Kidd G
Department of Communication Disorders, Boston University, Massachusetts 02215.
J Acoust Soc Am. 1993 Feb;93(2):992-6. doi: 10.1121/1.405405.
The effect of the number of nonsignal tones on the just-discriminable difference in spectral shape was evaluated for 12 naive subjects. The signal was an intensity increment to the center tone of a multitone complex. The tones were spaced at equal frequency ratios and, except for the signal increment, were equal in level. The number of nonsignal tones ranged from 2 to 20. Consistent with previous studies using equally spaced components, group mean thresholds improved monotonically with increasing number of tones. Group mean thresholds improved about 8 dB over the range of reference spectra employed. Large differences were found across subjects in overall performance, and in the magnitude of the improvement with increasing number of nonsignal tones. The greatest intersubject variability was observed for the stimuli composed of the fewest components. It was concluded that the differences reported in the literature concerning the presence and magnitude of the improvement are due both to differences in the procedures used and to the large individual differences that occur among subjects.
针对12名未受过训练的受试者，评估了非信号音数量对频谱形状可分辨差异的影响。信号是多音复合体中心音的强度增量。这些音以相等的频率比间隔排列，除了信号增量外，电平相等。非信号音的数量范围为2至20个。与之前使用等距分量的研究一致，组平均阈值随着音数量的增加而单调改善。在所采用的参考频谱范围内，组平均阈值提高了约8分贝。在整体表现以及随着非信号音数量增加而改善的幅度方面，受试者之间存在很大差异。对于由最少分量组成的刺激，观察到受试者间最大的变异性。得出的结论是，文献中报道的关于改善的存在和幅度的差异，既归因于所使用程序的差异，也归因于受试者之间出现的巨大个体差异。
