Distortion-product otoacoustic emissions in the common marmoset (Callithrix jacchus): parameter optimization

Abstract

Distortion-product otoacoustic emissions (DPOAEs) were measured in a New World primate, the common marmoset (Callithrix jacchus). We determined the optimal primary-tone frequency ratio (f(2)/f(1)) to generate DPOAEs of maximal amplitude between 3 and 24 kHz. The optimal f(2)/f(1), determined by varying f(2)/f(1) from 1.02 to 1.40 using equilevel primary tones, decreased with increasing f(2) frequency between 3 and 17 kHz, and increased at 24 kHz. The optimal f(2)/f(1) ratio increased with increasing primary-tone levels from 50 to 74 dB SPL. When all stimulus parameters were considered, the mean optimal f(2)/f(1) was 1.224-1.226. Additionally, we determined the effect of reducing L(2) below L(1). Decreasing L(2) below L(1) by 0, 5, and 10 dB (f(2)/f(1)=1.21) minimally affected DPOAE strength. DPOAE levels were stronger in females than males and stronger in the right ear than the left, just as in humans. This study is the first to measure OAEs in the marmoset, and the results indicate that the effect of varying the frequency ratio and primary-tone level difference on marmoset DPOAEs is similar to the reported effects in humans and Old World primates.

Figure 1

Frequency-ratio (f₂/f₁) functions showing L_DP (ordinate) vs. f₂/f₁ (abscissa) for each f₂ frequency (f₂ is specified above the individual panels). DPOAEs were obtained with equilevel primaries from 50 – 74 dB SPL (in 6-dB increments) as f₂/f₁ was increased from 1.02 – 1.40 in 0.04 increments (0.02 increments between 1.18 and 1.24). Responses were averaged across all ears (n = 17).

Figure 2

Frequency-ratio (f₂/f₁) functions for each f₂ frequency (axes as in Fig. 1). The DPOAEs shown here were obtained using equilevel primary tones at 62 dB SPL and were averaged across all animals and ears. Error bars are shown to demonstrate the typical amount of inter-subject and inter-ear variability observed in the DPOAE levels and the noise floor.

Figure 3

Mean optimal f₂/f₁ ratio (ordinate) vs. f₂ frequency (abscissa). A) The f₂/f₁ ratios which elicited the most robust mean DPOAEs (averaged across all ears) were considered as the mean optimal f₂/f₁ ratios. B) Right vs. left ear. DPOAEs were averaged only in subjects in which both ears were tested and the mean optimal f₂/f₁ was that which elicited the most robust mean DPOAE (n = 5). C) Males vs. females. DPOAEs were averaged across both ears in males and females and the f₂/f₁ which elicited the most robust mean DPOAE was considered optimal.

Figure 4

Mean optimal f₂/f₁ ratio (ordinate) vs. primary-tone level (abscissa). A) The f₂/f₁ ratios which elicited the most robust DPOAEs for each ear were averaged to determine the mean optimal f₂/f₁. B) Right vs. left ear. DPOAEs were averaged only in subjects in which both ears were tested and the mean optimal f₂/f₁ was that which elicited the most robust mean DPOAE (n = 10 ears). C) Male vs. female. DPOAEs were averaged across both ears in males and females and the f₂/f₁ which elicited the most robust mean DPOAE was considered optimal.

Figure 5

DPgram (L_DP vs. f₂ frequency) for f₂/f₁ values of 1.18 – 1.24. DPOAE levels were averaged across all ears (n = 17). Dashed line represents average noise floor.

Figure 6

I/O Functions (L_DP vs. L₁ Primary Tone) at each f₂ frequency (f₂ indicated above the panel) for each primary-tone level difference. DPOAE levels were averaged across all ears (n = 18).

Figure 7

DPgrams (L_DP vs. f₂ frequency) at each primary-tone level for each difference in primary-tone level. DPOAE levels were averaged across all ears (n = 18).

Figure 8

A) I/O Functions (L_DP vs. L₁ Primary Tone) for males vs. females (data was averaged across all f₂ frequencies). B) DPgrams (L_DP vs. f₂ frequency) for males vs. females (L₁ = L₂ = 62 dB SPL). C) I/O Functions (L_DP vs. L₁ Primary Tone) for right vs. left ear (data was averaged across all f₂ frequencies). D) DPgrams (L_DP vs. f₂ frequency) for right vs. left ear. Effect sizes are shown below the lower of the two compared mean L_DP values.

Figure 9

Slope Functions (Slope vs. f₂ frequency). The slopes of the linear-regression equations characterizing the I/O functions from 50 – 68 dB SPL (averaged across all ears of both sexes). Error bars are shown to indicate the amount of inter-subject and inter-ear variability in the slope of the I/O functions.