Model studies of source term and reflection components of synchronous-evoked and distortion product otoacoustic emissions

Abstract

The unified model of synchronous-evoked, distortion product, and spontaneous otoacoustic emissions of Talmadge et al. [J. Acoust. Soc. Am. 104, 1517 (1998); 105, 275 (1999); 108, 2570 (2000)] provides a useful testing ground for various paradigms that have been proposed for separately determining source terms and reflectance components of synchronous-evoked and distortion-product otoacoustic emissions. The source terms are the ear canal components that would exist in the absence of effects associated with cochlear wave reflections due to, e.g., distributed inhomogeneities [Zweig and Shera, J. Acoust. Soc. Am. 98, 2018 (1995)] and nonlinearity [Talmadge et al., 2000]. Model calculations are used to compare results for the unmixing of these components obtained using reflection-suppression techniques [see, e.g., Kalluri and Shera, J. Acoust. Soc. Am. 109, 622 (2001)] and various forms of time-domain latency windowing ones [see, e.g., Knight and Kemp, J. Acoust. Soc. Am. 107, 457 (2000)]. Because of multiple internal wave reflections at the cochlear base and the tonotopic place for the cochlear waves, the model implies (and experimental data indicate) that the reflectance components will generally exhibit fine structure effects similar to those of the total ear canal signal.