单音节词表双音节词表和句表的P-I函数曲线绘制

摘要

Objective To establish performance-intensity function (P-I) curves for monosyllable and di-syllable words lists and short sentences lists, in an attempt to obtain parameters of normal persons. Method Otologically normal subjects (n=20) were recruited from colleges and universities in Nanjing. Using new man-darin speech test materials including monosyllable and disyllable words lists and short sentences lists created by the Department of Otolaryngology, People’s Hospital of Jiangsu Province, intensity points across a speech recognition range of 20-80%and corresponding speech recognition scores, thresholds, as well as the 20-80%and 50%slopes, were obtained. Logistic regression functions were used to create P-I curves. Results Average speech recognition thresholds for monosyllable and disyllable words lists and short sentences lists were 28.5± 0.5 dB SPL, 24.4±0.5 dB SPL and 26.3±0.7 dB SPL, with average linear slopes over the 20%-80%recogni-tion range at 3.5 ± 0.1%/dB SPL, 6.3 ± 0.5%/dB SPL and 7.2 ± 0.4%/dB SPL, respectively. Conclusions This study has established preliminary normative P-I curves for six monosyllable words lists, eight disyllable words lists and five short sentences lists, which can be used in clinical applications as normal hearing stan-dards.%目的：绘制“男女声录制普通话言语测听材料”中单音节和双音节词表以及短句表的言语识别率-强度函数（Performance-intensity function，P-I）曲线，获得正常人P-I函数曲线的言语识别阈和斜率两个参数，并加以分析比较。方法选择耳科正常人20名，采用江苏省人民医院耳鼻咽喉科编写的“男女声录制普通话言语测听词表”，包括单音节词表6张、双音节词表8张及短句表5张，先分别获得其20％～80％的言语识别率范围所需的6个强度点，计算出各词表在各个强度下的识别率、言语识别阈、20％～80％的斜率和50％的斜率，采用Logistic回归函数进行分析，绘制其P-I曲线。结果（1）单音节词表：20％～80％言语识别率范围所需强度为21、25、25、30、35、40dB(SPL)，线性范围斜率为（3.5±0.1）%/dB，平均言语识别阈为28.5±0.5 dB SPL；（2）双音节词表：20％～80％言语识别率范围所需强度为20、22、24、26、28、30 dB(SPL)，线性范围斜率为（6.3±0.5）%/dB，平均言语识别阈为24.4±0.5 dB SPL；（3）句表：20％～80％言语识别率范围所需强度为22、24、26、28、30、32dB(SPL)，线性范围斜率为（7.2±0.4）%/dB，平均言语识别阈26.3±0.7 dB SPL。结论（1）三类词表正常人P-I曲线均呈侧向拉伸的S型，其言语识别率随着言语刺激信号强度的增加而增加，并在言语刺激信号强度上升一定程度后趋于稳定，从而达到最大言语识别率。（2）句表的20%-80%线性范围斜率最大,其P-I函数曲线走形最陡峭。