1. The objective of this study was to precisely evaluate the relationship between the threshold of neurons in the anteroventral cochlear nucleus (AVCN) and the properties of distortion product otoacoustic emissions (DPOAEs). Response areas of multiunit clusters in the AVCN and DPOAEs in the ear canal were measured alternately in the adult gerbil during furosemide-induced changes of the endocochlear potential. Stimulus frequencies of the probe tones for DPOAE measurement were in the range of f1 = 1.7-7.6 kHz and f2 = 2.0-9.0 kHz; the ratio f2:f1 was always 1.18. Stimulus amplitudes were varied in 5-dB steps from 30 to 80 dB SPL, with either equal amplitudes (L1 = L2) or unequal, with L1 set 10 dB above L2. Multiunit response areas were determined from cluster responses to a series of 100-ms tone bursts presented with a pseudo-random sequence in frequency and intensity. 2. Changes in the multiunit discharge properties after 50-75 mg/kg furosemide injection were as follows: the best frequency (BF) threshold increased from initial values in the range of 20-30 dB SPL to 50-80 dB SPL at 10-20 min postinjection and then recovered fully by 60-90 min. The spontaneous discharge activity decreased to zero before any changes in the frequency threshold curve were observed and did not return to initial values for several hours. Likewise, total discharge rates of stimulus elicited responses were reduced and tended to stay reduced even after BF threshold had fully recovered. 3. From the DPOAE measurements, the changes observed in the cubic distortion tone (CDT, 2f1-f2) emission after furosemide injection were as follows: at high levels of the probe tones, changes in the emission intensities generally stayed within a 10-dB range. The CDT amplitudes for low stimulus levels, however, were typically reduced by up to 40 dB, but recovered (depending on the furosemide dosage) by approximately 60-90 min. 4. At low to moderate stimulus levels of 40-60 dB SPL, there was a near perfect, minute-by-minute covariation of the ear canal CDT amplitude and the BF threshold measured in the AVCN. A 10-dB increase in threshold was associated with a 5- to 7-dB decrease in the CDT emission. 5. The optimum stimulus parameter set for the noninvasive estimation of cochlear performance from the CDT response was for stimulus amplitudes L1 = 50, L2 = 40 dB SPL. 6. This experiment demonstrates that CDT emissions at low stimulus levels are very good predictors of the thresholds of cochlear afferents, but this validity is lost for BF thresholds greater than approximately 60-70 dB SPL. 7. The ear canal CDT amplitude is better correlated with the BF threshold sensitivity of neuronal response areas in the AVCN than with the spontaneous discharge rate or absolute above-threshold discharge rates.
Reducing reflected contributions to ear-canal distortion product otoacoustic emissions in humans