Timing of spike initiation in cochlear afferents: dependence on site of innervation.

1. The phase of excitation of inner hair cells (IHCs) relative to basilar membrane motion has been estimated as a function of best frequency (BF) (or, equivalently, cochlear location) by recording responses to tones (100-1,000 Hz) from chinchilla cochlear afferent axons at their central exit from the internal auditory meatus. 2. The time of IHC excitation (i.e., the time of chemical transmitter release) was derived from the neural recordings at near-threshold levels by applying a correction for the latency of synaptic processes and the propagation time of action potentials. 3. The phase of basilar membrane motion at the appropriate innervation site was estimated on the basis of previously measured basilar membrane responses at a location close to the basal end of the cochlea and estimates of mechanical travel time from the basal end to the innervation site, derived from the neural latencies to intense rarefaction clicks, as a function of BF. 4. The derived near-threshold excitation of basal IHCs leads basilar membrane displacement toward scala tympani by approximately 40-60 degrees. 5. At BFs corresponding to midcochlear locations (2-6 kHz) there is an abrupt phase transition. The derived excitation for IHCs located at more apical locations (BFs large in relation to stimulus frequency) corresponds approximately to peak velocity of the basilar membrane toward scala vestibuli. 6. Although the derived response phases of apically located IHCs are consistent with intracellular recordings from IHCs, the derived near-threshold response phases of basal IHCs may be inconsistent with intracellular IHC recordings. 7. The foregoing results, based on responses of nearly 1,000 cochlear afferents to tones 100-1,000 Hz at near-threshold stimulus levels, amply confirm our previous conclusions that were based on a smaller sample of responses to very low frequency tones (less than or equal to 100 Hz): there is a spatial transition at midcochlear regions in the mode of excitation of IHCs, which does not seem to simply reflect the macromechanics of the basilar membrane. 8. It has been proposed that both the paradoxical response phases of high-BF afferents and the spatial phase transition arise from an influence of cochlear microphonics on the transmembrane potential of IHCs. The present results, which show that the spatial phase transition occurs for frequencies at least as high as 400 Hz, would appear to make such an electrical influence of outer hair cells on IHCs less likely. An alternative explanation might be that the phase transition has a mechanical basis, perhaps localized to micromechanical events in the subtectorial regio