Wilson R H, Carter A S
Auditory Research Laboratory, James H. Quillen VA Medical Center, Mountain Home, Tennessee 37684, USA.
J Am Acad Audiol. 2001 Jan;12(1):7-14.
This tutorial paper examines the relation between the slope of a mean word recognition function and the homogeneity or variability (with respect to recognition) of the individual stimulus items that compose the test materials. This was studied in terms of both the location (Cartesian) and slope of the psychometric functions of the individual words that compose the materials. Word recognition performances were measured for 100 CID W-22 (Hirsh) words and 100 PB-50 (Rush Hughes) words in quiet (0 to 56 dB HL in 8-dB steps) on 12 subjects with normal hearing. The functions for the individual W-22 words were more homogeneous (less variable) than were the functions for the individual PB-50 words. The mean function for the W-22 words was steeper (3.1%/dB) than the function for the PB-50 words (1.9%/dB). This evaluation of the individual words demonstrates the direct relation between variability of the test items and the slope of the mean psychometric function. The more homogeneous performance is on the individual test items with respect to both location and slope, the steeper the slope of the mean psychometric function.
本教程论文探讨了平均单词识别函数的斜率与构成测试材料的各个刺激项目在识别方面的同质性或变异性之间的关系。这是从构成材料的各个单词的心理测量函数的位置（笛卡尔坐标）和斜率两方面进行研究的。对12名听力正常的受试者在安静环境下（以8分贝为步长，从0到56分贝听力级）的100个CID W-22（赫什）单词和100个PB-50（拉什·休斯）单词的单词识别表现进行了测量。各个W-22单词的函数比各个PB-50单词的函数更具同质性（变异性更小）。W-22单词的平均函数比PB-50单词的函数更陡峭（3.1%/分贝比1.9%/分贝）。对各个单词的这种评估表明了测试项目的变异性与平均心理测量函数的斜率之间的直接关系。各个测试项目在位置和斜率方面的表现越具同质性，平均心理测量函数的斜率就越陡峭。
