| The ear not only receives sound but emits sounds of its own. These otoacoustic emissions (OAEs) originate as a byproduct of active cochlear processes, and provide a noninvasive window onto the mechanisms underlying peripheral hearing sensitivity and frequency tuning. However, despite the fact that changes in the active processes often start at the basal, high-frequency end of the cochlea, the potential utility of high-frequency OAE measurements has not been thoroughly explored. To address this, the global and local spectral features, or "macro-" and "microstructures", of behavioral hearing thresholds and tone-evoked, stimulus-frequency OAEs (SFOAEs) were characterized over a wide frequency range. First, SFOAEs and thresholds were compared from 0.5-20 kHz in young, normal-hearing individuals. While SFOAEs were measurable up to the highest frequencies of sensitive hearing, their amplitudes often substantially declined a half-octave or more lower, despite little change in threshold at the same frequency. To further explore the potential cochlear and middle ear factors influencing SFOAE measurement, the microstructures (i.e., quasiperiodic spectral fluctuations) of thresholds and SFOAEs were examined across frequency. Microstructure is thought to arise via multiple reflections of evoked SFOAE energy between its site of origin and the middle ear boundary. Comparison of the SFOAE amplitude in the ear canal with the strength of the microstructure may therefore indicate the relative magnitudes of the signals that are transmitted through the middle ear and those that are reflected back into the cochlea. The results indicate that (1) there is a common microstructure in behavioral thresholds and SFOAE responses, (2) the relationships between microstructure and SFOAE phase/magnitude are consistent with the theoretical origins of microstructure, and that (3) microstructure is very weak at high frequencies. There was therefore little evidence that large, high-frequency SFOAEs were generated inside the cochlea but not efficiently transmitted out to the ear canal. Overall, the results suggest that SFOAEs are indeed measurable, but low in amplitude, at high frequencies, and that their high-frequency decline is primarily attributable to a weakening of the cochlear mechanisms underlying their generation. If high-frequency SFOAEs are very sensitive to small changes in basal cochlear status, they may be useful in the early detection of damage due to aging, noise exposure, and ototoxicity. |
