40 individuals between 7 and 24 years of age were part of the study. Of these, 20 complained of difficulties associated with auditory processing, and had alterations in their assessments, making up the study group (TPA). The remaining (20) individuals were invited to participate in the study because they did not have complaints associated with auditory processing, learning or language, and did not show alterations in the auditory processing assessment, making up the control group (DT).

The subjects from both groups were tested as to BEAP responses for clicks and speech stimulus. The responses were recorded by means of the Traveler Express portable system by BioLogic.

We inspected the external acoustic canal using the otoscope in order to check for the presence of any factor that could prevent the exam from being performed. After skin cleaning with abrasive paste, the electrodes were placed on the vertex and on the right and left mastoids, using electrolytic paste and adhesive tape. Impedance values for the electrodes were below 5 KWs.

The click stimulus was presented to the right ear by mean of earphones at a rate of 19 stimuli per second, and we had 2,000 stimuli at 80 dBnHL being produced. The recording window was of 10 milliseconds and the stimuli were filtered from 100 to 3,000 Hz. A second stimulation was carried out in order to produce and confirm the wave trace. In this we identified and analyzed waves I, III and V.

For BEAP speech stimulation we had to produce such stimulus. We chose a natural stimulus instead of the synthesized one, thus the /da/ syllable was narrated in a recording studio by a male voice, recorded using the Sound Forge 6.0 (Sony) software and edited by the Vegas 4.0 (Sony) software, in such a way as to produce a stimulus with similar patterns to those described by King et al.8 and Wible et al.11. From the original syllable, only the five first formants were separated, which resulted in the 40 ms stimulus, with the transient portion of the syllable. Just like the stimulus used by Wible et al.11 - the vowel /a/ was shortened in order to increase the stimulation rate and, therefore, better activate the system.

The stimuli were organized in groups of 4, separated among themselves by 12 ms and, among each group of stimulus, the interval was of 30 ms. According to Wible et al.1112 ms is the shortest interstimuli interval that can be used without presenting a stimulus while the previous one is still active.

The speech stimulus was presented by means of headphones coupled to a disc-man, on the right ear at 80 dBH at a rate of 18 stimuli per second. A total of 3,000 stimuli were produced, 3 scans of 1,000 stimuli. The recording window was of 45 milliseconds.

The curves obtained in each scan were added up and on the resulting trace, waves V, A, C and F were identified and analyzed.

Analysis of waves I, III and V generated by the click stimulus was based on the conventional clinical analysis of the absolute latencies, interpeak latencies and amplitudes.

The analysis of waves V, A, C and F (chart 1) generated by the speech stimulus was based on the descriptions from Johnson et al.12.

The values for both stimuli went through statistical analysis in which the mean, median, standard deviation, minimum, maximum and confidence interval values were calculated for the sample. In order to check whether or not there were statistically significant differences regarding the results found in both groups, we studied the p value. The p value was calculated by means of the T student test and Variance Analysis (ANOVA) with significance levels of 5% (P < 0.05).

The trace analysis of all the subjects for click stimulus showed the presence of waves I, III and V with latency values as well as interpeak values within normal standards, according to Hall's parameters20. These data showed auditory pathway intactness all the way to the BS in all the subjects (n = 40).

When we compared the values of latency, amplitude and interpeak intervals (Table 1) we did not observe significant differences (p Value > 0.05) between the DT and TPA groups, and the averages of waves I, III and V latency values were very close. We noticed a trend towards having significant differences between the groups as to the mean values of the Interpeak I-III interval.

For all the subjects, waves V, A, C and F were identified in the tracing resulting from the summation of the three scans performed. In one individual only (2.5%), belonging to the TPA group, it was not possible to identify wave F, because he did not have it or it was in a greater latency than that of the interval studied (45 ms). For this reason we analyzed only the responses from 19 individuals from the study group. Thus, the mean, standard deviation and p values of the latencies and amplitudes were calculated and analyzed (Table 2).

There was a significant difference (p Value < 0.05) between the DT and TPA groups for the latency values of waves V and A. For the C wave latency, we noticed only one trend towards significant difference between the groups. Now, as far as amplitudes are concerned, we observed significant differences between the DT and TPA groups for the values of waves A, C and F.

The analysis of the VA Complex (Table 3) in relation to the mean values of latency, amplitude and slope (VA amplitude VA/ VA duration) showed a significant difference between the DT and TPA groups in regards of the VA complex amplitude values and, consequently, of slope.

In this study we observed normal response values for clicks, both for the DT and TPA group, without significant differences between the groups, leading us to believe that individuals with TPA process sounds such as the click, at the level of the BS, in a similar fashion to that of individuals with normal development.

Hall and Johnson21 state that, for individuals with TPA, the proportion of abnormalities in BEAP is much lower than that in the middle and long latency potentials, which strengthens our previous assumption.

Song et al.14 investigated the responses for clicks between a group of children with normal learning and a group of children with learning disorder. No significant differences were found between the groups, not even when contralateral masking was introduced. In this case, the amplitudes of both groups were similarly reduced.

Thus, we can state that individuals with auditory processing disorders did not have lesions in their auditory pathway all the way to the brainstem or, even, did not have difficulties in processing simple acoustic information such as the click.

In this study, for the TPA group in relation to the DT, speech stimuli responses with latency of waves A, C and the VA complex were significantly reduced, and also there was a lower VA complex slope. We can then suggest that children with TPA may have changes in the response generator synchronization (amplitude differences) and/or in the nervous signaling transmission velocity (latency differences)11. Other researchers5,8,11 studied the BEAPs for speech stimuli in subjects with normal development and those with learning difficulties and found results similar to the ones we did. King et al.8 noticed significant differences in relation to the A wave latency and suggested that at least some of the children with learning disorders have abnormalities in the acoustic representation of speech sounds in the low brainstem.

Kraus & Nicol5 found significantly increased onset response latencies (waves V, A and C) in children with learning disorders when compared to their counterparts with normal development. Wible et al.11 observed that the children with learning disorders had lower VA complex slope when compared to children with normal development.

When commenting the workshop “Speech Perception Nature “, Kraus & Nicol5 mention that: speech perception involves peripheral auditory analyses, automatic extraction of the brainstem nuclei characteristics, which lead both to the classification of words and phonemes. Thus, the alterations observed in speech processing at the BS could represent more specific alterations at a cortical level.

Although this relationship was not observed in this study, because cortical evaluations were not carried out, Wible et al.17 concluded that a greater coding synchronicity of the speech transient aspects (onset responses, waves V, A and C) at the level of BS contribute to a more robust processing at a cortical level, and vice-versa. Therefore, a BS alteration can suggest difficulties in speech cortical processing and, consequently, in the very development of speech and language.

This relationship becomes even clearer with the analysis of response alterations for speech in the BS based on Fant's Acoustic Theory of Speech Production19, as suggested by Kraus & Nicol22 and Johnson et al.12. Since the alterations noticed happened to waves V, A and C, which reflected the system's response to the speech filtering capacity, the difficulties found by TPA children for speech processing would happen in relation to linguistic aspects.

Johnson et al.12 suggest that the BS measures associated to the information generated by the speech filter may function as biological markers of neural asynchronicity in children with learning disorders such as dyslexia or in children with auditory processing disorder.

Therefore, based on the aforementioned, it would not be too difficult to state that the speech processing alteration in the BS causes a failed development in linguistic skills.

Although the study's goal was not to define normality parameters, we compared the DT group responses for speech stimulus with the parameters obtained by Russo et al.7 (Chart 2). The values found are similar, without significant differences, although in our study they are slightly increased, which can be justified by mild differences in response collection and analysis.

The results obtained from this study showed in the TPA group possible alterations as to neural signaling velocity and synchronicity in the processing of complex stimuli like speech, especially as to linguistic information.

Further studies must be carried out comparing the relationship of speech responses at the brainstem with those at a cortical level, and also the relationship between the electrophysiological responses and the behavioral ones, seeking a better understanding of the auditory processing of complex signals like speech.

Another goal that must be pursued associated with responses to complex stimuli at the brainstem level is the relationship with language disorders, speech, learning and auditory processing and the consequences of auditory treatment and Speech and Hearing Therapy in speech processing.