Reducing reflected contributions to ear-canal distortion product otoacoustic emissions in humans

Abstract

Distortion product otoacoustic emission (DPOAE) fine structure has been attributed to the interaction of two cochlear-source mechanisms (distortion and reflection sources). A suppressor presented near the 2f1-f2 frequency reduces the reflection-source contribution and, therefore, DPOAE fine structure. Optimal relationships between stimulus and suppressor conditions, however, have not been described. In this study, the relationship between suppressor level (L3) and stimulus level (L2) was evaluated to determine the L3 that was most effective at reducing fine structure. Subjects were initially screened to find individuals who produced DPOAE fine structure. A difference in the prevalence of fine structure in two frequency intervals was observed. At 2 kHz, 11 of 12 subjects exhibited fine structure, as compared to 5 of 22 subjects at 4 kHz. Only subjects demonstrating fine structure participated in subsequent measurements. DPOAE responses were evaluated in 1/3-octave intervals centered at 2 or 4 kHz, with 4 subjects contributing data at each interval. Multiple L3's were evaluated for each L2, which ranged from 20 to 80 dB SPL. The results indicated that one or more L3's at each L2 were roughly equally effective at reducing DPOAE fine structure. However, no single L3 was effective at all L2's in every subject.

Figure 1

DPOAE level (dB SPL) as a function of f₂ (kHz) for a single subject for the 1/3-octave interval surrounding f₂ = 2 kHz. Increasing line thickness indicates increasing L₂, which ranged from 20 to 80 dB SPL in 10-dB increments. Each trace has been offset by successive 10-dB increments to aid in visualization.

Figure 2

The number of subjects screened in each frequency interval who produced DPOAE fine structure, which was defined as a maxima-to-minima ratio ≥ 10 dB in one cycle of fine structure when L₂ = 30 dB SPL. The open bars represent the number of subjects evaluated and the cross-hatched region indicates the number who had DPOAE fine structure.

Figure 3

DPOAE level (dB SPL) as a function of f₂ (kHz) for all subjects screened in both the 2- and 4-kHz intervals. Data for f₂ ≈ 2 kHz are shown in the 1^st and 3^rd columns. Data for f₂ ≈ 4 kHz are shown in the 2^nd and 4^th columns. Adjacent panels within a row represent data from the same subjects, as identified in the upper right corner of each panel. All data in this figure were collected with L₂ = 30 dB SPL in the absence of a suppressor tone.

Figure 4

Inverse fast Fourier transforms (IFFTs) of the data shown in Fig. 3. The data are plotted as amplitude (relative units) as a function of time (ms). The layout is the same as in Fig. 3. Data for f₂ ≈ 2 kHz are shown in the 1^st and 3^rd columns, f₂ ≈ 4 kHz are shown in the 2^nd and 4^th columns, adjacent panels within a row are for the same subject.

Figure 5

Summary of the analyses completed on the data. The data shown in this figure are for a single subject (C110), with L₂ = 30 dB SPL, and f₂ ≈ 2 kHz. DPOAE responses are shown in the left column along with the DC component (overall DPOAE level) from the discrete cosine transform (DCT) of the data (see text for details regarding the DCT). The level of the higher “frequency” components of the DCT for the same data are shown in the middle column, and an inverse fast Fourier transform (IFFT) of the data in the left column is shown in the right column. In each row, the response in the control condition (no f₃) along with results for two suppressor levels (indicated by C value) is shown. Suppressor level increases from top to bottom in the figure.

Figure 6

The left column plots a summary of the discrete cosine transform (DCT) of the data shown in Fig. 5. The right column plots similar data from a different subject (C117) for the interval surrounding 4 kHz when L₂ = 30 dB SPL. The top row plots the DC component level, which corresponds to the overall DPOAE level, as a function of suppressor condition (C). The bottom row represents the rms level of the higher “frequency” components of the DCT (see middle column Fig. 5) and represents the magnitude of the fine structure in the DPOAE response for the various suppressor conditions. See the text for additional information regarding the derivation of these data.

Figure 7

Summary of the mean (± 1 standard deviation) discrete cosine transform (DCT) results across the 4 subjects. The left column plots data for the 2-kHz interval, with data for the 4-kHz interval shown in the right column. Data for L₂ = 20, 40, 60, and 80 dB SPL are shown, as indicated on the figure. For each L₂, the upper panel plots DC level (dB SPL) as a function of suppressor condition and the lower panel plots the rms level (dB_rms) of the fine structure. See the text for additional information regarding the derivation of these data.

Figure 8

. The relationship between suppressor level and L₂ in individual subjects. Data for the 2-kHz interval are shown in the left column and data for the 4-kHz interval are shown in the right column. Each row represents an alternate view of suppressor level (L₃). The top row plots C ( L₃ = 0.75L₂ + C ) as a function of L₂. The middle row plots absolute suppressor level (dB SPL) as a function of L₂, with the bottom row plotting suppressor level re: L₂ as a function of L₂. The solid slanting lines represent linear fits to these data; the fits are shown as insets in each panel. The r² values associated with these fits are also shown in each panel. The circles filled with asterisks in the right column highlight data from an individual subject, as indicated in the text.