Cullen J K, Long G R
J Acoust Soc Am. 1986 Jan;79(1):114-9. doi: 10.1121/1.393762.
Difference limens for trains of 30-microseconds pulses were determined for repetition rates of 50, 100, 200, 400, and 800 pulses per second under conditions of no filtering and high-pass filtering (115 dB/oct) with corner frequencies of 2.5, 5.0, 7.5, and 10 kHz. Low-pass-filtered noise was mixed with the trains of impulses to preclude discrimination on the basis of potential low-frequency signal components. Measures were obtained from four trained listeners at a signal level of 30 dB SL relative to individually determined thresholds for each filter condition and repetition rate. The data support the hypothesis that resolution of pulse-train repetition rate involves both temporal- and frequency-based processes--the latter becoming ineffective when frequency resolution of the ear is insufficient to resolve separate harmonics of the signal. Inter- and intra-individual differences are interpreted as reflecting frequency resolution capacity.
在无滤波和高通滤波（115分贝/倍频程）且转折频率分别为2.5、5.0、7.5和10千赫的条件下，针对每秒50、100、200、400和800个脉冲的重复率，测定了30微秒脉冲串的辨别阈限。低通滤波噪声与脉冲串混合，以排除基于潜在低频信号成分进行辨别的可能性。在相对于每个滤波条件和重复率单独确定的阈值的30分贝感觉级信号水平下，从四名经过训练的听众那里获取了测量结果。数据支持这样的假设，即脉冲串重复率的分辨涉及基于时间和基于频率的过程——当耳朵的频率分辨率不足以分辨信号的单独谐波时，后者变得无效。个体间和个体内差异被解释为反映频率分辨能力。
