Long-latency component of stimulus frequency otoacoustic emissions: Relationship to behavioral measures

Abstract

Uncertainties regarding the generation mechanism and source of human stimulus frequency otoacoustic emissions (SFOAEs), as well as the possible generation of artifacts, have previously restricted the use of SFOAEs in clinical settings for assessing cochlear function. Over the years, advanced time-frequency analysis of SFOAEs, using models of SFOAE delays as a function of frequency, have been implemented in an effort to isolate the long-latency (LL) component of SFOAEs—improving the signal-to-noise ratio and the artifact rejection as noteworthy byproducts. In order to test the hypothesis that behavioral measures and SFOAE LL component both reflect the activity of limited cochlear regions, the relationship between SFOAE LL component and behavioral measures was investigated. SFOAEs, behavioral thresholds and psychophysical tuning curves (PTCs) were measured for probe frequencies centered around 0.75, 1, 2, 4, 8, 10, 12.5 and 14 kHz. Stimulus levels for PTCs (10 dB SL) and SFOAEs (10, 20 and 30 dB SL) were referenced to behavioral thresholds at each frequency. Using the same in situ calibration technique using Thevenin-equivalent source parameters, behavioral thresholds and SFOAE-estimated thresholds from input-output functions of the SFOAE LL component were strongly correlated. Furthermore, there was a strong correlation between SFOAE-based and psychophysical tuning estimates.