Narrow-band AP latencies in normal and recruiting human ears.

Derived narrow-band action potential latencies increase monotonically with decreasing central frequency, and can be interpreted as reflecting the traveling wave delay in the cochlea. It was found that, for recruiting human ears with average flat hearing losses around 40 dB, this accumulating latency increase was smaller than for normal ears. A comparison of 15 normal ears and 37 recruiting ears showed, however, that in only half of the recruiting ears this difference was significant. These recruiting ears were therefore divided in two groups based on the waveform of the narrow-band action potential AP, which correlated well with the subdivision according to latency. The findings have been explained on the basis that latency of the narrow-band APs is not determined solely by the mechanical traveling-wave delay, but also by the response time of the (second?) cochlear filter. When this filter broadens, one expects a decrease in its impulse response time. Since this impulse response time. Since this impulse response depends on the sum of the high- and low-frequency slope values of the cochlear filter, one expects only a latency decrease when the steep high-frequency slope also becomes more shallow. A support for the influence of the response times of the cochlear filter is found in the narrow-band AP latencies for restricted cochlear losses (e.g., in a 4-kHz noise dip). It appears that the latency in that area actually is shorter than for the higher central frequencies, a fact which cannot be explained solely on the basis of a traveling wave phenomenon.