Nonlinear interactions that could explain distortion product interference response areas

Suppression and/or enhancement of third- and fifth-order distortion product s by a third tone that can have a frequency more than an octave above and a level more than 40 dB below the primary tones have recently been measured by Martin et al. [Hear. Res. 136, 105-123 (1999)]. Contours of iso-suppress ion and iso-enhancement that are plotted as a function of third-tone freque ncy and level are called interference response areas. After ruling out orde r aliasing, two possible mechanisms for this effect have been developed, a harmonic mechanism and a catalyst mechanism. The harmonic mechanism produce s distortion products by mixing a harmonic of one of the primary tones with the other primary tone. The catalyst mechanism produces distortion product s by mixing one or more intermediate distortion products that are produced by the third tone with one or more of the input tones. The harmonic mechani sm does not need a third tone and the catalyst mechanism does. Because the basilar membrane frequency response is predicted to affect each of these me chanisms differently, it is concluded that the catalyst mechanism will be d ominant in the high-frequency regions of the cochlea and the harmonic mecha nism will have significant strength in the low-frequency regions of the coc hlea. The mechanisms are dependent on the existence of both even- and odd-o rder distortion, and significant even- and odd-order distortion have been m easured in the experimental animals. Furthermore, the nonlinear part of the cochlear mechanical response must be well into saturation when input tones are 50 or more dB SPL. (C) 2000 Acoustical Society of America. [S0001-4966 (00)04510-0].