Measurements of human middle ear forward and reverse acoustics: Implications for otoacoustic emissions

Middle and inner ears from human cadaver temporal bones were stimulated in the forward direction by an ear-canal sound source, and in the reverse direction by an inner-ear sound source. For each stimulus type, three variables were measured: (a) Pec--ear-canal pressure with a probe-tube microphone within 3 mm of the eardrum, (b) Vst--stapes velocity with a laser interferometer, and (c) Pv--vestibule pressure with a hydrophone. From these variables, the forward middle-ear pressure gain (M1), the cochlear input impedance (Zc), the reverse middle-ear pressure gain (M2), and the reverse middle-ear impedance (M3) are directly obtained for the first time from the same preparation. These measurements can be used to fully characterize the middle ear as a two-port system. Presently, the effect of the middle ear on otoacoustic emissions (OAEs) is quantified by calculating the roundtrip middle-ear pressure gain Gme(RT) as the product of M1 and M2. In the 2-6.8 kHz region, absolute value(Gme(RT)) decreases with a slope of -22 dB/oct, while OAEs (both click evoked and distortion products) tend to be independent of frequency; this suggests a steep slope in vestibule pressure from 2 kHz to at least 4 kHz for click evoked OAEs and to at least 6.8 kHz for distortion product OAEs. Contrary to common assumptions, measurements indicate that the emission generator mechanism is frequency dependent. Measurements are also used to estimate the reflectance of basally traveling waves at the stapes, and apically generated nonlinear reflections within the vestibule.