Since ASSR shows a pronounced laterality effect,24 the outcomes of the present study might not be applicable to the left ear. In this regard, future studies are warranted to compare the ASSR thresholds between ears, as well as to further support the relevance of recording mode and modulation frequency in ASSR recording.