Cochlear nonlinearity and gain control as determinants of the response of primary auditory neurons to harmonic...

Earlier papers reported a calculational "model" (Greenwood, 1985, 1986a, b, c), composed of an initial compressive nonlinearity accompanied by a nonlinear gain control and "rectification", to demonstrate in principle certain "two-tone suppression" effects seen in primary fiber response. This paper presents simulations of primary fiber response to more complex stimuli. When harmonic stimuli of Horst et al. (1985a, b; 1986a, b) are used, the calculated spectra closely agree with the empirical spectra they report, from fibers with both low or high spontaneous activity. Thus, the simulated nonlinear motion of the cochlear partition reflects a sufficient degree of compressive distortion and an appropriate form to calculate spectra closely matching the empirical spectra in respect to the pattern of intermodulation products and the main "enhancement" and cancellation effects. Comparison of spectra of full and truncated waveforms assess added effects of "rectification". The gain control, as for simpler stimuli, leads to appropriate growth and saturation of Fourier coefficients and average firing rate. The implication that the measured transverse motion of the cochlear partition could provide sufficient distortion to account for the main features of the histograms has a two-part corollary: that the mechanism which produces severe overall saturation does not add much compressive distortion to the waveform passed radially to IHC and primary fibers, and that the IHC is kept operating chiefly in an approximately linear part of its range, by a prior gain control that is, perforce, cochlear and mechanical. Further consequences, for IHC synapses and fiber types, stem from the sufficiency of negatively biased waveform centerlines in generating simulated period histograms and spectra which match those of at least some primary fibers characterized by low spontaneous activity.