Discrimination of Japanese monosyllables in patients with high-frequency hearing loss

To analyze the difficulty of discriminating Japanese nonsense monosyllables in each of several grades of high-frequency hearing loss and to evaluate the limitations of amplification.

We collected retrospective data on the discrimination of Japanese nonsense monosyllables by patients with three grades of high-frequency hearing loss who fulfilled or nearly fulfilled the Japanese criteria for EAS. Discrimination of the twenty monosyllables included in the 67-S speech audiometric test, which is approved by the Japan Audiological Society, was evaluated under quiet conditions.

One hundred and five ears of ninety-one adults with high-frequency hearing loss were tested. We classified the ears according to hearing threshold at 1000 Hz; Group 1: <45 dB; Group 2: ≥45 dB and <70 dB; Group 3: ≥70 dB. Under the best conditions, the best speech discrimination scores were 72.3 ± 18.6% (mean ± SD, N=11), 56.9 ± 19.9% (N=57) and 38.1 ± 22.6% (N=37) in Group 1, Group 2 and Group 3, respectively. For most of the monosyllables, discrimination score declined gradually as high-frequency hearing loss became more severe. The high incidence in the Japanese language of [k], an easy consonant to distinguish, may be an advantage for patients with high-frequency hearing loss who use hearing aids. By employing a new confusion matrix that displays consonants and the following vowels separately, we were able to reveal the interactions of those two components. We observed that discrimination of preceding nasal consonants and that of the following vowels were not independent in patients with high-frequency hearing loss.

Our classification based on threshold at 1000 Hz was useful to predict the effectiveness and limitations of amplification in high-frequency hearing loss. Threshold at 1000 Hz can be an index enabling us to refine the indications of EAS for native Japanese speakers to maximize its effectiveness against high-frequency hearing loss.