Cochlear preprocessing in analog models, in digital models and in human inner ear

Data of spontaneous, delayed, and simultaneously evoked otoacoustic emissions produced in human subjects are compared with data produced in an analog and in a digital wave-parameter realization of our model. The high probability of the frequency distance of the emission's extreme values at 0.4 Bark, corresponding to a local distance of 0.5 mm along the basilar membrane is found also in the models. The influence of the lateral feedback coupling in the models on the frequency selectivity and on the appearance of emissions show the high quality of simulating human peripheral signal processing by the models.