Frequency selectivity mediated by inner hair cells alone

A well-recognized measure of frequency selectivity can be obtained by a tone-on-tone masking paradigm, which generates a so-called “psychophysical tuning curve” [e.g., A. M. Small Jr., J. Acoust. Soc. Am. 31, 1619–1625 (1959); E. Zwicker in Facts and Models in Hearing (Springer-Verlag, New York, 1974), pp. 132–139]. This procedure was adapted to chinchillas trained with the shock-avoidance method. In order to eliminate outer hair cells from controlled regions of the cochlea, kanamycin was administered. We have demonstrated previously that this method is suitable for producing relatively pure outer hair cell losses in the basal part of the cochlea, with the immediate psychophysical correlate of an approximately 40-dB threshold shift [A. Ryan and P. Dallos, Nature (in press, 1975)]. Measurements of the “tuning curves” both before and after kanamycin treatment indicated that (1) for maskee frequencies in the unaffected frequency region the “tuning curves” are unchanged, and (2) when the maskee is in the affected region, the segment of the “tuning curve” also within that region is simply shifted proportional to the threshold shift. The curves do not become wider. This suggests that the sharpness of tuning in the auditory system can be determined by inner hair cells, acting without significant influence by the outer hair cells. [Work supported by grants from the NINDS. We thank T. McGee and C. Sanes for their contribution to this research.]