Skip to main content
NCBI home page
The Journal of the Acoustical Society of America logoLink to The Journal of the Acoustical Society of America
. 2013 Aug;134(2):1121–1133. doi: 10.1121/1.4812766

Influence of stimulus parameters on amplitude-modulated stimulus frequency otoacoustic emissionsa

Tiffany A Johnson 1,a), Laura Beshaler 1
PMCID: PMC3745488  PMID: 23927112

Abstract

The present study evaluated the influence of suppressor frequency (fs) and level (Ls) on stimulus-frequency otoacoustic emissions (SFOAEs) recorded using the amplitude-modulated (AM) suppressor technique described by Neely et al. [J. Acoust. Soc. Am. 118, 2124-2127 (2005a)]. Data were collected in normal-hearing subjects, with data collection occurring in two phases. In phase 1, SFOAEs were recorded with probe frequency (fp) = 1, 2, and 4 kHz and probe levels (Lp) ranging from 0 to 60 dB sound pressure level (SPL). At each fp, Ls ranged from Ls = Lp to Ls = Lp + 30 dB. Additionally, nine relationships between fs and fp were evaluated, ranging from fs/fp = 0.80 to fs/fp = 1.16. Results indicated that for low suppressor levels, suppressors higher in frequency than fp (fs > fp) resulted in higher AM-SFOAE levels than suppressors lower in frequency than fp (fs < fp). At higher suppressor levels, suppressors both higher and lower in frequency than fp produced similar AM-SFOAE levels, and, in many cases, low-frequency suppressors produced the largest response. Recommendations for stimulus parameters that maximize AM-SFOAE level were derived from these data. In phase 2, AM-SFOAEs were recorded using these parameters for fp = 0.7-8 kHz and Lp = 20-60 dB SPL. Robust AM-SFOAE responses were recorded in this group of subjects using the parameters developed in phase 1.

INTRODUCTION

Of the various types of evoked otoacoustic emissions (OAEs), distortion product otoacoustic emissions (DPOAEs), transient-evoked otoacoustic emissions (TEOAEs), and stimulus-frequency otoacoustic emissions (SFOAEs), SFOAEs may be the simplest to interpret. SFOAEs occur in response to a single pure-tone stimulus, a simpler and more frequency-specific stimulus than what is used to elicit either DPOAEs or TEOAEs. Despite the apparent simplicity in the eliciting stimulus, recording SFOAEs is more technically challenging than recording DPOAEs or TEOAEs. The SFOAE response overlaps in both time and frequency with the tone used to elicit it. Thus, separating the response from the stimulus involves more complex signal processing than is required to record DPOAEs or TEOAEs. Recently, Kalluri and Shera (2007) illustrated that, when careful attention is paid to the measurement technique, SFOAEs recorded using any of the commonly used approaches yield similar estimates of the SFOAE response in individual ears. This finding suggests that SFOAEs are not measurement artifacts but reflect cochlear processing. Not surprisingly, numerous investigators have used SFOAEs to investigate various aspects of cochlear function (e.g., Brass and Kemp, 1993; Schairer et al., 2003; Schairer et al., 2006; Shera and Guinan, 2003; Schairer and Keefe, 2005; Keefe et al., 2008; Bergevin et al., 2012; Lichtenhan, 2012).

Despite the relative success with which SFOAEs have been used to assess cochlear function, less effort has focused on translating SFOAEs into a clinical tool for identifying and diagnosing auditory pathology. Although data have been reported from studies where SFOAEs were recorded in ears with normal hearing (NH) and in ears with hearing loss (e.g., Konrad-Martin et al., 2002; Konrad-Martin and Keefe, 2005), few have recorded SFOAE responses in large groups of normal and impaired ears. Data from these large-scale studies are necessary to describe the accuracy with which SFOAEs could be used to identify differences in responses from normal and impaired ears. Currently, the only large-scale studies in both normal and impaired ears are those described by Avan et al. (1991) and Ellison and Keefe (2005). General applicability of the Avan et al. findings is limited because they reported results for a limited range of frequencies (all ≤1000 Hz). Ellison and Keefe reported data for a broader range of frequencies, and their data suggest that SFOAEs may have clinical utility beyond their primary use as a research tool. However, Ellison and Keefe explored a broad parametric space for each subject to determine the probe-suppressor combination resulting in the largest SFOAE response in each ear. This necessitated several hours of data collection, spread over multiple days, a time commitment that is not possible under clinical constraints. To be a viable clinical tool, appropriate stimulus parameters need to be determined in advance so that measurements can be made in a short period of time. It is unclear how well SFOAEs would have performed in the Ellison and Keefe study if a fixed set of stimulus parameters had been used for all subjects.

In addition to the generally encouraging results obtained by Ellison and Keefe (2005), there has been some speculation that SFOAEs may be useful in clinical applications because the cochlear mechanisms contributing to the SFOAE response are different from those contributing to other types of evoked OAEs, particularly DPOAEs. Currently, there are believed to be two fundamentally different cochlear mechanisms by which OAEs are generated (e.g., Shera and Guinan, 1999). These mechanisms include a coherent-reflection mechanism, arising from a reflection source, and a nonlinear distortion mechanism, arising from a distortion source. DPOAEs are believed to include contributions from both mechanisms (e.g., Talmadge et al., 1998; Talmadge et al., 1999; Shera and Guinan, 1999), while SFOAEs primarily may arise via a coherent-reflection mechanism (Shera and Guinan, 1999, 2003; Goodman et al., 2003; Shairer et al., 2006; Shera et al., 2008; Choi et al., 2008). Shera and Guinan (1999) have speculated that the two cochlear mechanisms underlying OAE generation may depend on different aspects of cochlear mechanics for their generation. Additional studies where SFOAEs and other types of OAEs are recorded in the same subject would help to explain if useful knowledge regarding auditory pathology can be obtained by comparing responses from OAEs arising through different cochlear mechanisms.

Exploration of these clinical questions requires an efficient and reliable technique for recording SFOAEs. Neely et al. (2005a) recently described a technique for recording SFOAEs using an amplitude-modulated (AM) suppressor that may be more efficient than other approaches to recording SFOAEs. This technique is a variation of the classic two-tone suppression approach to extracting the SFOAE response from the pure-tone stimulus (e.g., Kemp and Chum, 1980; Brass and Kemp, 1993; Keefe, 1998). In this approach, an AM suppressor is presented in conjunction with a probe tone that is either fixed or modulated in level. Details of this technique will be described further in Sec. 2. Briefly, the AM-suppressor interacts with the SFOAE elicited by the probe stimulus resulting in AM in the SFOAE response. The AM of the SFOAE response is then detectable in sidebands that occur at frequencies just higher and just lower than the probe frequency (see Fig. 1). Spectral techniques can then be used to estimate the SFOAE response present in these sidebands.

Figure 1.

Figure 1

Open in a new tab

Demonstration of the AM-SFOAE response in an acoustic cavity and a NH ear. In both columns, fp = 2 kHz and fs = 2.16 kHz. Left column: Spectra of the AM-SFOAE stimuli and response for Lp = 40 dB SPL and Ls = 54 dB SPL (measured level, nominal level = 60 dB SPL; see text for the distinction between measured and nominal levels). The gray shading near the bottom of each panel represents the maximum noise level observed for the frequency interval 1992–2008 Hz (i.e., the interval encompassing fp and the spectral sidebands). Evidence of AM in the suppressor is apparent in the sidebands surrounding fs, which are present in both the cavity and in the NH ear. Evidence of the suppressor modulating the response to the probe is present in the NH ear where large sidebands representing the AM-SFOAE response are present at fp ± 8 Hz. The sidebands at fp ± 8 Hz are not seen in the acoustic cavity. Right column: Data plotted as input/output functions. The open symbols represent the response level in each sideband, the filled circles represent the response level after summing the sidebands, and the solid lines represent the associated noise levels (3 lines are plotted, one for each sideband and one representing the average sideband noise).

Potential gains in efficiency for the AM-SFOAE technique relative to other approaches come from the fact that the suppressor and probe are presented simultaneously in a single recording interval. In contrast, the more common implementation of the suppression technique involves multiple recording intervals, some containing the probe stimulus presented alone and others containing the probe in combination with a suppressor (e.g., Brass and Kemp, 1993; Souter, 1995; Keefe, 1998; Kalluri and Shera, 2007; Keefe et al., 2008). In this approach, the SFOAE is obtained by subtracting the intervals with the suppressor from the intervals without the suppressor, essentially requiring multiple measurements to determine the response for a single condition. The AM-SFOAE technique does not require separate recordings for a probe alone and a probe plus suppressor condition.

Another potential benefit of the AM-SFOAE technique is that the use of spectral techniques to estimate the response allows for the exclusion of some non-cochlear, physiologic noise sources. For example, Neely et al. (2005a) showed evidence of spectral components that were believed to arise from a physiologic process such as the heartbeat. These biologic (non-OAE) components could be excluded from the AM-SFOAE response because their frequency content differs from the frequencies where the AM-SFOAE response occurs.

These factors suggest that the AM-SFOAE approach may be an efficient technique for recording SFOAE responses, particularly if stimulus parameters yielding robust responses are known in advance. The parametric space has been thoroughly investigated for DPOAEs (e.g., Gaskill and Brown, 1990; Harris et al., 1989; Kummer et al., 1998; Neely et al., 2005b; Johnson et al., 2006) but has received less attention in the SFOAE literature. Neely et al. (2005a) demonstrated the feasibility of the AM-SFOAE technique but did not explore the suppressor-probe parametric space. Therefore the optimal frequency and level relationships for the probe and AM-suppressor tones are not known.

The purpose of the present study was to define stimulus parameters for recording AM-SFOAEs using the Neely et al. (2005a) method with the idea that these predefined stimulus parameters could be used in future investigations into the clinical applications of AM-SFOAEs. More specifically, the goals of the present project were (1) to explore the influence of manipulating the suppressor frequency and suppressor level on the AM-SFOAE response recorded in NH ears, (2) to develop stimulus parameters that are expected to produce robust AM-SFOAE responses in NH ears, and (3) to evaluate the extent to which these stimuli generalized by recording AM-SFOAEs using these parameters in a new group of NH subjects.

METHODS

Subjects

Data were collected from one ear of 40 NH adult subjects (30 females, 10 males). Subjects ranged in age from 18 to 28 yr (mean = 22 yr). NH was defined as pure-tone behavioral thresholds ≤20 dB hearing level (HL) (re: ANSI, 2004) for the octave frequencies from 250 to 8000 Hz. Pure-tone behavioral thresholds were evaluated on the first day of data collection. Subjects were required to have a normal 226-Hz tympanogram in the test ear prior to each test session. Data collection occurred in two phases (described in the following text) with 20 subjects (15 females, 5 males) participating in each phase. During data collection, subjects reclined in a comfortable chair and were free to sleep, read, or watch a silent, captioned movie. Phase 1 data collection required approximately 7-8 h of participation from each subject; data collection was spread over three to four 2-h sessions. Phase 2 data collection was completed in a single 1.5-2 h session for each subject. The subjects participating in phase 2 differed from those participating in phase 1.

Procedures

Equipment and calibration

All AM-SFOAE data were collected using custom-designed software (emav; Neely and Liu, 1993) that controlled a 24-bit sound card (CardDeluxe, Digital Audio Labs) housed in a PC. Stimuli were presented, and responses were recorded through an ER-10C (Etymotic Research) probe-microphone system. The ER-10C was modified to remove 20 dB of attenuation from each channel. As a result, it was possible to achieve stimulus levels as high as 80 dB sound pressure level (SPL) in each subject's ear canal. Stimulus levels were calibrated in situ in sound pressure at the plane of the ER-10C probe. Equipment and calibration approaches were identical for the two phases of data collection.

The AM-SFOAE technique

In the AM-suppressor approach used here, an AM suppressor is presented in conjunction with a probe tone that is fixed in level. This differs from the original description of the AM-SFOAE technique (Neely et al., 2005a), where the probe tone was modulated slowly in level. Modulation of probe level is not a critical feature of the technique. In other words, to implement the technique, the suppressor must be AM, but the probe can remain fixed in level as was done in the present study.

The suppressor frequency (fs) and level (Ls) are chosen so that the suppressor will interact with the cochlear response to the probe tone (the SFOAE) causing the SFOAE to alternate between suppression and no suppression at a rate equal to the suppressor modulation frequency (mf). In other words, the suppressor produces AM in the SFOAE and, as a result, the SFOAE response will be apparent in spectral sidebands present at frequencies both higher and lower than the probe frequency (fp). Specifically, the sideband components will be present at fp ± mf. Spectral techniques can then be used to estimate the SFOAE response present in these sidebands.

Stimulus conditions

The stimulus conditions differed across the two phases. Phase 1 primarily was a stimulus-development phase where the influence of a range of fs and Ls conditions on the AM-SFOAE level was evaluated. Phase 2 primarily was a stimulus-validation phase where the stimulus parameters developed during phase 1 were tested in a new group of subjects.

During phase 1, AM-SFOAE responses were recorded in the form of input–output (I/O) functions for probe levels (Lp) ranging from 0 to 60 dB SPL in 10-dB steps. The I/O functions were recorded at 3 fp's: 1, 2, and 4 kHz. The influence of fs and Ls on the level of the AM-SFOAE was explored by recording responses to nine different suppressor-probe frequency ratios (fs/fp) and four different suppressor-probe level relationships (Ls, Lp). The stimulus conditions are summarized in Table TABLE I.. The test stimuli included conditions where fs was lower in frequency than fp (fs/fp < 1.0) and conditions where fs was higher in frequency than fp (fs/fp > 1.0). For each fp and fs/fp combination, the Ls, Lp relationship ranged from one where Ls was equal to Lp (Ls = Lp + 0) to one where Ls was 30 dB higher than Lp (Ls = Lp + 30). In all cases, the suppressor was 100% AM at a rate of 8 Hz (i.e., mf = 8 Hz).

TABLE I.

Summary of stimulus conditions for phase 1.

fp (kHz) Lp (dB SPL) fs/fp Ls, Lp
1, 2, 4 0-60 0.80, 0.84, 0.88, 0.92, 0.96 Ls = Lp+0
  10-dB steps 1.04, 1.08, 1.12, 1.16 Ls = Lp+10
      Ls = Lp+20
      Ls = Lp+30
Open in a new tab

It should be noted that our software (emav) sets signal level for both the probe and suppressor stimuli before AM is added to the suppressor. With the addition of AM, the average level of the suppressor is lower than it would be if it were unmodulated. Therefore the recorded levels for the suppressor tone were lower than the nominal levels used as the target during the calibration stage. For example, if the nominal (target) level for Ls was 40 dB SPL, the recorded level was lower (approximately 34 dB SPL) because the recorded level includes the reduction in energy that occurs with the addition of AM. Unless otherwise noted, the levels reported for the suppressor tones in this manuscript are the nominal levels. For cases where Ls was 30 dB higher than Lp, the maximum level of the suppressor was limited to 80 dB SPL (nominal level). This was done to limit the exposure of the subjects to high levels of sound and to minimize the contribution of system distortion (discussed in the following text) to the recorded response.

During phase 2, AM-SFOAE responses were again recorded as I/O functions with Lp ranging from 20 to 60 dB SPL in 5-dB steps. Here, fp ranged from 0.7 to 8 kHz in half-octave steps. During this phase, Ls and fs/fp were set to those values identified in phase 1. Specific information regarding the Ls and fs/fp values used in phase 2 are described in Sec. 3.

Measurement-based stopping rules

During data collection, emav alternately stored 0.25-s samples of the recorded response in one of two buffers. These buffers were summed to provide an estimate of stimulus and response level and were subtracted to provide an estimate of noise level at the same frequencies. Measurement-based stopping rules were used during both phases of data collection. Using these rules, averaging continued until the noise level was ≤ −25 dB SPL or until 32 s of artifact-free averaging was completed, whichever occurred first. Because averaging time was allowed to vary according to the noise level for a given condition, it was possible to obtain more consistent and lower noise levels across conditions and across subjects. For the purposes of the measurement-based stopping rules, noise level was estimated from the fp – 8 Hz frequency bin (i.e., from the lower frequency sideband in the AM-SFOAE response).

Quantification of AM-SFOAE response and noise levels

The spectrum of an AM-SFOAE stimulus and response is plotted in the left column of Fig. 1, where component level (dB SPL) is plotted as a function of frequency (Hz) for data collected in an acoustic cavity (IEC 711 coupler, top panel) and in a NH ear (bottom panel). In both the acoustic cavity and the NH ear, spectral components are seen at 2.16 kHz, which correspond to fs, along with sidebands ±8 Hz re: fs that reflect the 100% AM of fs. Likewise, spectral components are observed at 2 kHz in both the acoustic cavity and in the NH ear, which corresponds to fp. In the NH ear, spectral sidebands with levels of approximately −1 dB SPL and signal-to-noise ratios (SNRs) of nearly 30 dB are evident at fp ± 8 Hz. In contrast, for the acoustic cavity, the spectral sidebands at fp ± 8 Hz have levels of approximately −30 dB SPL and SNRs ≤0 dB. The presence of sidebands with positive SNRs in the NH ear, and their absence in the acoustic cavity, provides evidence that the suppressor produces AM of the SFOAE in the NH ear but that there was no interaction of the suppressor and probe in the measurement system, at least for these conditions.

The growth of sideband level as a function of Lp is plotted in the right column of Fig. 1, for the acoustic cavity (upper panel) and the NH ear (lower panel). In the acoustic cavity, the sideband level (open circles) is relatively constant across Lp, with levels between −20 and −30 dB SPL observed for most conditions. Noise levels (solid lines) in the cavity were typically less than −25 dB SPL for all Lp. In the NH ear, sideband level emerged from the noise for Lp > 10 dB SPL, reaching a maximum of nearly 9 dB SPL for Lp = 60 dB SPL. The difference in response behavior for the NH ear compared to the acoustic cavity was taken as further evidence that, for these conditions, there was no interaction between the suppressor and probe in the measurement system.

In the NH ear, the response levels in the two sidebands were within 1 dB of each other for all cases where the response had emerged from the noise. Evaluation of the responses from other ears and for other conditions suggested no systematic difference in response behavior between the two sidebands. As a result, the sidebands were summed (using vector summation) to yield a single estimate of SFOAE level. In the right column of Fig. 1, the summed response is plotted as filled circles. As expected, the summed response level is approximately 6 dB higher than the response in an individual sideband. Deviations from the expected 6-dB increase for the summed response were observed when either the level or the phase differs between the two sidebands, an occurrence that was common in the cavity and when the response was not well separated from the noise. To derive a single estimate of the noise, the noise levels observed at the two sideband frequencies were averaged to yield a single value. In all subsequent figures and in all further analyses, only the summed response levels and averaged noise levels are considered.

RESULTS

System distortion and valid emissions

In a linear measurement system, there should be no interaction between the probe and the suppressor, regardless of the frequencies or levels tested. The data shown in Fig. 1 for the acoustic cavity suggest that for some frequencies and levels, this assumption is true. However, it appears that the assumption was not valid for the measurement system across all conditions evaluated. To quantify the dynamic range of the system, a modified version of the approach for quantifying system distortion described by Dorn et al. (2001) was used. Here responses were recorded in four cavities: (1) An IEC 711 coupler, (2) a standard 2-cm3 coupler, (3) a 10-m PVC tube (internal diameter = 8 mm), and (4) an open syringe with the plunger in close proximity to the ER-10C probe tip (enclosed volume of air approximately 0.1 cm3). The IEC 711 and standard 2-cm3 couplers were included as cavities that provide an approximation of an average human ear. The syringe and 10-m PVC tube were included to represent extremes of cavities sizes. The syringe represents a smaller cavity than would be encountered in a human ear but represents a case where high signal levels could be achieved for low driving voltages. Conversely, the 10-m PVC tube represents a cavity much larger than the human ear where higher driving voltages were necessary to achieve target signal levels. These cavities were intended to span the range of acoustic conditions that might be expected across a group of subjects and would allow insight into those conditions where nonlinearities (non-biologic contributions) from the measurement system were present in the measured response. The response levels recorded in these cavities were taken as an indication of the level of system distortion. Data were collected in each of these four cavities for the stimulus combinations described in Table TABLE I..

As shown in Fig. 1, for many conditions, nonlinear contributions did not emerge from the noise floor. For other conditions, however, nonlinear contributions were observed that equaled the responses observed in NH ears. An example of this is shown in Fig. 2. Here, response level is plotted as a function of Lp. The response in the NH ear (bottom panel) suggests a robust SFOAE response that emerges from the noise floor and grows to a maximum of approximately 12 dB SPL for probe levels between 50 and 60 dB SPL. Examination of the response in the IEC 711 coupler (middle panel) suggests a nearly identical growth function with response levels in the coupler falling within 1 dB of the levels observed in the NH ear for Lp = 40 to 60 dB SPL. For these conditions, it appears that the response recorded in the NH ear may be largely (or entirely) non-biologic in origin. Examination of the results recorded by connecting the soundcard DAC directly to the ADC (top panel) suggest the origin of the nonlinearities primarily resides in the ER-10C transducers as the response level at the soundcard output never exceeds −25 dB SPL, even for Lp = 60 dB SPL. This pattern of findings is similar to what was reported by Neely et al. (2005a) in the original description of the AM-SFOAE technique. Neely et al. attributed the nonlinearities to those arising through intermodulation distortion in the ER-10C. Cross-talk in the ER-10C could also contribute to artifacts in the measured response, although cross-talk in the ER-10C would be expected to be a factor at frequencies higher than the frequency where this measurement was made (Siegel, 2007).

Figure 2.

Figure 2

Open in a new tab

Input/output functions recorded with fp = 1 kHz, fs/fp = 1.04, and Ls = Lp + 30 dB. Because the maximum stimulus level was limited to 80 dB SPL, when Lp = 60 dB SPL, a lower Ls was used (Ls = Lp + 20). Data in the top panel were recorded by connecting the DAC output of the sound card directly to the ADC input of the sound card; data in the middle panel were recorded in the IEC 711 coupler; and data in the bottom panel were recorded in a NH ear. The filled symbols represent the summed sideband response (the AM-SFOAE) and the solid line represents the average noise floor for the two sidebands.

Figure 2 represents one of the most extreme examples of system distortion observed in the data; however, non-biologic energy that exceeded the noise floor was observed for other combinations of fs, fp, and Ls, Lp. Data representing the range of system distortion observed in the data are plotted in Fig. 3 as cumulative distribution functions. The data plotted here suggest that system distortion decreases as fp increases from 1 to 4 kHz and that system distortion increases as suppressor level increases from top to bottom in the figure. Although in many cases, it is not possible to distinguish the functions across the different fs/fp's (see, for example, the upper right panel), in other cases, the level of system distortion varied across fs/fp. Specifically, when fp = 1 and 2 kHz, fs/fp = 0.96 and 1.04 produced higher levels of system distortion than other ratios. This is evident in Fig. 3 in that the cumulative distribution functions corresponding to these ratios (thickest lines) were associated with the highest system distortion levels at each cumulative percentile when fp = 1 and 2 kHz. For the condition where fp = 1 kHz and Ls= Lp+30 (bottom left panel), fs/fp = 0.92 and 1.08 also produced higher system distortion levels than was observed with other ratios. These findings are consistent with the idea that higher levels of system distortion were observed when fs and fp were more closely spaced compared to conditions where they were more broadly spaced, particularly when high Ls were used.

Figure 3.

Figure 3

Open in a new tab

Cumulative distributions for system distortion measured in the four acoustic cavities described in the text. As indicated on the figure, each column represents a different fp and each row represents a different Ls, Lp relationship, with Ls increasing from top to bottom. Within each panel, the individual curves represent the data recorded for a given fs/fp as Lp is increased to a maximum of 60 dB SPL. Additionally, each curve represents the data recorded across all four cavities for a given combination of fp and fs/fp. Cases where fs/fp < 1.0 are plotted with dashed lines, and cases where fs/fp> 1.0 are plotted as solid lines. Line thickness is used to indicate the spacing between fs and fp. The thickest lines indicate the narrowest spacing (fs/fp = 0.96 and 1.04) with decreasing thickness indicating broader spacing. Also shown in each panel is a vertical reference line at −20 dB SPL, which marks the point above which system distortion levels would exceed the biologic noise floor observed in ears.

Although the influence of Lp on these data is not apparent in Fig. 3, the conditions associated with the highest levels of system distortion (the 100% point on the cumulative distribution function) were those recorded with the highest Lp (50 or 60 dB SPL). In other words, in all cases, the highest levels of system distortion were observed when Lp was at or near its maximum level, which was when Ls also was at its maximum level. Please note that due to the decision to limit the maximum stimulus level used during data collection in ears to 80 dB SPL, no data were collected with Lp = 60 for the Ls = Lp + 30 conditions. In some cases, this produced the impression that less system distortion was obtained for the Ls = Lp + 30 condition than for the Ls = Lp + 20 condition (compare the maximum system distortion recorded with fp = 2 kHz in the bottom 2 rows). The apparently lower levels of maximum system distortion for the highest suppressor levels is a result of the decision to not test Lp = 60 dB SPL for this condition.

It should be noted that the biologic noise floor in ears varied across fp but was generally between −20 and −30 dB SPL. Therefore the system distortion would not be expected to influence the SFOAE level recorded in ears until it exceeded −20 to −30 dB SPL. The system distortion values recorded in the acoustic cavities were frequently lower than −20 to −30 dB SPL; however, the data plotted in Fig. 3 suggest that system distortion exceeding the biologic noise floor may be present under some conditions, particularly when recording with high Lp and Ls and when fs and fp are closely spaced.

Given these observations, inclusion criteria were developed and data were required to meet these criteria to be considered primarily biologic in origin. Before applying these inclusion criteria to the data collected in phase 1, the mean AM-SFOAE response level for the 20 subjects participating in this phase was computed. Likewise, the mean noise level across the 20 subjects was calculated. After averaging the data across the 20 subjects, the following inclusion criteria were applied. First, the mean response level in ears must exceed the mean noise level in ears by 6 dB (the SNR must exceed 6 dB). Second, for those cases where the SNR was >6 dB, the response must also exceed the maximum system distortion level observed across all four cavities by 6 dB (the signal-to-distortion ratio, SDR, must exceed 6 dB). The SNR criterion was always invoked first because the SDR is of little interest for responses that are in the noise floor, i.e., these responses are not interpretable regardless of their origin, biologic or otherwise. Responses meeting these inclusion criteria were considered to be “valid” responses, where the average response in NH ears exceeded the noise floor by more than 6 dB and also exceeded the maximum response in cavities by more than 6 dB. More stringent inclusion criteria were used when making the determination of stimulus parameters for use when recording AM-SFOAEs. These more stringent criteria are described in Sec. 3C.

Influence of suppressor frequency and level

The influence of fs and Ls on AM-SFOAE level is plotted in Figs. 45. In Fig. 4, mean AM-SFOAE and noise levels (and standard errors) were obtained by averaging the responses obtained in the 20 subjects participating in phase 1 (15 females and 5 males). Data plotted on this figure represent cases where both the SNR and SDR were >6 dB. Missing data are observed in most panels. In the case of the lowest Lp's (0–10 dB SPL), data typically were excluded for not meeting the SNR criterion. In contrast, at higher Lp's and at narrow fs, fp spacing (i.e., fs/fp = 0.96 and 1.04), data more frequently were excluded for failing to meet the SDR criterion. These missing-data trends can be seen in the leftmost column (fp = 1 kHz) in the Ls = Lp + 20 panel. Here data are plotted for Lp = 10–60 dB SPL. No data are shown for Lp = 0 dB SPL because for all fs/fp's, this Lp resulted in AM-SFOAEs with SNR <6 dB. Likewise, in this panel, data points are missing for fs/fp = 0.96 and 1.04 when Lp = 50 and 60 dB SPL. These data were excluded for not meeting the SDR inclusion criterion.

Figure 4.

Figure 4

Open in a new tab

AM-SFOAE level is plotted as a function of fs/fp for the three fp's evaluated in phase 1. Suppressor level increases from top to bottom in the figure as indicated within each panel. The symbols represent mean AM-SFOAE level (±1 standard error). The parameter in each panel is Lp, decreasing in 10-dB increments from a maximum of 60 dB SPL, indicated with decreasing line thickness and change of symbol (as indicated on the figure). The dashed vertical line in each panel represents the division between conditions where fs < fp and fs > fp. See text for discussion of missing data points in this figure. Note that maximum stimulus level was limited to 80 dB SPL. Because of this limitation in maximum level, when Lp = 60 dB SPL, no data were collected for Ls = Lp + 30. Therefore the highest Lp plotted in the bottom row of the figure is 50 dB SPL.

Figure 5.

Figure 5

Open in a new tab

Average AM-SFOAE level as a function of Ls for the three fp's evaluated in phase 1. Each column represents data for a different fp, with Lp increasing from top to bottom in the figure. The parameter within each panel is fs/fp. As in Fig. 3, cases where fs/fp < 1.0 are plotted with dashed lines and cases where fs/fp > 1.0 are plotted as solid lines. Line thickness is used to indicate the spacing between fs and fp. The thickest lines indicate the narrowest spacing (fs/fp = 0.96 and 1.04) with decreasing thickness indicating broader spacing. As in Fig. 4, when Lp = 60 dB SPL, no data were collected with Ls= Lp +30. As a result, data for this condition are not plotted (see bottom row).

For low Ls (top 2 rows), fs/fp influences the level of the recorded SFOAE. For a fixed Lp, suppressors higher in frequency than the probe (fs/fp > 1.0) resulted in higher SFOAE levels than suppressors lower in frequency than the probe (fs/fp < 1.0). However, at higher Ls (bottom 2 rows), for any given Lp, similar AM-SFOAE levels were observed for suppressors both higher and lower in frequency than fp and in some cases fs/fp < 1.0 produced larger AM-SFOAEs.

The influence of Ls and fs on AM-SFOAE level for a fixed Lp is shown in Fig. 5. As was the case for Fig. 4, the data plotted in Fig. 5 represent the mean SFOAE levels for the 20 subjects participating in phase 1. These mean SFOAE levels are plotted as a function of Ls (in dB re: Lp). As in Fig. 4, all data plotted here met the inclusion criteria of SNR >6 dB and SDR >6 dB. The trends shown in Fig. 5 are consistent with those shown in Fig. 4, although the relative influence of Ls and fs at a fixed Lp are seen more easily here. The data are consistent with the idea that for a given Lp, suppressors higher in frequency than the probe were more effective at suppressing the AM-SFOAE response when Ls is low (<20 dB re: Lp). However, for Ls ≥ 20 dB re: Lp, suppressors lower in frequency than the probe result in AM-SFOAE levels that were equal to or slightly higher than those observed with a high-frequency suppressor. For most Lp's, the SFOAE level saturated indicating that a range of fs/fp and Ls combinations are equally effective at suppressing the SFOAE, which results in equal AM-SFOAE levels across multiple stimulus combinations.

Stimulus parameters for recording AM-SFOAEs

The data plotted in Figs. 45 describe the influence of fs and Ls on the mean AM-SFOAE response. Because exploration of this parametric space is time consuming, one goal of this study was to develop a set of stimulus parameters that would be expected to produce robust AM-SFOAE responses in individuals with NH. This was accomplished by first imposing more stringent inclusion criteria on the data to be included in this analysis. Data used for the purpose of identifying optimal stimulus parameters were required to meet the following three criteria: (1) The SNR must exceed 6 dB, (2) the AM-SFOAE level must exceed the maximum level of system distortion measured across all cavities by ≥10 dB, and (3) the AM-SFOAE level must exceed the maximum level of system distortion measured in the “ear-like” cavities (IEC 711 coupler and 2-cm3 coupler) by ≥15 dB. As was done with the data plotted in Figs. 45, the inclusion criteria were applied to the SFOAE data after the SFOAE data had been averaged across subjects. This resulted in the exclusion of the same conditions producing missing data points in Figs. 45 along with several additional conditions. These additional excluded conditions included data recorded with fp = 1 kHz when fs/fp = 0.96 and 1.04 for all Lp and Ls. Data recorded at these ratios with fp = 2 kHz for the highest Ls conditions (Ls = Lp + 20 and Ls = Lp +30) also were excluded. Recall that these were the conditions where the cumulative distribution functions indicated higher system distortion levels than those obtained with other fs/fp's (see Fig. 3). These more stringent criteria were imposed to reduce the likelihood that the conditions producing the highest AM-SFOAE levels included contributions from non-biologic sources.

After the data set was reduced through the more stringent inclusion criteria, the averaged data were examined to identify those suppressor frequencies and levels that resulted in the largest AM-SFOAE responses at each fp and Lp. Because we observed differences in the Ls at which the largest AM-SFOAE responses were observed when fs < fp compared to fs > fp, we determined the Ls associated with the highest AM-SFOAE level for the low-frequency suppressors separately from the high-frequency suppressors. Figure 6 summarizes the data obtained through this analysis. In Fig. 6, the symbols represent the combination of Ls (top row) and fs/fp (bottom row) that produced the highest average AM-SFOAE level for each fp and Lp. For both fs < fp and fs > fp, the relationship between Ls and Lp was statistically significant suggesting that the highest AM-SFOAE levels were obtained when Ls was set higher than Lp and increased as Lp increased according to the best fit equation shown in each panel. As expected from the data plotted in Figs. 45, when fs < fp, the largest AM-SFOAE response required a higher Ls than was required to record the largest response for fs > fp. In contrast to the results suggesting Ls should vary with Lp, the relationship between fs/fp and Lp was not statistically significant. This suggests that fs/fp should remain fixed across Lp.

Figure 6.

Figure 6

Open in a new tab

The approach to specifying stimulus parameters for recording AM-SFOAEs. The symbols represent the Ls (top row) and fs/fp (bottom row) associated with the highest AM-SFOAE level at each Lp. The determination of parameters associated with the highest AM-SFOAE level was restricted to conditions where fs < fp in the left column and fs > fp in the right column. The solid lines in each panel represent the best fit line with the results of the regression analysis shown as insets in each panel. Note that the symbols are offset slightly in the figure for clarity. See text for information regarding inclusion parameters that were imposed on the data included in this figure.

Based on this regression analysis, the recommended stimulus parameters for recording AM-SFOAEs are shown in Table TABLE II.. The recommendations for Ls were taken directly from the best fit equations shown in the upper row of Fig. 6. The recommendations for fs/fp were determined by taking the intercepts of the best fit lines in the bottom row of Fig. 6 and rounding to the nearest fs/fp that was used during data collection and that passes through the individual data points plotted in the lower row. In other words, for fs > fp, the intercept of 1.1 was rounded down to 1.08 and not up to 1.12 because 1.08 represents the individual data better than 1.12. The stimulus conditions specified in Table TABLE II. are expected to produce robust AM-SFOAE responses across a range of frequencies. Furthermore, slightly higher response levels might be expected by using the conditions specified for fs < fp than for the conditions specified for fs > fp (see Figs. 45). These predictions were directly evaluated in phase 2.

TABLE II.

Stimulus parameters for recording AM-SFOAEs.

  fs < fp fs > fp
Ls 0.6Lp + 42 0.5Lp + 38
fs/fp 0.88 1.08
Open in a new tab

Stimulus parameters in new subjects

In phase 2, a new group of 20 NH subjects (15 females, 5 males) participated in data collection. Data were collected in this group of subjects for fp = 0.7–8 kHz in half-octave steps with Lp ranging from 20 to 60 dB SPL using the stimulus conditions specified in Table TABLE II.. Figure 7 summarizes the data collected in these subjects. The data plotted in Fig. 7 represent the median and interquartile ranges for AM-SFOAE and noise levels. Also shown in Fig. 7 is a thick slanting line, which represents the estimate of system distortion. This estimate was obtained by recording responses in the four cavities described previously (IEC 711 coupler, 2-cm2 coupler, 10 -m PVC tube, and a syringe) and developing a regression model to predict system distortion levels based on stimulus conditions.

Figure 7.

Figure 7

Open in a new tab

Input/output functions for AM-SFOAE and noise levels recorded using the stimulus parameters developed in Fig. 6 and specified in Table TABLE II.. Probe frequency increases from top to bottom in the figure as indicated within each panel. Solid lines represent median levels, with the dashed lines representing the interquartile range. The thick, solid line within each panel represents the estimate of system distortion.

The data plotted in Fig. 7 suggest we were able to record AM-SFOAEs that exceed both the noise level and the estimate of system distortion across a range of Lp's for the majority of the new group of subjects. Positive SNRs and SDRs were observed across all frequencies, even those not evaluated during phase 1. The stimulus conditions derived in phase 1 appear to have generalized to a new group of subjects. Across most fp's, AM-SFOAE levels recorded using fs/fp = 0.88 were slightly higher than those recorded with fs/fp = 1.08. This observation was tested using repeated-measures analysis of variance (RMANOVA) with three factors, ratio (fs/fp), frequency (fp), and level (Lp). Only response levels recorded for Lp = 40–60 dB SPL were included in the RMANOVA. We excluded the data for Lp < 40 dB SPL to reduce the influence of noise at these lower probe levels where poor (<6 dB) SNRs were observed for many subjects. The results of this analysis are summarized in Table TABLE III.. The results of the RMANOVA suggest that SFOAE level varies significantly with both Lp and fp. These findings are not surprising and are consistent with similar observations made for DPOAEs (e.g., Gorga et al., 1993; Dorn et al., 2001) and for SFOAEs (e.g., Kemp and Chum, 1980; Schairer et al., 2003; Schairer and Keefe, 2005).

TABLE III.

Results of the RMANOVA for AM-SFOAE Level. The asterisk (*) denotes conditions reaching statistical significance (p < 0.05).

Source df F ratio p value
Main effects
*Ratio (fs/fp) 1, 17 46.06 <0.001
*Frequency (fp) 4.3, 72.4 5.07 0.001
*Level (Lp) 2.2, 38 184.84 <0.001
Interactions
*fs/fp and fp 3.8, 63.8 6.63 <0.001
fs/fp and Lp 2.4, 41.5 1.62 0.206
fp and Lp 28, 476 0.82 0.730
fs/fp, fp, and Lp 28, 476 1.486 0.054
Open in a new tab

For the purposes of determining stimulus parameters to be used in future studies with AM-SFOAEs, the significant results for the main effect of fs/fp and for the interaction of fs/fp and fp are of interest. These results suggest that when the data were collapsed across both fp and Lp, higher AM-SFOAE levels were recorded with fs/fp = 0.88 than for fs/fp = 1.08. The significant interaction of fs/fp and fp, however, suggests that this trend was not present across all fp's. Examination of the data in Fig. 7 suggests that AM-SFOAE level was higher when recorded with fs/fp = 0.88 at all fp's except 8 kHz, where AM-SFOAE level for fs/fp = 1.08 was 3.4 dB higher, on average. For all other fp's, the average (across Lp) AM-SFOAE level was higher with fs/fp = 0.88. The higher average level for fs/fp = 0.88 ranged from as little as 0.763 dB when fp = 2 kHz to as much as 3.472 dB when fp = 1 kHz. These data suggest that, for most frequencies, higher AM-SFOAE levels would be expected for fs/fp = 0.88 than for fs/fp = 1.08.

DISCUSSION

The data plotted here extend the original report by Neely et al. (2005a) describing the AM-SFOAE technique by demonstrating the AM-SFOAE response in a larger group of NH subjects across a broader range of stimulus conditions. These data suggest that it is possible to obtain AM-SFOAE responses that exceed the estimates of both noise and system distortion for a range of stimulus frequencies and stimulus levels. We were able to develop a set of stimulus parameters (Ls, Lp and fs/fp relationships) that are expected to result in large SFOAE responses in NH ears. The data reported for phase 2 suggest that these stimulus parameters generalized to a new group of subjects; this suggests that the stimuli could be used in future investigations into clinical applications of AM-SFOAEs using large groups of NH and hearing impaired ears where an efficient protocol is necessary.

Contribution of system distortion to the AM-SFOAE response

All measurement systems have a limited dynamic range. Observations like those shown in Fig. 2, where the cavity response was identical to the response from the NH ear, suggest it is important to define those conditions under which system nonlinearities may contaminate the response. The data shown in Fig. 2 suggest that the major source of nonlinearities is the ER-10C probe-microphone system. Examination of the responses recorded across the four cavities suggests that a large portion of the distortion components arise from the AM suppressor tone. Examination of the spectrum of the AM suppressor indicated the presence of the expected sidebands at fs ± mf, along with additional distortion components. The amplitude of these distortion components grew as suppressor level increased. When fs and fp are closely spaced, such as when fs/fp = 0.96 and 1.04, the energy from the distortion components may overlap the sideband frequencies at which the determination of the SFOAE response is made. These additional distortion components more frequently were observed when recording low- and mid-frequency AM-SFOAEs (fp = 1 and 2 kHz) than when recording higher frequency AM-SFOAEs (fp = 4 kHz), consistent with the trends observed in Fig. 3. They also were more apparent at the highest suppressor levels.

It appears that this source of system distortion may be specific to the AM-SFOAE technique because it was observed for the AM suppressor more frequently than for the probe, which was fixed in level in this implementation of the AM-SFOAE technique. Other suppression approaches that used fixed-levels for both the probe and suppressor would not be expected to be contaminated by this type of system distortion. It appears that when using the AM-SFOAE technique, care must be given to the spacing of the suppressor and probe frequencies to avoid interference from the distortion components introduced to the AM suppressor by the ER-10C system. These distortion components were observed across the three models of the ER-10C that are used in the laboratory. It is not clear if another probe system would be less susceptible to this type of nonlinearity.

Although we have attempted to minimize the influence of system distortion on the AM-SFOAE data through cavity measures and the implementation of inclusion criteria, an additional level of verification would include recording the AM-SFOAE response in NH and hearing-impaired ears to evaluate the extent to which responses differ in these groups. Although it is difficult to fully capture system distortion for all possible acoustic environments, we believe the stringent inclusion criteria used when developing the stimulus parameters eliminated the majority of the non-biologic contributions to the measured response.

Influence of suppressor frequency and level on the AM-SFOAE

The data plotted in Figs. 45 suggest that the AM-SFOAE response is influenced by the frequency and level relationships between the suppressor and probe tones in a manner that is consistent with what would be expected from mechanical two-tone suppression data. For a fixed Lp, the response to fp that is generated by the cochlea (the cochlear SFOAE) is constant. Thus, in Fig. 4, the cochlear SFOAE response for any single curve within each panel is constant because all points on the curve were recorded with a fixed Lp. Likewise, the cochlear SFOAE response within each panel of Fig. 5 is fixed due to the common Lp in each panel. For a fixed Lp, deviations in the measured AM-SFOAE response level, therefore, indicate variations in the extent of interaction between the probe and suppressor responses along the basilar membrane. In other words, although the cochlear SFOAE is constant for a given probe level, measurement of the response depends on the interaction between the suppressor and probe responses; therefore, variability in the measured AM-SFOAE levels for a given probe level must be due to variability in the extent to which the suppressor and probe interact.

For the lowest suppressor levels (Ls = Lp + 0 or 10), the highest AM-SFOAE levels were observed for suppressors that were higher in frequency than the probe (fs/fp > 1.0). This pattern is consistent with mechanical two-tone suppression data that suggest suppression is observed at lower suppressor levels when the suppressor is higher in frequency than the probe compared to when the suppressor is lower in frequency (e.g., Cooper, 1996; reviewed in Robles and Ruggero, 2001). Similar patterns also have been reported for SFOAE data (e.g., Brass and Kemp, 1993; Keefe et al., 2008).

For the higher suppressor levels in Figs. 45 (Ls = Lp + 20 or 30), the AM-SFOAE response recorded for the suppressors lower in frequency than the probe (fs/fp < 1.0) are equal to or, in some cases, larger than those observed with the high-frequency suppressors. This finding suggests that once an effective suppressor level is used, the low-frequency suppressor yielded a more fully suppressed SFOAE. These patterns also were observed in Fig. 7 and were confirmed in the statistical analyses completed for those data. Ruggero et al. (1992) reported similar observations in their data on mechanical two-tone suppression in chinchilla where suppressors that were lower than the characteristic frequency resulted in greater suppression magnitude than suppressors that were higher in frequency. Brass and Kemp (1993) and Keefe et al. (2008) reported a similar pattern in their SFOAE data.

Brass and Kemp (1993) provide a simple explanation of this pattern that is based on consideration of the relative overlap of the probe and suppressor tones along the basilar membrane. When presented at equivalent low levels, a high-frequency suppressor will more fully overlap the probe traveling wave than a low-frequency suppressor. As a consequence, the high-frequency suppressor will provide more complete suppression, particularly in the region just basal to the peak of the probe traveling wave. As suppressor level is increased, however, the low-frequency suppressor becomes more effective due to the basal spread of its tail which produces increasing overlap (and suppression) of the response to the probe. In contrast, for all but the narrowest fs/fp's, a high-frequency suppressor is more restricted in the extent of overlap with the probe because traveling-wave spread in the apical direction is more limited.

Stimulus parameters for recording AM-SFOAEs

The stimulus parameters identified in Table TABLE II. can be used in future work with AM-SFOAEs. Slightly (but significantly) larger response levels are expected for the parameters associated with fs < fp compared to fs > fp. It should be noted that the largest fs/fp evaluated was 1.16. Keefe et al. (2008) have noted that the most sensitive region for suppressors higher in frequency than the probe occurs in the vicinity of fs/fp = 1.19, a slightly larger value than the maximum used in this study. It is possible that the differences in SFOAE level observed for fs < fp compared to fs > fp may have been reduced if we had collected data for fs/fp > 1.16. In any case, for the stimulus parameters described here, slightly larger SFOAE levels would be expected at most frequencies when using fs/fp = 0.88 compared to fs/fp = 1.08.

Data shown in Fig. 7 suggest that the stimulus parameters described in Table TABLE II. generalized to new subjects and are expected to produce robust SFOAEs in NH ears. The validation of these stimulus parameters in a new sample of subjects during phase 2 is an important aspect of the work describe here, suggesting that the stimulus parameters were not idiosyncratic to the ears on which they were developed. Additionally, although the stimulus parameters were based on data collected in ears that primarily were females, examination of the responses recorded in the five males participating in phase 2 suggested that their responses were distributed throughout the interquartile ranges plotted in Fig. 7. These stimulus parameters appear to produce robust responses in both male and female ears.

Based on the results of phase 2, where robust AM-SFOAEs were observed in a new group of subjects, future work could focus on questions related to the accuracy with which AM-SFOAEs identify auditory pathology. Some have speculated that OAEs arising from a single cochlear-source mechanism may more accurately identify auditory pathology than OAEs arising from multiple cochlear-source mechanisms (e.g., Shera and Guinan, 1999; Dhar and Shaffer, 2004; Vetesnick et al., 2009). Although this has been a matter of some debate (compare Siegel et al., 2005 and Shera et al., 2008), low-to-moderate level SFOAEs are believed to primarily arise through the reflection-source mechanism (Shera and Guinan, 1999, 2003; Goodman et al., 2003; Schairer et al., 2006; Choi et al., 2008). The use of suppressors that are within 12-octave of the probe frequency, as is the case for all fs/fp combinations used in the present study, may also help to restrict the cochlear contributions to a single source mechanism (Shera et al., 2004; Kalluri and Shera, 2007). It remains to be seen if SFOAEs could be used to provide any additional information beyond what is currently obtained from DPOAE or TEOAEs.

CONCLUSIONS

AM-SFOAEs can be recorded across a range of frequencies and levels. Variations in suppressor-probe frequency and level relationships influence the AM-SFOAE level in a manner that is consistent with mechanical two-tone suppression data. The stimulus parameters developed in phase 1 of this project generalized to new subjects in phase 2. These stimuli could be used in future investigations with AM-SFOAEs.

ACKNOWLEDGMENTS

This work was supported by the University of Kansas General Research Fund Allocation No. 2302177 and the American Speech-Language-Hearing Foundation. Manuscript preparation was supported by NIH NIDCD Grant No. R03 DC011367. The authors thank Steve Neely for his assistance with the implementation of the AM-SFOAE technique. We also thank Katie Shatzer for assistance with data collection and Elizabeth Stewart and Janna Sorensen for assistance with data collection and figure preparation.

a

Portions of this work were presented at the 2010 Annual Convention of the American Speech-Language-Hearing Association, Philadelphia, PA, and at the 2010 Annual Meeting of the American Auditory Society, Scottsdale, AZ.

References

  1. ANSI (2004). 3.6, Specifications for Audiometers (Acoustical Society of America, New York: ). [Google Scholar]
  2. Avan, P., Bonfils, P., Loth, D., Narcy, P., and Trotoux, J. (1991). “ Qualitative assessment of human cochlear function by evoked otoacoustic emissions,” Hear. Res. 52, 99–112. 10.1016/0378-5955(91)90191-B [DOI] [PubMed] [Google Scholar]
  3. Bergevin, C., Fulcher, A., Richmond, S., Velenovsky, D., and Lee, J. (2012). “ Interrelationships between spontaneous and low-level stimulus-frequency otoaoustic emissions in humans,” Hear. Res. 285, 20–28. 10.1016/j.heares.2012.02.001 [DOI] [PubMed] [Google Scholar]
  4. Brass, D., and Kemp, D. T. (1993). “ Suppression of stimulus frequency otoacoustic emissions,” J. Acoust. Soc. Am. 93, 920–939. 10.1121/1.405453 [DOI] [PubMed] [Google Scholar]
  5. Choi, Y.-S., Lee, S. -Y., Parham, K., Neely, S. T., and Kim, D. O. (2008). “ Stimulus-frequency otoacoustic emission: Measurements in humans and simulations with an active cochlear model,” J. Acoust. Soc. Am. 123, 2651–2669. 10.1121/1.2902184 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cooper, N. P. (1996). “ Two-tone suppression in cochlear mechanics,” J. Acoust. Soc. Am. 99, 3087–3098. 10.1121/1.414795 [DOI] [PubMed] [Google Scholar]
  7. Dhar, S., and Shaffer, L. A. (2004). “ Effects of a suppressor tone on distortion-product otoacoustic emissions fine structure: Why a universal suppressor level is not a practical solution to obtaining single-generator DP-Grams,” Ear Hear. 25, 573–585. 10.1097/00003446-200412000-00006 [DOI] [PubMed] [Google Scholar]
  8. Dorn, P. A., Konrad-Martin, D., Neely, S. T., Keefe, D. H., Cyr, E., and Gorga, M. P. (2001). “ Distoriton product otoacoustic emission input/output functions in normal-hearing and hearing-impaired ears,” J. Acoust. Soc. Am. 110, 3119–3131. 10.1121/1.1417524 [DOI] [PubMed] [Google Scholar]
  9. Ellison, J. C., and Keefe, D. H. (2005). “ Audiometric predictions using stimulus-frequency otoacoustic emissions and middle ear measurements,” Ear. Hear. 26, 487–503. 10.1097/01.aud.0000179692.81851.3b [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gaskill, S. A., and Brown, A. M. (1990). “ The behavior of the acoustic distortion product, 2f1-f2, from the human ear and its relation to auditory sensitivity,” J. Acoust. Soc. Am. 88, 821–829. 10.1121/1.399732 [DOI] [PubMed] [Google Scholar]
  11. Goodman, S. S., Withnell, R. H., and Shera C. A. (2003). “ The origin of SFOAE microstructure in the guinea pig,” Hear. Res. 183, 7–17. 10.1016/S0378-5955(03)00193-X [DOI] [PubMed] [Google Scholar]
  12. Gorga, M. P., Neely, S. T., Bergman, B., Beauchaine, K. L., Kaminski, J. R., Peters, J., and Jesteadt, W. (1993). “ Otoacoustic emissions from normal-hearing and hearing-impaired subjects: Distortion product responses,” J. Acoust. Soc. Am. 93, 2050–2060. 10.1121/1.406691 [DOI] [PubMed] [Google Scholar]
  13. Harris, F. P., Lonsbury-Martin, B. L., Stagner, B. B., Coats, A. C., and Martin, G. K. (1989). “ Acoustic distortion products in humans: Systematic changes in amplitude as a function of f2/f1 ratio,” J. Acoust. Soc. Am. 85, 220–229. 10.1121/1.397728 [DOI] [PubMed] [Google Scholar]
  14. Johnson, T. A., Neely, S. T., Garner, C. A., and Gorga, M. P. (2006). “ Influence of primary-level and primary-frequency ratio on human distortion product otoacoustic emissions,” J. Acoust. Soc. Am. 119, 418–428. 10.1121/1.2133714 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kalluri, R., and Shera, C. A. (2007). “ Comparing stimulus-frequency otoacoustic emissions measured by compression, suppression, and spectral smoothing,” J. Acoust. Soc. Am. 122, 3562–3575. 10.1121/1.2793604 [DOI] [PubMed] [Google Scholar]
  16. Keefe, D. H. (1998). “ Double-evoked otoacoustic emissions. I. Measurement theory and nonlinear coherence,” J. Acoust. Soc. Am. 103, 3489–3498. 10.1121/1.423057 [DOI] [PubMed] [Google Scholar]
  17. Keefe, D. H., Ellison, J. C., Fitzpatrick, D. F., and Gorga, M. P. (2008). “ Two-tone suppression of stimulus frequency otoacoustic emissions,” J. Acoust. Soc. Am. 123, 1479–1494. 10.1121/1.2828209 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kemp, D. T., and Chum, R. A. (1980). “ Observations on the generator mechanism of stimulus frequency acoustic emissions—Two tone suppression,” in Psychophysical Physiological and Behavioural Studies in Hearing, edited by Brink G. V. D. and Bilsen F. A. (Delft University Press, Delft: ), pp. 34–42. [Google Scholar]
  19. Konrad-Martin, D., and Keefe, D. H. (2005). “ Transient-evoked stimulus-frequency and distortion-product otoacoustic emission in normal and impaired ears,” J. Acoust. Soc. Am. 117, 3799–3815. 10.1121/1.1904403 [DOI] [PubMed] [Google Scholar]
  20. Konrad-Martin, D., Neely, S. T., Keefe, D. H., Dorn, P. A., Cyr, E., and Gorga, M. P. (2002). “ Sources of DPOAEs revealed by suppression experiments, inverse fast Fourier transforms, and SFOAEs in impaired ears,” J. Acoust. Soc. Am. 111, 1800–1809. 10.1121/1.1455024 [DOI] [PubMed] [Google Scholar]
  21. Kummer, P., Janssen, T., and Arnold, W. (1998). “ The level and growth behavior of the 2f1-f2 distortion product otoacoustic emission and its relationship to auditory sensitivity in normal hearing and cochlear hearing loss,” J. Acoust. Soc. Am. 103, 3431–3444. 10.1121/1.423054 [DOI] [PubMed] [Google Scholar]
  22. Lichtenhan, J. T. (2012). “ Effects of low-frequency biasing on otoacoustic and neural measures suggest that stimulus-frequency otoacoustic emissions originate near the peak region of the traveling wave,” J. Assoc. Res. Otolaryngol. 13, 17–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Neely, S. T., Johnson, T. A., Garner, C. A., and Gorga, M. P. (2005a). “ Stimulus-frequency otoacoustic emissions measured with amplitude-modulated suppressor tones,” J. Acoust. Soc. Am. 118, 2124–2127. 10.1121/1.2031969 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Neely, S. T., Johnson, T. A., and Gorga, M. P. (2005b). “ Distortion-product otoacoustic emission measured with continuously varying stimulus level,” J. Acoust. Soc. Am. 117, 1248–1259. 10.1121/1.1853253 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Neely, S. T., and Liu, Z. (1993). emav: Otoacoustic Emission Averager, Technical Memorandum 17 (Boys Town National Research Hospital, Omaha, NE: ). [Google Scholar]
  26. Robles, L., and Ruggero, M. A. (2001). “ Mechanics of the mammalian cochlea,” Physiol. Rev. 81, 1305–1352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ruggero, M. A., Robles, L., and Rich, N. C. (1992). “ Two-tone suppression in the basilar membrane of the cochlea: Mechanical basis of auditory-nerve rate suppression,” J. Neurophysiol. 68, 1087–1099. [DOI] [PubMed] [Google Scholar]
  28. Schairer, K. S., Ellison, J. C., Fitzpatrick, D., and Keefe, D. H. (2006). “ Use of stimulus-frequency otoacoustic emission latency and level to investigate cochlear mechanics in human ears,” J. Acoust. Soc. Am. 120, 901–914. 10.1121/1.2214147 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Schairer, K. S., Fitzpatrick, D., and Keefe, D. H. (2003). “ Input-output functions for stimulus-frequency otoacoustic emissions in normal-hearing adult ears,” J. Acoust. Soc. Am. 114, 944–966. 10.1121/1.1592799 [DOI] [PubMed] [Google Scholar]
  30. Schairer, K. S., and Keefe, D. H. (2005). “ Simultaneous recording of stimulus-frequency and distortion product otoacoustic emission input-output functions in human ears,” J. Acoust. Soc. Am. 117, 818–832. 10.1121/1.1850341 [DOI] [PubMed] [Google Scholar]
  31. Shera, C. A., and Guinan, J. J., Jr. (1999). “ Evoked otoacoustic emissions arise by two fundamentally different mechanisms: A taxonomy for mammalian OAEs,” J. Acoust. Soc. Am. 105, 782–798. 10.1121/1.426948 [DOI] [PubMed] [Google Scholar]
  32. Shera, C. A., and Guinan, J. J., Jr. (2003). “ Stimulus-frequency-emission group delay: A test of coherent reflection filtering and a window on cochlear tuning,” J. Acoust. Soc. Am. 113, 2762–2772. 10.1121/1.1557211 [DOI] [PubMed] [Google Scholar]
  33. Shera, C. A., Tubis, A., and Talmadge, C. L. (2008). “ Testing coherent reflection in chinchilla: Auditory-nerve responses predict stimulus-frequency emissions,” J. Acoust. Soc. Am. 123, 381–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Shera, C. A., Tubis, A., Talmadge, C. L., and Guinan, J. J., Jr. (2004). “ The dual effect of “suppressor” tones on stimulus-frequency otoacoustic emissions,” Assoc. Res. Otolaryngol. Abstr. 27, 181. [Google Scholar]
  35. Siegel, J. H. (2007). “ Calibrating otoacoustic emission probes,” in Otoacoustic Emissions – Clinical Applications, Third Edition, edited by Robinette M. S. and Glattke T. J. (Thieme Medical Publishers, New York: ), pp. 403–427. [Google Scholar]
  36. Siegel, J. H., Cerka, A. J., Recio-Spinoso, A., Temchin, A. N., van Dijk, P., and Ruggero, M. A. (2005). “ Delays of stimulus-frequency otoacoustic emissions and cochlear vibratio contradict the theory of coherent reflection filtering,” J. Acoust. Soc. Am. 118, 2434–2443. 10.1121/1.2005867 [DOI] [PubMed] [Google Scholar]
  37. Souter, M. (1995). “ Stimulus frequency otoacoustic emissions from guinea pig and human subjects,” Hear. Res. 90, 1–11. 10.1016/0378-5955(95)00124-9 [DOI] [PubMed] [Google Scholar]
  38. Talmadge, C. L., Long, G. R., Tubis, A., and Dhar, S. (1999). “ Experimental confirmation of the two-source interference model for the fine structure of distortion product otoacoustic emissions,” J. Acoust. Soc. Am. 105, 275–292. 10.1121/1.424584 [DOI] [PubMed] [Google Scholar]
  39. Talmadge, C. L., Tubis, A., Long, G. R., and Piskorski, P. (1998). “ Modeling otoacoustic emission and hearing threshold fine structures,” J. Acoust. Soc. Am. 104, 1517–1543. 10.1121/1.424364 [DOI] [PubMed] [Google Scholar]
  40. Vetesnik, A., Turcanu, D., Dalhoff, E., and Gummer, A. W. (2009). “ Extraction of sources of distortion product otoacoustic emissions by onset-decomposition,” Hear. Res. 256, 21–38. 10.1016/j.heares.2009.06.002 [DOI] [PubMed] [Google Scholar]

Articles from The Journal of the Acoustical Society of America are provided here courtesy of Acoustical Society of America

RESOURCES