The purpose of this study is to establish the standardization and our method of EEG-computer audiometry available in practice, because variable responses obtained depend on each of the age groups, especially in case of infant and child, and changes in the stages of sleep.
Averaged auditory evoked responses consist of fast and slow components, of which the latter is most available. The slow components are stable response of long duration at the time of awaking. Each 30-100 times averaged auditory response, obtained at random in the data recorder, was almost invariable.
The averaged auditory response in each 30-100 times, peak latency and amplitude of which was stable. The pattern of the latter component as response was clearer than that of early component. This is a fact that there were some tendency showing similar patterns of component to normal awaking adult on its response in both cases of conductive and sensory deafness. Studying for standardization of the EEG-computer audiometry and the investigation of the fundamental judgement in objective audiometry under the various condition s, we take into consideration the clinical evaluation of our new technique.
On the frequency analysis evoked response
The results were as follows: the patterns of the averaged auditory responses corresponded to narrow frequency-band of 3-7 Hz in an awaking adult. If we could amplify the frequency narrow-band within the averaged evoked responses obtained, not any noises could be noticed in the original averaged evoked response.
According to our ideas, it could be obtained some averaged evoked responses, using narrow-band amplifier in an awaking adult which were called the frequency analysis evoked response.
We compared an N₁-P₂ of the averaged evoked response to 100 times of 1KHz, 60 dB in upper channel of computer (Signal: S₁) with a maximum amplitude of the frequency analysis evoked response in its lower channel (Signal: S₂). The amplitude of the noise-components were indicated as a Noise: N which was not more than the ratio 0.27 mV in this technique even though over 100 times summated, and it was also obtained a ratio S/N of each response value and noise value in the correlation between response and noise level.
For example, in some cases ratio S/N which was 1 or less, was impossible to read, but it was easily recognized, when ratio S/N of the response is larger than 1.
A ratio S/N in auditory evoked response at under 20 dB, was often clinically impossible to read, while a ratio S/N in frequency analysis evoked response was obtained clinically larger than a ratio S/N in an averaged evoked response.
In case of high intensity of stimulus for example 60 dB, it could be easily developed the ratio S/N even though in fewer times of summation. In case of 10 dB of intensity, ratio S/N of averaged evoked response showed about 1, which was not developed enough. As compared with former results, we found that ratio S/N of frequency analysis evoked response showed 2.5 of ratio in 60 times, which was easily recognized.
It was summarized that frequency analysis evoked response used in narrowband filter, dominantly 3-7 Hz were indicated to control or reduce some noises or artifacts and also showed some pattern as a clear sign of averaged evoked response. In the objective auditory test, in the case of sleeping stage either infant and child, it is necessary to use the lower frequency band filter than in the case of an awaking adult. The reasons of our method are as follows:

View full abstract