Moore B C
Department of Experimental Psychology, University of Cambridge, England.
Ann N Y Acad Sci. 1993 Jun 14;682:119-36. doi: 10.1111/j.1749-6632.1993.tb22964.x.
The ear contains an array of filters that separate the components of a complex signal into "channels" tuned to different center frequencies. Temporal analysis can be considered as two processes: analysis of the time pattern occurring within each channel, and comparison of the time patterns across channels. Within-channel acuity can be characterized by tasks such as gap detection, or by the ability to detect amplitude modulation as a function of modulation rate. The smallest detectable gap duration for a white noise stimulus is 2-3 ms. The results can be modeled by an array of filters, with each filter followed by a nonlinearity and a (central) sliding temporal integrator. Hearing impairment of cochlear origin can have adverse effects on temporal resolution because it often reduces the audible bandwidth of the stimuli, and because it results in a reduced sensation level of the stimuli. The sliding temporal integrator appears to be unaffected by hearing loss, although the nonlinearity preceding the integrator may be abnormal, and this can lead to reduced temporal resolution for sounds with slowly fluctuating envelopes. Hearing impairment of more central origin may also adversely affect temporal resolution, but the mechanisms responsible for this are not known. The acuity of across-channel temporal analysis depends on whether the task is one of discrimination or of identification of temporal order. The finest acuity (1-2 ms) occurs for discrimination tasks. Identification of temporal order is an order of magnitude worse. When the elements of a sequence of sounds are perceived as more than one source (more than one perceptual stream), the ability to judge the order of the elements can be very poor. Perceptual grouping processes can also have dramatic effects on the perceived temporal structure of sound. Conversely, temporal structure can have a powerful influence on perceptual grouping.
耳朵包含一系列滤波器，这些滤波器将复杂信号的各个成分分离到调谐至不同中心频率的“通道”中。时间分析可被视为两个过程：对每个通道内出现的时间模式进行分析，以及对不同通道间的时间模式进行比较。通道内的敏锐度可用诸如间隙检测之类的任务来表征，或者用检测作为调制率函数的幅度调制的能力来表征。对于白噪声刺激，最小可检测间隙持续时间为2 - 3毫秒。这些结果可用一系列滤波器来建模，每个滤波器后面跟着一个非线性元件和一个（中心）滑动时间积分器。耳蜗源性听力损失会对时间分辨率产生不利影响，这是因为它常常会降低刺激的可听带宽，还因为它会导致刺激的感觉水平降低。尽管积分器前面的非线性元件可能异常，但滑动时间积分器似乎不受听力损失的影响，而这可能会导致对包络缓慢波动的声音的时间分辨率降低。更中枢性起源的听力损失也可能对时间分辨率产生不利影响，但其背后的机制尚不清楚。跨通道时间分析的敏锐度取决于任务是辨别还是识别时间顺序。辨别任务的敏锐度最高（1 - 2毫秒）。识别时间顺序的能力则要差一个数量级。当一系列声音的元素被感知为不止一个声源（不止一个感知流）时，判断这些元素顺序的能力可能会非常差。感知分组过程也会对声音的感知时间结构产生显著影响。反之，时间结构会对感知分组产生强大影响。
