Cochlear microphonic evidence for mechanical propagation of distortion products (f2 - f1) and (2f1 - f2)

The present cochlear microphonic (CM) study was undertaken to help resolve a conflict in the literature regarding cochlear nonlinear properties. The CM studies of Dallos and his coworkers have concluded that "(up to 70-80 dB SPL [re 20 mu Pa]), all orders of distortion components...do not seem to be accompanied by traveling waves of their own" (Dallos, P. (1973): The Auditory Periphery: Biophysics and Physiology. Academic Press, New York). However, studies of spatial distributions of cochlear nerve fiber responses, acoustic distortion products in the ear canal, and related modeling studies of Kim et al. (Kim, D.O. and Molnar, C.E. (1975): in: The Nervous System, Vol. 3: Human Communication and Its Disorders. Editor: D.B. Tower. Raven Press, New York; Kim, D.O., Molnar, C.E. and Matthews, J.W. (1980): J. Acoust. Soc. Am. 67, 1704-1721) have led to conclusions to the contrary. In the present study, CM data were obtained from the second and third turns of the chinchilla cochlea using fluid-filled glass micropipettes in scala media and nichrome wire electrodes in scala vestibuli and scala tympani. We sought responses containing predominant distortion products (f2 - f1) and (2f1 - f2) by fixing a distortion frequency (fD) near the characteristic frequency (CF) of the recording site and varying the stimulus frequencies f1 and f2 and SPLs (with L1 = L2). By subsequently varying the distortion frequency around the CF, e.g., fixing f1 well above the CF and varying f2, we measured the tuning characteristics of the distortion products (f2 - f1) and (2f1 - f2). Tuning characteristics of single-tone responses were measured by applying single-tone stimuli of various frequencies with a constant SPL at the eardrum. We have observed, with SPLs as low as 25 dB, that these distortion products in CM display tuning similar to the single-tone response which is consistent with above neural results. From these tuning similarities, we conclude that our CM data reflect the presence of mechanically propagated distortion products at low SPLs, in agreement with the above studies by Kim et al. Validity of our results is supported by the sensitivity and sharp tuning of our CM data and, in the case of the scala media recordings, by the presence of a normal d.c. endolymphatic potential. Plausible explanations for the opposing conclusions of previous studies of Dallos et al. and the present study are discussed.