The "inverse problem" solved for a three-dimensional model of the cochlea. II. Application to experimental data sets.

With "classical" nonactive models of the cochlea it is impossible to simulate the degree of frequency selectivity that is revealed by modern mechanical measurements of the motion of the basilar membrane (BM). Locally active models have been proposed to alleviate this problem, but it remains uncertain whether the actual cochlea is active in this sense. In the first paper of this series [E. de Boer, J. Acoust. Soc. Am. 98, 896-903 (1995)], the "inverse" problem is solved for a (classical) three-dimensional model and a procedure is developed for recovering the BM impedance needed to simulate a given BM response function. It was found that the results of this procedure will be more accurate in the region of the response peak than in the more basal region of the model. In the present paper the same procedure is applied to data of recent mechanical experiments. For the peak region the outcome is unequivocal: Recent measurement results can only be simulated by the classical model when it is made locally active. Resynthesis of the model response, on the basis of the recovered BM impedance, confirms the validity of the method in the peak region and strengthens this conclusion.