Download Distortion product otoacoustic emissions provide clues to hearing

Distortion product otoacoustic emissions provide clues
to hearing mechanisms in the frog eara)
Pantelis N. Vassilakisb) and Sebastiaan W. F. Meenderinkc)
Department of Physiological Science, University of California at Los Angeles, Los Angeles,
California 90095-1606
Peter M. Narins
Departments of Physiological Science and Organismic Biology, Ecology, and Evolution, University
of California at Los Angeles, 621 Charles E. Young Drive, South, Los Angeles, California 90095-1606
共Received 5 March 2004; revised 11 August 2004; accepted 12 September 2004兲
2 f 1 - f 2 and 2 f 2 - f 1 distortion product otoacoustic emissions 共DPOAEs兲 were recorded from both
ears of male and female Rana pipiens pipiens and Rana catesbeiana. The input-output 共I/O兲 curves
obtained from the amphibian papilla 共AP兲 of both frog species are analogous to I/O curves recorded
from mammals suggesting that, similarly to the mammalian cochlea, there may be an amplification
process present in the frog AP. DPOAE level dependence on L1 -L2 is different from that in
mammals and consistent with intermodulation distortion expectations. Therefore, if a mechanical
structure in the frog inner ear is functioning analogously to the mammalian basilar membrane, it
must be more broadly tuned. DPOAE audiograms were obtained for primary frequencies spanning
the animals’ hearing range and selected stimulus levels. The results confirm that DPOAEs are
produced in both papillae, with R. catesbeiana producing stronger emissions than R. p. pipiens.
Consistent with previously reported sexual dimorphism in the mammalian and anuran auditory
systems, females of both species produce stronger emissions than males. Moreover, it appears that
2 f 1 - f 2 in the frog is generated primarily at the DPOAE frequency place, while 2 f 2 - f 1 is generated
primarily at a frequency place around the primaries. Regardless of generation place, both emissions
within the AP may be subject to the same filtering mechanism, possibly the tectorial
membrane. © 2004 Acoustical Society of America. 关DOI: 10.1121/1.1811571兴
PACS numbers: 43.80.Lb, 43.64.Jb 关JAS兴
I. INTRODUCTION
A. Background—definitions
Distortion product otoacoustic emissions 共DPOAEs兲
arise when the ear is stimulated acoustically by two sinusoidal signals 共primaries兲 with appropriately chosen frequencies ( f 1 , f 2 ) and stimulus levels (L1 , L2 ) 共e.g., Kemp, 1979;
Probst et al., 1991兲. In human subjects, otoacoustic emissions 共OAEs兲 in general are often used for clinical diagnostic
screening of frequency-dependent cochlear function.
DPOAEs may be used for early screening of outer hair cell
共OHC兲 damage 共e.g., Lonsbury-Martin et al., 1993; Stover
et al., 1996; Wagner and Plinkert, 1999兲, for monitoring inner ear function after exposure to ototoxic drugs and/or noise
共e.g., Brown et al., 1989; Emmerich et al., 2000兲, for diagnosis of middle ear damage 共e.g., Owens et al., 1992, 1993;
Zhang and Abbas, 1997兲, and may also be used for assessing
OHC maturation in premature babies 共e.g., Abdala, 2000兲. In
addition, OAEs have provided noninvasive means of exama兲
Portions of this work were presented at the 146th meeting of the Acoustical
Society of America 关P. Vassilakis and P. M. Narins, J. Acoust. Soc. Am.
114, 2414 共2003兲兴.
b兲
Present address: School of Music, De Paul University, 804 West Belden
Avenue,
Chicago,
Illinois 60614-3296.
Electronic
mail:
[email protected]
c兲
Present address: Department of Otorhinolaryngology and Head and Neck
Surgery, University Hospital Maastricht, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands.
J. Acoust. Soc. Am. 116 (6), December 2004
Pages: 3713–3726
ining inner ear mechanisms in mammals 共e.g., Maison et al.,
1997; Shera and Guinan, 2003兲 and other vertebrates 共e.g.,
Rosowski et al., 1984兲, as well as in insects 共e.g., Kössl and
Boyan, 1998兲. The present study uses DPOAE recordings
from two frog species to examine hearing mechanisms in the
frog ear.
The vast majority of physiological and psychophysical
studies involving DPOAEs concentrate on the 2 f 1 - f 2 distortion product, principally because it usually represents the
strongest DPOAE and therefore the easiest one to measure.
OAEs, including DPOAEs, originate in the inner ear and, in
most mammals, have been linked to outer hair cell 共OHC兲
motility and receptor potential 共Probst et al., 1991; Robles
and Ruggero, 2001兲. More specifically, the shape of mammalian DPOAE input/output 共I/O兲 curves 共i.e., compressive
growth at low stimulus levels, although not necessarily as
compressive as in the case of basilar membrane and hair cell
I/O curves兲 provides further evidence that may link DPOAE
generation to the cochlear amplifier, a process hypothesized
to account for the observed increased sensitivity of the mammalian ear at low stimulus levels and for the mammalian
ear’s compressive and saturating response as stimulus levels
increase from low to intermediate. Zheng et al. 共2000兲
showed that prestin is the motor protein related to OHC motility, while Liberman et al. 共2002兲 demonstrated a direct
coupling between electrically stimulated OHC motility and
cochlear amplification. In general, DPOAE generation in
mammals has also been associated with the overlap of the
0001-4966/2004/116(6)/3713/14/$20.00
© 2004 Acoustical Society of America
3713
primary tones’ basilar membrane 共BM兲 disturbances.
DPOAEs have been recorded from the ears of all terrestrial vertebrate classes, including most frog species tested.1
The presence of OAEs in general and DPOAEs in particular
in the frog raises numerous questions with regards to emission origin, especially due to the fact that the frog inner ear
has neither a BM 共Capranica, 1976兲 nor morphologically distinct outer and inner hair cells 共Lewis, 1977; Lewis et al.,
1982兲. Based on observations that direct current injection
may influence both the frequencies and amplitudes of OAEs
共Wit et al., 1989兲, Probst et al. 共1991兲 suggested that frog
OAEs may be related to the electrical tuning of hair cells.
Others 共e.g., Van Dijk et al., 2002兲 have suggested that
OAEs in frogs may be linked to spontaneous hair bundle
movement, which may be analogous to the previously mentioned OHC motility in mammals and has been shown to
occur in the bullfrog sacculus 共Martin and Hudspeth, 1999兲.
Hair bundle movement has been observed in vitro on several
nonmammalian vestibular organs 共review in Hudspeth, 1997兲
and the hearing organ of turtles 共Crawford and Fettiplace,
1985兲, and in vivo on the bobtail lizard 共Manley et al., 2001兲.
Recently, hair bundle motility was also observed in rats
共Kennedy et al., 2003兲.
The frog ear responds to airborne sound via two separate
hearing organs, the amphibian papilla 共AP兲 and the basilar
papilla 共BP兲. The AP nerve fibers are tuned to low and
middle frequencies while the BP fibers are tuned to high
frequencies 共Feng et al., 1975兲. More specifically, in R. p.
pipiens the AP responds best to frequencies approximately
within the range 100–1250 Hz, while the BP responds best
to frequencies approximately within the range 1250–2400
Hz, with a ‘‘break’’ in the frequency response of the two
papillae at approximately 1250 Hz 共Ronken, 1991兲. In R.
catesbeiana, the AP range is approximately 100–1100 Hz
and the BP range is approximately 1100–1700 Hz, with the
frequency response border at approximately 1100 Hz 共Feng
et al., 1975; Lewis et al., 1982兲. The present study reveals
similarities in the behavior of DPOAEs originating in the
frog AP and the mammalian cochlea, supporting analogies
between these two organs. Several such analogies and distinctions have already been pointed out 共Lewis and Narins,
1999兲, guiding the present study.
B. Exploration of the stimulus parameter space
The results of parametric studies 共e.g., Whitehead et al.,
1995a, b; Knight and Kemp, 2000; Kummer et al., 2000;
Mills, 2002; Schneider et al., 2003兲 have revealed a complex
interrelationship between DPOAE levels and the fourdimensional primary-tone parameter space and have guided
mammalian models of DPOAE generation 共e.g., Shera and
Guinan, 1999; Talmadge et al., 1998, 1999; Tubis et al.,
2000; Knight and Kemp, 2000; Fahey et al., 2000; Lukashkin et al., 2002; Mills, 2002兲. The precise relationship may
not be the same for all mammals, but it exhibits several
distinguishing trends. As an example, a generally accepted
protocol leading to optimal DPOAE generation within the
frequency ranges of interest in clinical DPOAE measurements includes the following
3714
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
共a兲
共b兲
共c兲
1.2⭐ f 2 / f 1 ⭐1.25.2 Shera and Guinan 共1999兲, Faulstich
and Kössl 共2000兲, and others have argued that this
value must be at least f 2 / f 1 ⫽1.15, to avoid phase
complications. In human subjects, f 2 / f 1 ⫽1.22 results
in the strongest 2 f 1 - f 2 emissions 共Probst et al., 1991兲.
L1 ⭐65 dB SPL. For most mammals, DPOAE I/O
curves show saturation at stimulus levels around 60–70
dB SPL 共Probst et al., 1991兲, followed by a DPOAE
level decrease.
5 dB⭐L1 -L2 ⭐10 dB, for f 2 / f 1 ⬵1.2 共e.g., Whitehead
et al., 1995a, b; Kummer et al., 2000; Fahey et al.,
2000兲. As primary-tone levels become higher and/or
frequency ratio becomes smaller, this level difference
becomes increasingly unnecessary for the maximization of DPOAE levels.
In frog DPOAE studies, the selection of primary tone parameters has in general been less systematic. For example, Van
Dijk et al. 共2002兲 used f 2 / f 1 ⫽1.1 and L1 ⫽L2 ⫽90 dB SPL.
Van
Dijk
and
Manley
共2001兲
used
1.05⭐ f 2 / f 1 ⭐1.5 and 37⭐L1 ⫽L2 ⭐85 dB SPL, but only reported the relationship between DPOAE audiograms and
f 2 / f 1 . In the absence of both a full parametric study and a
theoretical model of DPOAE generation customized to the
amphibian ear, parameter choices have been based on mammalian theoretical grounds, on DPOAE data from studies on
other nonmammals, or on a trial-and-error basis.3
The present work provides a more systematic exploration of the primary-tone parameter space in frogs. DPOAE
amplitudes were recorded from two ranid frog species in a
series of experiments that examine the dependence of
DPOAE amplitude on the absolute frequencies and on the
absolute and relative levels of the primaries. In addition, we
examined the effects of frog species and sex on DPOAE
levels, controlling for possible effects of the degree of anesthesia.
II. MATERIALS AND METHODS
A. Stimulus parameter space
The dependence of frog DPOAE levels on primary-tone
frequency ratio 共explored extensively in mammals, e.g.,
Knight and Kemp, 2000; Schneider et al., 2003兲 was not
addressed in the present study and the primary-tone frequency ratio was set at f 2 / f 1 ⫽1.15 for all experiments. In
the absence to date of a parametric study examining the relationship between DPOAEs and f 2 / f 1 in the frog, this value
was chosen based on the previously mentioned recommendation ( f 2 / f 1 ⭓1.15) and to facilitate comparison with previous frog research 共Van Dijk et al., 2002: f 2 / f 1 ⫽1.1).4
More specifically, the study addresses the following primarytone parameters:
共a兲
Frequency range: For the DPOAE audiograms, the
primary-tone frequencies were selected based on the
tuning characteristics of the R. p. pipiens and R. catesbeiana ears, spanning the frequency range of each papilla 共AP and BP兲 and at frequency steps approximately
equal to or less than 10% of f 1 .
Vassilakis et al.: Emissions and frog hearing mechanisms
共b兲
共c兲
Primary-tone absolute levels: DPOAE I/O curves were
obtained for equal-level primaries. Highest DPOAE
levels and lowest DPOAE thresholds have in general
been measured for primaries within the frequency
ranges associated with the most sensitive hearing regions of the species examined 共Probst et al., 1991兲. Additionally, there is a close relationship between auditory threshold and EOAE levels 共Wagner and Plinkert,
1999兲. Therefore, for the DPOAE I/O curves, f 1 was
selected so that all frequencies of interest ( f 1 , f 2 ,
2 f 1 - f 2 , 2 f 2 - f 1 ) fell near the center of each papilla’s
frequency range.
Primary-tone relative levels DPOAE audiograms were
also obtained for unequal levels of the primaries
(0⭐ 兩 L1 -L2 兩 ⭐10 dB) from R. p. pipiens at 300⭐ f 1
⭐1600. The frequency range examined was selected
based on the tuning characteristics of the R. p. pipiens’
AP.
B. Subjects and procedure
DPOAEs were recorded from both ears of
共a兲
共b兲
Ten northern leopard frogs 共R. p. pipiens兲, five males
共body mass 25.04 –54.31 g, snout-vent length 6.08 –
8.08 cm, cranial width 2.09–3.05 cm, tympanic membrane diameter 0.41–0.53 cm兲 and five females 共body
mass 26.59–56.02 g, snout-vent length 6.94 –7.61 cm,
cranial width 2.32–2.57 cm, tympanic membrane diameter 0.41–0.52 cm兲 and
Ten bullfrogs 共R. catesbeiana兲, five males 共body mass
267–370 g, snout-vent length 15.29–17.41 cm, cranial
width 5.24 –5.95 cm, tympanic membrane diameter
1.63–2.16 cm兲 and five females 共body mass 300– 435
g, snout-vent length 14.76 –17.55 cm, cranial width
4.82– 6.02 cm, tympanic membrane diameter 1.21–
1.48 cm兲.
DPOAE amplitudes are thought to be independent of stimulation side 共i.e., left versus right ears兲 and hearing loss in one
ear of a subject does not affect emission levels in the other
共Probst et al., 1991; Kastanioudakis et al., 2003兲. Therefore,
each ear tested in our study was considered a separate data
point.
The same subjects participated in all experiments, performed in a single session per subject, lasting 4 –9 h. Animals were anaesthetized with an intramuscular injection of
pentobarbital sodium solution 共Nembutal, Abbott Laboratories, 50 mg/ml: ⬃0.9– 1.0 ␮ l/g body mass兲 and were covered with wet gauze that was regularly moistened with water
to facilitate cutaneous respiration. Measurements began between approximately 1.5 and 3 h after Nembutal injection,
with the animals deeply anesthetized 共no toe-pinch or eye
reflex and no acoustic artifacts—i.e., breathing/swallowing
noise—in the recorded signal兲. Some of the DPOAE measurements on R. p. pipiens were repeated 5– 6 h later under
light anesthesia 共all reflexes had returned and there were
acoustic artifacts present in the microphone signal兲 to control
for possible effects of the degree of anesthesia on emission
levels.5
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
FIG. 1. DPOAE I/O curves (2 f 1 - f 2 ) from the AP ( f 1 ⫽600 Hz) of male
共-䉱-兲 and female 共-䊐-兲 R. catesbeiana. Average DPOAE levels and standard
errors 共ten ears each sex: both ears of five males and five females兲. Although
there are level differences between male and female DPOAEs, the overall
shape of the I/O curves is similar regardless of subjects’ sex.
C. Equipment and signal analysis parameters
Measurements were performed in a single-wall soundattenuating chamber with the subjects placed on a vibration
isolation table 共Newport VH Isostation兲. Emissions were recorded with a probe assembly 共ER-10C, Etymotic Research,
Elkgrove, IL兲 that includes a sensitive microphone connected
to a preamplifier 共amplification: 20 dB兲 and two miniature
speakers. The ER-10C probe is designed and calibrated for
use with human subjects. Since the frog species examined
have no ear canals, the probe ear-tip was placed inside a
plastic tube, which was coupled to the frog ear by placing its
open end adjacent to the skin around the tympanic membrane. The small gap between the plastic tube and the frog’s
head was sealed using high vacuum grease, a silicone lubricant. The integrity of the acoustic seal was tested using a
sweep tone 共WG1 waveform generator, Tucker-Davis Technologies, Alachua, FL兲 to ensure that sound level loss at 250
Hz due to diffraction was ⭐5 dB, relative to the level measured at 1000 Hz. Because of the large differences in tympanic membrane diameter between R. p. pipiens and R.
catesbeiana and between male and female R. catesbeiana,
three different ‘‘ear canal’’ tubes were used: 共1兲 R. p. pipiens:
diameter⬵0.95 cm, length⬵1.20 cm, volume⬵0.68 cm3 ;
共2兲 male R. catesbeiana: diameter⬵2.20 cm, length
⬵0.55 cm, volume⬵2.10 cm3 , and 共3兲 female R. catesbeiana:
diameter⬵1.25 cm,
length⬵1.40 cm,
volume
⬵1.72 cm3 .
Primary frequencies ( f 1 ) were selected within the range
240–3000 Hz6 and stimuli were generated using a Real Time
Processor 共TDT-RP2兲, controlled by software written in Matlab 共version 12, MathWorks, Inc., Natick, MA兲 and RPvds
共version 5.0, TDT, Alachua, FL兲. The stimuli were fed to the
two miniature speakers on the probe via two attenuators
共TDT-PA4兲, used for manual control of the signal levels. The
Vassilakis et al.: Emissions and frog hearing mechanisms
3715
FIG. 2. DPOAE I/O curves for f 2 / f 1 ⫽1.15 from R. p. pipiens 共-䉱-, f 1 ⫽800 Hz, 1800 Hz兲 and R. catesbeiana 共-䊐-, f 1 ⫽600 Hz, 1500 Hz兲. Average DPOAE
levels and standard errors 共20 ears each species: both ears of five males and five females兲. 共a兲 and 共b兲 2 f 1 - f 2 ; 共c兲 and 共d兲 2 f 2 - f 1 ; 共a兲 and 共c兲 AP; 共b兲 and
共d兲 BP.
probe’s microphone signal was fed to a frequency analyzer
共SRS-SR770 FFT Network Analyzer, Stanford Research Systems, Sunnyvale, CA兲.7 The bandwidth of the Fourier analysis changed as a function of f 1 . 8 For each stimulus tone-pair,
the DPOAE level reported was the maximum amplitude of
five adjacent analysis bands centered at the band containing
the DPOAE frequency. The noise level was calculated by
averaging the levels of 18 analysis bands surrounding the
five bands used to determine the DPOAE level. These
DPOAE- and noise-level calculation methods minimized the
influence of spectral leakage and gave noise levels comparable to system distortion (⬃83 dB below the primaries for
L1 ⫽L2 ⭓70 dB SPL and ⬃⫺12 dB SPL for L1 ⫽L2
⬍70 dB SPL兲, but may have overestimated DPOAE levels
that were near the calculated noise floor. Therefore, any
emission levels in the results that are ⬍3 dB above the estimated noise-floor may be considered indistinguishable from
noise. All levels reported have been corrected to compensate
for the frequency response of the ER-10C microphone.
3716
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
III. RESULTS
A. Experiment 1: DPOAE IÕO curves
DPOAE I/O curves at 2 f 1 - f 2 and 2 f 2 - f 1 were obtained
for f 2 / f 1 ⫽1.15 and 35⭐L1 ⫽L2 (⫾1 dB) ⭐85 dB SPL 共in
2.5-dB steps兲 with f 1 ⫽800 Hz 共AP兲 and 1800 Hz 共BP兲 for R.
p. pipiens and f 1 ⫽600 Hz 共AP兲 and 1500 Hz 共BP兲 for R.
catesbeiana. Although the absolute DPOAE amplitudes were
different for male and female subjects, the overall shapes of
DPOAE I/O curves within each species were similar between
the two sexes 共Fig. 1兲, so data obtained from male and female animals were averaged. Mean levels and standard errors for each distortion product from each papilla and for
both species are displayed in Fig. 2.
In Fig. 2共a兲, the 2 f 1 - f 2 DPOAE I/O curves obtained
from the AP of both R. p. pipiens and R. catesbeiana display
non-monotonic growth. For low primary levels (⬍60 dB
SPL兲, the DPOAE level growth rate is ⬍1 dB/dB 共compressive兲, saturating at primary levels ⬇60 dB SPL, turning
negative, and reaching a notch at primary levels ⬇70 dB
SPL (⬇67 dB SPL for R. catesbeiana; ⬇72 dB SPL for R. p.
Vassilakis et al.: Emissions and frog hearing mechanisms
across different DP-audiograms, indicating that the finestructure is independent of stimulus level. Similarly to Figs.
2共a兲 and 共b兲, Fig. 3 supports the observation that, for primary
frequencies in the AP and stimulus levels ⬍70 dB there is a
compressive growth of DPOAE levels as primary levels increase. This does not appear to be the case for primary frequencies in the BP. More specifically, for both DPOAE components 关 2 f 1 - f 2 : Fig. 3共a兲; 2 f 2 - f 1 : Fig. 3共b兲兴, the L1 , L2
⫽60 dB plot is close to the top of the graph for frequencies
in the AP (⬃⬍900 Hz) but in between the L1 , L2
⫽57.5 dB and L1 , L2 ⫽62.5 dB plots for frequencies in the
BP (⬃1000 Hz).
B. Experiment 2: DPOAE level dependence on
relative levels of primaries
FIG. 3. DPOAE audiograms from R. p. pipiens 共average DPOAE levels
from 20 ears: both ears of five males and five females兲 for f 2 / f 1 ⫽1.15,
300⭐ f 1 ⭐1600 Hz, and 50⭐L1 ⫽L2 ⭐70 dB SPL. Only the AP may be exhibiting compressive DPOAE level growth. 共a兲 2 f 1 - f 2 ; 共b兲 2 f 2 - f 1 .
pipiens兲. At higher primary levels, DPOAEs grow linearly at
a rate ⬎1 dB/dB (⬇1.3– 2.2 dB/dB). The 2 f 2 - f 1 DPOAE
I/O curves obtained from the AP of both frog species 关Fig.
2共c兲兴 show similar trends but with no pronounced notches.
For primary frequencies in the BP frequency range 关Figs.
2共b兲 and 共d兲兴, DPOAE levels are much lower than those recorded from the AP, with the DPOAE I/O curves for primary
levels ⬍70 dB being very close to the noise-floor. For primary levels ⬎70 dB, the slopes of the DPOAE I/O curves
from the BP are similar to those from the AP. Also note that,
in the AP I/O curves for R. catesbeiana, the notch 共or deviation from linear growth兲 is observed at approximately the
same primary tone levels for both distortion products 关Figs.
2共a兲 and 2共c兲; L1 ⫽L2 ⬵67 dB SPL兴. This does not seem to
be the case in the AP I/O curves for R. p. pipiens. Here, the
notch in the 2 f 1 - f 2 I/O curve occurs at L1 ⫽L2 ⬵72 dB SPL,
while the deviation from linear growth in the 2 f 2 - f 1 I/O
curve is observed at L1 ⫽L2 ⬵67 dB SPL.
A portion of the same results 共only R. p. pipiens and
50⭐L1 ⫽L2 (⫾1 dB) ⭐70 dB SPL兲 is displayed in Fig. 3 in
the form of DP audiograms, for 300⭐ f 1 ⭐1600 Hz. All individual DP-audiograms in Fig. 3 show fine-structure, i.e., a
periodic variation in DPOAE amplitude as a function of frequency. The frequencies for which respective maxima and
minima in DPOAE amplitude were found did not change
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
The relationship between DPOAE level and L1 -L2 was
examined in R. p. pipiens 共both ears of five males and five
females兲 for 50⭐L1 ⭐60 dB SPL (L2 ⫽60 dB SPL兲 and
50⭐L2 ⭐60 dB SPL (L1 ⫽60 dB SPL兲, with 0⭐ 兩 L1 -L2 兩
⭐10 dB
共2.5-dB
steps兲,
300⭐ f 1 ⭐1600 Hz,
and
f 2 / f 1 ⫽1.15. Figure 4 displays the mean DPOAE levels for
each L1 , L2 combination at f 1 ⫽800 and 1500 Hz. These
frequencies were selected to represent DPOAEs recorded
from each papilla. Similar responses were observed at other
frequencies 共Fig. 5兲. Again, although the absolute DPOAE
amplitudes were different for male and female subjects, there
were no significant sex differences in the shape/slopes of the
curves describing the relationship between L1 -L2 and
DPOAE levels. Therefore, data obtained from male and female animals were averaged.
Independent of species, sex, and DPOAE (2 f 1 - f 2 or
2 f 2 - f 1 ), strongest emissions were recorded for 兩 L1 -L2 兩 ⫽0
or 2.5 dB. As 兩 L1 -L2 兩 increased, DPOAE amplitudes gradually dropped. Figure 5 displays the same results in the form
of DPOAE audiograms, showing maximum DPOAE levels
for L1 -L2 ⬵0 for most frequencies tested and both 2 f 1 - f 2
and 2 f 2 - f 1 .
C. Experiment 3: DPOAE audiograms
1. AP versus BP; 2 f1-f2 and 2 f2-f1
DPOAE audiograms were obtained from both ears of all
subjects for 240⭐ f 1 ⭐3000 Hz at ⬃0.1f 1 steps, with
L1 ⫽L2 ⫽60 dB SPL, and f 2 / f 1 ⫽1.15. Figure 6 共R. p. pipiens兲 and Fig. 7 共R. catesbeiana兲 display the levels of 2 f 1 - f 2
and 2 f 2 - f 1 DPOAEs as a function of 共a兲 stimulus frequency
f 1 , 共b兲 stimulus frequency f 2 , and 共c兲 distortion product
共DP兲 frequency. Each plot includes a dashed vertical line that
indicates the approximate frequency value separating the frequency ranges of the two papillae. For both species, the frequency separation between the two papillae is based on neural tuning curve characteristics 共R. p. pipiens: 1250 Hz,
Ronken, 1991; R. catesbaiena: 1100 Hz, Feng et al., 1975兲.9
DPOAE data from both sexes were averaged. The error bars
in Figure 9 and 10 are an indication of the variability in the
male and female date.
In both figures, two distinct frequency regions can be
identified where relative maximum emission amplitudes are
found. These frequency regions correspond roughly to the
Vassilakis et al.: Emissions and frog hearing mechanisms
3717
FIG. 4. Average DPOAE levels and
standard errors versus L2 , for
L1 ⫽60 dB 关共a兲 and 共c兲兴, and versus
L1 , for L2 ⫽60 dB 关共b兲 and 共d兲兴 from
R. p. pipiens 共20 ears: both ears of five
males and five females兲. Strongest
emissions were generally recorded for
L1 ⬇L2 . As expected, based on intermodulation distortion properties, the
level of each distortion product was
affected most by changes in the level
of its neighboring 共in frequency兲 primary 关共b兲 and 共c兲兴.
frequency tuning of the AP 共lower frequency region兲 and the
BP 共higher frequency region兲 of each species, respectively.
For the primary level tested 共60 dB SPL兲, the emission levels
recorded from the AP were higher than those recorded from
the BP.
Figure 6 indicates that, for R. p. pipiens, alignment of
the 2 f 1 - f 2 DPOAE audiogram dip and the break in the frequency coverage between the two papillae 共1250 Hz;
Ronken, 1991兲 is best when 2 f 1 - f 2 amplitude is plotted as a
function of distortion product frequency 关Fig. 6共c兲兴. For the
2 f 2 - f 1 DPOAE, on the other hand, closest alignment occurs
when 2 f 2 - f 1 amplitude is plotted as a function of f 2 关Fig.
6共b兲兴. Near alignment is also observed when this distortion
product is plotted as a function of f 1 关Fig. 6共a兲兴. Similar
results were obtained for R. catesbeiana 共Fig. 7兲. All values
reported are averages.
To control for possible effects of the degree of anesthesia on DPOAE amplitudes, the experiment was repeated 5– 6
h later as the anesthesia was wearing off 共eye and toe-pinch
reflexes accompanied by acoustic artifacts—i.e., breathing/
swallowing noise—in the recorded signal兲. ‘‘Light’’ anesthesia measurements were made on all R. p. pipiens for
300⭐ f 1 ⭐1600 Hz, L1 ⫽L2 ⫽60 dB SPL, and f 2 / f 1 ⫽1.15.
3718
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
The mean results are displayed in Fig. 8 and indicate that, for
the anesthetic agent used and the frequencies and levels
tested, degree of anesthesia did not affect DPOAE amplitudes.
2. Female versus male; R. p. pipiens and R.
catesbeiana
Separate analyses of DPOAE data from male and female
R. p. pipiens and R. catesbeiana revealed clear sex differences in DPOAE levels. Figures 9 and 10 compare male and
female DPOAE audiograms, demonstrating that, in both species, over most of the frequency range tested, and for the
levels of primary tones used, female subjects produce higher
emission levels than male subjects. This is more pronounced
in the BP.
With the exception of emissions from the highfrequency end of the BP, R. catesbeiana demonstrate larger
emission level differences between sexes and, as also indicated in Fig. 2, larger emission levels in general.
Vassilakis et al.: Emissions and frog hearing mechanisms
FIG. 5. DPOAE audiograms from R. p. pipiens 共average DPOAE levels
from 20 ears: both ears of five males and five females兲 for 0⭐ 兩 L1 -L2 兩
⭐10 dB, 300⭐ f 1 ⭐1600, and f 2 / f 1 ⫽1.15. At most frequencies tested,
highest DPOAE levels were recorded for L1 -L2 ⬇0. 共a兲 2 f 1 - f 2 ; 共b兲
2 f 2- f 1 .
IV. DISCUSSION
A. DPOAE IÕO curves provide evidence of an AP
amplifier
The 2 f 1 - f 2 I/O curves obtained from the AP of both
frog species tested 关Figs. 1 and 2共a兲兴 have a salient characteristic in common with the DPOAE I/O curves obtained
from mammals: nonlinear, non-monotonic growth. The general shape of I/O curves recorded in mammals is described in
a manner strikingly similar to the one used to describe the
2 f 1 - f 2 DPOAE I/O AP curves obtained in our study 关Fig.
2共a兲兴 and, with the absence of a clear notch, to the one describing the 2 f 2 - f 1 DPOAE I/O AP curves 关Fig. 2共c兲兴. The
slope of mammalian DPOAE I/O curves depends on the
stimulus levels used. For primary levels ⬍60 dB SPL slopes
are ⭐1 dB/dB. For intermediate stimulus levels (L1 ⫽L2
⬇60– 70 dB SPL兲 the growth saturates, sometimes even resulting in a decrease in DPOAE amplitude 共i.e., a notch兲. For
higher stimulus levels (L1 ⫽L2 ⬎70 dB SPL兲 the slope increases again, often to a value ⬎1 dB/dB 共e.g., Probst et al.,
1991兲. In mammals, it has been shown that DPOAE levels in
general 共e.g., Knight and Kemp, 2000兲 and the exact shape
of the DPOAE I/O curves in particular 共e.g., Mills, 2002兲
depend on the distortion product under study, and on the
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
FIG. 6. Average levels of the 2 f 1 - f 2 共-䉱-兲 and 2 f 2 - f 1 共-䊐-兲 DPOAEs
from R. p. pipiens 共20 ears: both ears of five males and five females兲 for
f 2 / f 1 ⫽1.15, L1 ⫽L2 ⫽60 dB SPL (⫾2.5 dB), and 240⭐ f 1 ⭐3000 Hz as a
function of 共a兲 f 1 , 共b兲 f 2 , and 共c兲 DP frequency. The dashed vertical line
marks the ‘‘break’’ in the frequency coverage of the AP and the BP 共1250
Hz兲. For 2 f 1 - f 2 , the dip in the plot aligns best with the vertical line when
DPOAE level is plotted as a function of DP frequency 共c兲. For 2 f 2 - f 1 , the
alignment is best when DPOAE level is plotted as a function of f 2 共b兲,
although there is also near alignment in the plot plotted against f 1 .
relative levels (L1 -L2 ) and frequencies ( f 2 / f 1 ) of the primaries. In our DPOAE I/O experiment, L1 -L2 and f 2 / f 1 were
not varied. The AP data, however, do exhibit differences in
I/O curve shapes between distortion products 关Fig. 2共a兲 versus 2共c兲兴, with only the I/O curve for the 2 f 1 - f 2 DPOAE
exhibiting a clear notch. Our results are in general agreement
with the DPOAE I/O curve slopes measured by Meenderink
and Van dijk 共2004兲 on R. p. pipiens, although, in their results, both the low- and high-level I/O slopes obtained from
the BP appear relatively steep.
Until recently, the main hypothesis 共e.g., in Probst et al.,
1991兲 introduced to explain the DPOAE I/O curve shape
共especially the observed notches兲 in mammals assumes that
there is interference between 2 DPOAE components:10
共a兲
共b兲
a level dependent, nonlinear component, which dominates the recorded DPOAEs when the stimulus levels
are low but saturates for higher stimulus levels and
a monotonic component that, as the stimulus levels inVassilakis et al.: Emissions and frog hearing mechanisms
3719
FIG. 7. Average levels of the 2 f 1 - f 2 共-䉱-兲 and 2 f 2 - f 1 共-䊐-兲 DPOAEs
from R. catesbeiana 共20 ears: both ears of five males and five females兲 for
f 2 / f 1 ⫽1.15, L1 ⫽L2 ⫽60 dB SPL (⫾2.5 dB), and 240⭐ f 1 ⭐3000 Hz as a
function of 共a兲 f 1 , 共b兲 f 2 , and 共c兲 DP frequency. The alignment between the
dip in the audiograms and the frequency coverage border between the AP
and the BP follows the same pattern as in the R. p. pipiens data 共Fig. 6兲.
crease and the level-dependent component saturates,
becomes dominant in the recorded DPOAEs.
The precise nature of these two components and their place
of origin are subjects of ongoing research.
Lukashkin et al. 共2002兲 proposed an alternative hypothesis with subtle but important differences to the previous
one. Based on the observation that cochlear function impairment results in an upward level shift of the DPOAE I/O
curve notch 共as opposed to the downward level shift predicted by the earlier hypothesis兲 they concluded that DPOAE
I/O curve shapes can be accounted for by a single source: a
nonlinear amplifier with saturating I/O characteristics. A
similar hypothesis was introduced by Mom et al. 共2001兲
based on their observation that DPOAEs measured during
complete cochlear ischemia are highly vulnerable to mild
auditory fatigue induced prior to ischemia. Their results indicate that there is no ‘‘passive’’ component in the response
of the mammalian cochlea and support the single-component
model. Whitehead 共1998兲 and clinical data reviewed in Martin 共2002兲 also support such a model.11
The common thread in all explanations is the hypoth3720
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
FIG. 8. DPOAE audiograms from R. p. pipiens 共average DPOAE levels
from 20 ears: both ears of five males and five females兲 for f 2 / f 1 ⫽1.15,
L1 ⫽L2 ⫽60 dB SPL, and 300⭐ f 1 ⭐1600 Hz, recorded from the same animals while heavily and lightly anesthetized. No systematic effect of the
degree of anesthesia on DPOAE levels was observed. 共a兲 2 f 1 - f 2 ; 共b兲
2 f 2- f 1 .
esized presence 共whether on its own or along with some
other passive nonlinearity兲 of a saturating nonlinear inner ear
amplifier 共i.e., cochlear amplifier兲. The DPOAE I/O curves in
Fig. 2共a兲 of the present study, along with the fact that SOAEs
have been recorded from the frog AP, are consistent with the
inner ear amplifier hypothesis, despite the fact that the frog
ear lacks OHCs, believed to be at the center of the mammalian ear’s amplification process.12 Amplification and OAEs,
in the AP of the frog, may be linked to spontaneous movement of the hair cell bundle, which has been shown to occur
in the frog sacculus 共Martin and Hudspeth, 1999兲, while
there is indirect evidence that it may also occur in the mammalian ear 共Kennedy et al., 2003兲. In other words, hair cell
bundle movement in the frog AP may be analogous to OHC
and/or hair bundle motility in the mammalian ear.
For primary frequencies in the BP, DPOAE levels were
much lower than those recorded from the AP. The portions of
the DPOAE I/O curves that are distinguishable from the
noise-floor appear monotonic, implying that, in contrast to
the AP, an amplification mechanism may not be present in
the frog BP. This potential difference is consistent with the
absence of SOAEs in the frog BP 共Van Dijk et al., 1989兲 and
the differential physiological vulnerability of frog DPOAEs
Vassilakis et al.: Emissions and frog hearing mechanisms
FIG. 9. Comparison between DPOAE audiograms from both ears of five
male and five female R. p. pipiens. 共a兲 The level of 2 f 1 - f 2 is plotted as a
function of DP frequency and 共b兲 the level of 2 f 2 - f 1 is plotted as a function
of f 2 共see Fig. 6兲. Female subjects exhibited stronger emissions than males,
especially from the BP.
共Van Dijk et al., 2003兲, and may echo several physiological
differences between the amphibian and basilar papilla. In the
bullfrog ear, for example, the AP has approximately 15 times
as many hair cells and is innervated by approximately three
times as many afferent nerve fibers as the BP 共Geisler et al.,
1964; Lewis et al., 1985兲. At the same time, efferent nerve
fibers innervate the AP but not the BP 共Robbins et al., 1967兲.
The possible relationship between these physiological differences and the observed difference in DPOAE I/O curves,
assumed to reflect amplification-process differences, remains
to be examined. As a general observation, the fact that only
the AP is tonotopically organized 共Lewis et al., 1982; Simmons et al., 1992兲 is consistent with a greater relevance of an
amplification mechanism that would increase this papilla’s
tuning sharpness. In contrast, the BP responds as a band-pass
filter with a single characteristic frequency 共e.g., Van Dijk
and Manley, 2001兲, rendering the development of an amplification and tuning-sharpening mechanism less essential.
In both R. p. pipiens and R. catesbeiana, the peak of the
compressive-with-saturation portion of the DPOAE I/O
curves occurs for L1 ⫽L2 ⬵60 dB SPL. Since in mammals
the amplified or ‘‘active’’ portion of the I/O curves is presumed to be dominant only at low stimulus levels, DPOAE
audiograms 共experiment 3兲 were obtained for L1 and L2 cenJ. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
FIG. 10. Comparison between DPOAE audiograms from both ears of five
male and five female R. catesbeiana. 共a兲 The level of 2 f 1 - f 2 is plotted as a
function of DP frequency and 共b兲 the level of 2 f 2 - f 1 is plotted as a function
of f 2 共see Fig. 7兲. Female subjects exhibited stronger emissions than males.
tered at ⬃60 dB SPL. The relative levels of the primaries
(L1 -L2 ) corresponding to the highest DPOAE levels were
determined in a separate experiment, discussed below.
B. DPOAE amplitude dependence on the relative
levels of primaries
In mammals, the amplitude of each distortion product is
influenced most by level changes in its neighboring 共in frequency兲 primary tone, consistent with intermodulation distortion expectations. However, and contrary to intermodulation distortion expectations, the highest DPOAE levels are
obtained for unequal levels of the primaries 共i.e., L1 ⬎L2 ;
Kummer et al., 2000兲. This asymmetry in primary tone levels has been linked to the asymmetry of basilar membrane
disturbance envelopes and is less critical at low frequencies,
high primary stimulus levels, and low frequency ratios of the
primaries. Our findings in the frog are more consistent with
intermodulation distortion expectations alone, with highest
emission levels recorded for equal levels of the primaries
共Figs. 4 and 5兲.13 The difference between mammalian and
frog data regarding the effect of L1 -L2 on DPOAE levels
suggests that, if a filtering mechanism analogous to the mammalian BM is present in the frog 共i.e., tectorial membrane兲, it
must be more broadly tuned. This notion is supported by the
Vassilakis et al.: Emissions and frog hearing mechanisms
3721
shape of frequency threshold curves 共i.e., tuning curves兲 recorded from frog VIIIth nerve auditory fibers 关e.g., R. catesbeiana in Feng et al. 共1975兲; Eleutherodactylus coqui and
Bombina orientalis in Narins and Hillery 共1983兲; R. p. pipiens in Ronken 共1991兲兴. In both mammals and frogs, tuning
curves display an asymmetry, with relatively shallow lowfrequency limbs and high-frequency abrupt cutoffs, which is
stronger for high-frequency fibers. However, neural tuning
curves recorded from mammals 共e.g., Robles and Ruggero,
2001兲 exhibit narrower tips than those recorded from frogs.
In mammals, the above noted asymmetry, along with several
other properties of auditory nerve responses including the
L1 , L2 DPOAE asymmetry, is thought to reflect corresponding features of BM vibration 共Robles and Ruggero, 2001兲.
Our results 共Figs. 4 and 5兲, along with the difference between
mammalian and frog neural tuning curves, suggest that, although the AP and BP of the frog ear lack a BM, some other
mechanical structure 共possibly the tectorial membrane兲 may
in some respects function analogously to the mammalian
BM, while more broadly tuned. This observation is in agreement with phase data from auditory nerve fiber measurements in the frog 共Hillery and Narins, 1987兲, pointing to the
tectorial membrane–receptor interface as a possible candidate for the mechanical generation of intermodulation distortion products.14 Further evidence supporting this suggestion
is discussed below. No sex differences in the shape/slopes of
the curves describing the relationship between L1 -L2 and
DPOAE levels were observed, suggesting that the mechanisms involved in DPOAE generation may be the same for
both sexes.
C. DPOAE audiograms
1. DPOAEs can be evoked in both papillae
The correspondence between frequency ranges to which
nerve fibers are tuned and the frequency ranges for which
relative maximum DPOAE amplitudes are recorded indicates
that DPOAEs are produced in both papillae present in the
frog inner ear 共Figs. 6 and 7兲. Our findings are in agreement
with previous results on R. p. pipiens 共Van Dijk et al., 2002;
Meenderink and Van Dijk, 2004兲. For the primary level
tested in our DPOAE audiogram experiment 共60 dB SPL兲,
the emission levels recorded from the AP were higher than
those recorded from the BP. Van Dijk et al. 共2002兲 reported
almost equal DPOAE levels from both papillae of R. p. pipiens for primary levels of 90 dB SPL. The difference in the
results of the two studies probably reflects differences in the
primary-tone levels used, an explanation supported by our
study’s DPOAE I/O data 共Fig. 2兲. More specifically, Fig. 2
indicates that, for low primary levels, DPOAE amplitudes
recorded from the AP are higher than those recorded from
the BP while, for high primary levels, the DPOAE amplitudes recorded from the AP and the BP are comparable.
2. Different generation mechanisms for the 2 f1-f2 and
2 f2-f1 distortion products
In mammals, the level of the DPOAE at 2 f 1 - f 2 is typically plotted as a function of f 2 , reflecting very specific theoretical assumptions/hypotheses regarding the generation
3722
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
mechanism and place of origin of the 2 f 1 - f 2 DPOAE. In
studies on frog DPOAEs 共Baker et al., 1989; Van Dijk and
Manley, 2001; Van Dijk et al., 2002兲, DPOAE levels have
been plotted as a function of f 1 , a choice based on convention 共following Köppl et al., 1993兲. Due to the absence of a
theoretical framework for DPOAE generation in the frog ear,
we plotted DPOAE levels in three different ways.
The dashed vertical lines in Figs. 6 and 7 mark the frequency separation between the AP and BP for the species
tested, as determined by neural tuning curves, and are therefore within a frequency range of low sensitivity. The assumption of lower sensitivity at the edges of a papilla’s frequency
response versus the center is supported by the results of several studies 共e.g., Ehret and Capranica, 1980; Narins and Hillery, 1983; Zelick and Narins, 1985兲. Since lower DPOAE
levels and higher DPOAE thresholds have in general been
measured at frequency ranges associated with the least sensitive hearing regions of the species examined 共e.g., Probst
et al. 1991兲, the degree of alignment between the DP audiogram dips and the vertical lines in Figs. 6 and 7 provides
useful information regarding the site of DPOAE generation.
For example, if a plot in Figs. 6 and 7 has its dip and vertical
line misaligned, the expectation that emission levels should
be minimal at the frequency range of minimal response is not
satisfied. From this observation it can be deduced that the
generation site of the respective distortion product cannot be
related to the generation site of the frequency component
against which the distortion product levels have been plotted.
Figure 6 indicates that, for R. p. pipiens, alignment of the
2 f 1 - f 2 DPOAE audiogram dip and the break in the frequency coverage between the two papillae 共1250 Hz;
Ronken, 1991兲 is best when 2 f 1 - f 2 amplitude is plotted as a
function of DP frequency 关Fig. 6共c兲兴. This suggests that the
generation of the 2 f 1 - f 2 distortion product is related to the
generation region corresponding to the DP frequency and not
to that corresponding to the frequencies of the primaries. For
the 2 f 2 - f 1 DPOAE, on the other hand, alignment occurs
when 2 f 2 - f 1 amplitude is plotted as a function of f 2 关Fig.
6共b兲兴. Near alignment is also observed when this distortion
product is plotted as a function of f 1 关Fig. 6共a兲兴. This suggests that, in contrast to 2 f 1 - f 2 , the generation of the
2 f 2 - f 1 DPOAE is related to the frequencies of the primaries
and not to the distortion product frequency. Similar results
were obtained for R. catesbeiana 共Fig. 7兲, supporting a similar interpretation.15
The implication is that the generation of the 2 f 1 - f 2
DPOAE in the frog may primarily occur at or near the
DPOAE frequency place, while the generation of the 2 f 2 - f 1
DPOAE may primarily occur at a frequency place between
the two primaries. These results differ from those observed
in mammals. As indicated by suppression studies using a
variety of methods and confirmed by studies assessing recovery functions following short exposures to pure-tone or noise
stimuli, the generation of the 2 f 1 - f 2 DPOAE in mammals
occurs primarily at a frequency place between the primaries.
The distortion product itself then acts as a weak stimulus and
may induce stimulus-frequency emissions at the distortion
product frequency 共Shera and Guinan, 1999兲. In contrast, the
generation of the 2 f 2 - f 1 DPOAE occurs at or near the
Vassilakis et al.: Emissions and frog hearing mechanisms
DPOAE frequency place on the basilar membrane 共Probst
et al., 1991; Knight and Kemp, 2000兲. The opposite behavior
of the 2 f 2 - f 1 and 2 f 1 - f 2 DPOAEs observed between mammalian and frog DPOAE data may be related to various anatomical differences between the mammalian and frog ears
and needs to be researched further.
Figures 6共c兲 and 7共c兲 also reveal that, for frequencies in
the AP, the 2 f 2 - f 1 and 2 f 1 - f 2 curves are very similar in
shape and overlaying when DPOAE levels are plotted
against emission frequency. This suggests that, regardless of
the place of generation and prior to their reemission from the
AP, both DPOAEs may be subject to the same mechanical
filtering by a structure with broad filter characteristics, possibly the tectorial membrane. Several studies have used
DPOAE data to derive filter characteristics 共e.g., Brown
et al., 1992; Allen and Fahey, 1993; Kössl and Vater, 1996兲,
while a recent study 共Lukashkin and Russell, 2003兲 specifically argues for the existence of a possible tectorial membrane filter in mammals, based on the phase characteristics of
different order DPOAEs.
3. Sexual dimorphism manifested in DPOAEs
Figures 9 and 10 indicate that, for the selected primarytone levels, and over most of the frequency range tested,
female subjects produced higher emission levels than male
subjects in both species.16 Overall, R. catesbeiana produced
higher emission levels than R. p. pipiens.17 In addition, there
is a difference in the DPOAE audiogram peaks between
males and females, suggesting that there are differences in
the frequency tuning between sexes for both species.18
Several studies discussed in Probst et al. 共1991兲 have
shown significantly higher levels of spontaneous OAEs for
females in adult, children, and infant humans compared to
males. The cause of this sex difference is not clear. In the
female organ of Corti, OHCs are arranged much more irregularly than in males, possibly due to the smaller physical dimensions of the female cochlea. It has been argued that such
irregularities may be linked to OAE generation 共LonsburyMartin et al., 1988兲.19 McFadden and Pasanen 共1998兲 linked
EOAE level differences between males and females to hormonal differences. They reported lower emission levels for
homosexual than heterosexual females, and argued that the
nature of this difference is consistent with the idea that homosexual females are exposed prenatally to higher levels of
androgens than are heterosexual females.
Another possible explanation may lay in sex differences
in middle ear transfer functions. In humans, for example, all
OAEs can be detected reliably within a range 共1– 8 kHz兲
determined to a large extent by the frequency range 共1– 6
kHz兲 of effective reverse transmission from the middle ear
共Probst et al., 1991兲. Puria 共2003兲 argues that the forward
and reverse transfer characteristics of the mammalian middle
ear alter significantly the levels of DPOAEs recorded in the
ear canal, in a frequency specific manner 共higher frequencies
are increasingly attenuated兲. Johansson and Arlinger 共2002兲
recorded significantly higher DPOAEs from human female
subjects and linked high middle-ear compliance present in
males to low emission levels. Additionally, clinical OAE
studies in infants routinely record higher evoked OAEs emisJ. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
sions from female newborns 共e.g., Newmark et al., 1997兲,
with this difference increasing with the frequency of the primaries 共e.g., Cassidy and Ditty, 2001兲. The frequencydependent change in EOAE levels may therefore be linked to
the frequency-dependent transfer characteristics of the
middle ear demonstrated by Puria 共2003兲, which, as Johansson and Arlinger 共2002兲 have argued in terms of middle ear
compliance, may be sexually dimorphic.
In a recent study of bullfrogs, Mason et al. 共2003兲
showed that, according to the current models of middle ear
function, there is considerably higher middle ear impedance
transform ratio in male bullfrogs compared to females. Although, according to the authors, it is not certain whether this
reflects the assumptions of the model or actual physiological
differences, such a transform ratio difference between male
and female bullfrog middle ears corresponds well with the
observed sex differences in the DPOAEs. In addition to
middle ear transfer differences, sex differences in the frog
have been reported in relation to peripheral auditory response. Narins and Capranica 共1976兲 demonstrated such differences for the neotropical treefrog Eleutherodactylus coqui,
associating sexual selectivity in the response properties of
the BP to sexual selectivity in behavior. Since higher emission levels are thought to correspond to higher auditory sensitivity, frequency-dependent differences between sexes in
OAE levels may indeed reflect sex differences in auditory
sensitivity, possibly associated with male-female differences
in what constitutes a biologically relevant sound.
V. SUMMARY AND CONCLUSIONS
DPOAE measurements were made for the 2 f 1 - f 2 and
2 f 2 - f 1 distortion products from male and female R. p. pipiens and R. catesbeiana. DPOAE I/O curves were obtained at
frequencies corresponding to the best response in the AP and
BP of both species. The 2 f 1 - f 2 I/O curves obtained from the
AP of both species are very similar to mammalian DPOAE
I/O curves, exhibiting a non-monotonic response, with a distinct notch for primary levels ⬃70 dB SPL. This similarity
suggests that, like in the mammalian ear, there may be an
amplification process present in the frog AP manifested as a
compressive nonlinearity with saturation. In mammals this
process is linked to prestin-mediated OHC activity. Since the
frog ear lacks OHCs and its hair cells contain no prestin in
their basolateral membranes, an alternative amplifier may be
based on hair cell bundle spontaneous movement, shown to
occur in the frog sacculus. The DPOAE I/O curves obtained
from the BP of both species exhibit a rather monotonic response, suggesting that no amplification process may be
present in this papilla. The difference in the response between amphibian and basilar papilla may be linked to respective differences in degree and type of innervation. It is also
consistent with the fact that only the AP is tonotopically
organized and could take better advantage of an amplification mechanism that increases tuning sharpness, as well as
with the observation that only this papilla exhibits SOAEs.
The dependence of DPOAE level on the relative levels
of primaries (L1 -L2 ) was investigated in R. p. pipiens, within
the range 50⭐L1 , L2 ⭐60 dB SPL. Consistent with intermodulation distortion expectations alone and differently
Vassilakis et al.: Emissions and frog hearing mechanisms
3723
from mammalian results, highest emissions were generally
obtained for L1 ⬵L2 . This difference between mammalian
and frog data is consistent with a difference in neural tuningcurve shape of mammalian and frog nerve fibers with comparable characteristic frequencies. In the mammalian case,
neural tuning curves are thought to reflect BM displacement
envelopes, which are in turn considered responsible for the
dependence of DPOAE level on the relative levels of the
primaries. Since the frog ear lacks a BM, the observed
DPOAE level dependence on L1 -L2 in the frog suggests that,
if some mechanical structure 共i.e., the tectorial membrane兲
functions analogously to the mammalian BM, it must be
more broadly tuned. Our interpretation is supported by the
rest of our results, outlined below, and does not exclude the
possibility of the optimal primary-level difference to be
L1 -L2 ⫽0 for primary levels outside the level-range examined.
DPOAE audiograms were obtained from males and females of both species for primary frequencies spanning the
animals’ hearing range and primary levels determined by the
present study. The results indicate the following:
共a兲
共b兲
共c兲
In agreement with previous DPOAE frog results 共Van
Dijk et al., 2002; Meenderink and Van Dijk, 2004兲,
DPOAEs are produced in both frog papillae and, except for primary frequencies at the high-frequency end
of the BP, R. catesbeiana produce stronger emissions
than R. p. pipiens.
In contrast to mammalian results, the generation of the
2 f 1 - f 2 DPOAE in the frog appears to occur primarily
at or near the DPOAE frequency place, while the generation of the 2 f 2 - f 1 DPOAE occurs primarily at a
frequency place between the primaries. Further study is
needed to confirm the observed difference in DPOAE
generation sites between mammals and frogs and determine the precise reason behind it. Regardless of
DPOAE generation site, both DPOAEs within the AP
may be subject to the same filtering mechanism, possibly the tectorial membrane.
For the primary-tone levels used, females of both R. p.
pipiens and R. catesbeiana produce higher emissions
than males in both papillae, with the difference being
larger in the BP. The observed similarities in male and
female frog DPOAE data in terms of I/O curve shapes,
dependence on L1-L2, etc. suggest that sex differences
in overall DPOAE levels are not related to DPOAE
generation mechanism differences. DPOAE-based
sexual dimorphism, also reported in humans and other
mammals, may be linked to inner ear anatomical differences, middle ear transfer impedance differences,
hormonal differences, nerve-fiber tuning differences,
and/or behavioral and environmental factors.
ACKNOWLEDGMENT
Work supported by NIH Grant No. DC-00222 to PMN.
1
In frogs, OAEs have been recorded from the ears of Rana pipiens pipiens
关DPOAEs: Van Dijk et al., 2002; Meenderink and Van Dijk, 2004; spontaneous OAEs 共SOAEs兲: Van Dijk et al., 1996兴, Rana catesbeiana 共SOAEs:
Gennosa et al., 1989兲, Rana esculenta 关SOAEs: Palmer and Wilson, 1982;
3724
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
Van Dijk et al., 1989; Van Dijk and Wit, 1987; stimulus-frequency OAEs
共SFOAEs兲: Palmer and Wilson, 1982兴, Rana temporaria 共SOAEs: Baker
et al., 1989兲, Xenopus laevis 共DPOAEs: Van Dijk et al., 2002兲, Hyla cinerea 共DPOAEs: Van Dijk and Manley, 2001; SOAEs: Van Dijk et al.,
1996兲, Hyla chrysoscelis, Hyla versicolor, and Leptodactylus albilabris
共SOAEs: Van Dijk et al., 1996兲. No measurable SOAE levels were obtained from Xenopus laevis 共Van Dijk et al., 1996兲 and no measurable
DPOAE levels were obtained from Scaphiopus couchi 共Van Dijk et al.,
2002兲. Neither SOAEs nor DPOAEs could be measured from Bombina
orientalis 共Van Dijk et al., 1996 and Van Dijk et al., 2002 respectively兲.
2
E.g., f 2 / f 1 ⫽1.2 in Kalluri and Shera 共2001兲; f 2 / f 1 ⫽1.22 in Emmerich
et al. 共2000兲; f 2 / f 1 ⫽1.25 in Kemp and Brown 共1983兲 共in Probst et al.,
1991兲.
3
The only available parametric DPOAE study on the frog to date 共Meenderink and Van Dijk, 2004兲 examines in detail the I/O characteristics of the
two papillae in R. p. pipiens, offering results that, in general, agree with the
results of the present study.
4
Although exploration of the f 2 / f 1 space in mammals has revealed several
discontinuities, we assume that the value selected by the present study is
close enough to the value used in previous frog DPOAE studies to allow for
a meaningful comparison.
5
The body temperature of the animals was not monitored systematically.
6
Below 200 Hz the noise floor of the probe/analysis system was too high
(⬎25 dB SPL兲 and the speaker response too weak (⬍65 dB SPL兲, while
the hearing range of neither frog species examined exceeds 3000 Hz.
7
Windowing: Blackman and Harris; averaging: 200 time frames with 90%
overlap; input range: adjusted for each primary tone level combination to
ensure consistent use of the analyzer’s full dynamic range; harmonic/
intermodulation distortion of the speaker–microphone–analyzer system
⭓83 dB below the primaries at all frequencies tested. System distortion
was determined by attaching the probe tube system to a hard surface and
measuring the level recorded at frequencies corresponding to DPOAE frequencies for selected primary-tone frequency and amplitude values. The
dBV reading of the analyzer was converted to dB SPL using a measuring
amplifier 共B&K 2609, Brüel & Kjær, Nærum, Denmark兲.
8
The analysis bandwidth was 0.98 Hz for 240 Hz⭐ f 1 ⭐800 Hz, 1.95 Hz for
850 Hz⭐ f 1 ⭐1600 Hz, and 3.9 Hz for 1700 Hz⭐ f 1 ⭐3000 Hz.
9
Shofner and Feng 共1984兲 indicate 1000 Hz as the frequency value marking
the separation between AP and BP response in R. catesbeiana.
10
One alternative hypothesis involves transition among different vibration
modes of the inner ear structures as the levels of the primaries change,
claimed to be responsible for the observed DPOAE I/O curve notch. In
such an explanation, it is expected that, for a given set of primary tone
frequencies, the DPOAE I/O curve notches should be observed at the same
stimulus levels, regardless of distortion product. Our data indicate that this
may be the case for the R. catesbeiana I/O curves but it certainly is not for
the R. p. pipiens I/O curves 关Figs. 2共a兲 and 2共c兲兴, suggesting that the
‘‘vibration mode transition’’ hypothesis may be excluded as an explanation
for the DPOAE I/O curve shapes observed. Another explanation for the
presence of notches in the DPOAE I/O curves is based on a hypothesized
frequency shift of DPOAE fine-structure in human subjects with changing
stimulus levels 共He and Schmiedt, 1993; Whitehead, 1998兲. Figure 3 demonstrates that this is not the case for the frog species tested. In mammals,
the presence of DP audiogram fine structure has been explained in terms of
interference between two emission components 共e.g., Talmage et al., 1998,
1999兲. Further study is needed to determine the origin of the DPOAE fine
structure observed in the frog.
11
In the case of the frog ear, however, the DPOAE phase data obtained and
analyzed by Meenderink and Van Dijk 共2004兲 do not appear to support
such a model.
12
The presence of an inner ear amplifier in the echidna, a Monotreme with
features common to early mammals, birds, and reptiles and which has no
organ of Corti, has also been inferred based on DPOAE data 共Mills and
Shepherd, 2001兲.
13
The range of primary stimulus levels tested 共50– 60 dB SPL兲 is below the
DPOAE I/O curve saturation point for R. p. pipiens 共Fig. 2兲, supporting the
suggestion that the observed effect of L1 -L2 on DPOAE levels is indeed
indicative of a difference between mammalian and frog responses as opposed to simply a level saturation response in the frog.
14
In an alternative interpretation of the data by Hillery and Narins 共1987兲,
Pitchford and Ashmore 共1987兲 argue that the large phase accumulations in
the phase-locked responses of auditory-nerve fibers in frogs may not reflect phase accumulations of a mechanical traveling wave 共e.g., Hillery
Vassilakis et al.: Emissions and frog hearing mechanisms
and Narins, 1984兲 but may result from electrical filtering within the hair
cells.
15
The DP audiogram dips for R. catesbeiana are shallower than those for R.
p. pipiens, due to the fact that, in R. catesbeiana, there is a slight overlap
in the coverage of the two papillae, while in R. p. pipiens there is not.
16
These results are consistent with the results obtained from the first two
experiments, examining the relationship of absolute and relative levels of
the primaries to emission levels.
17
With the exception of emissions from the high-frequency end of the BP
associated possibly with the reduced response of the male and female
bullfrog middle ear at these frequencies (⬎⬃1200 Hz; Mason and Narins,
2002a, b兲.
18
Although the dependence of DPOAE levels on animal weight and/or the
size of the eardrum was not systematically examined, our data suggest that
any DPOAE level difference between species and sexes is most likely not
related to such factors.
19
Sex differences in DPOAE phase delay measures have also been attributed
to anatomical differences in cochlear length 共Bowman et al., 2000兲.
Abdala, C. 共2000兲. ‘‘Distortion product otoacoustic emission (2 f 1 - f 2 ) amplitude growth in human adults and neonates,’’ J. Acoust. Soc. Am. 107,
446 – 456.
Allen, J. B., and Fahey, P. F. 共1993兲. ‘‘A second cochlear-frequency map that
correlates distortion product, neural tuning measurements,’’ J. Acoust. Soc.
Am. 94, 809– 816.
Baker, R. J., Wilson, J. P., and Whitehead, M. L. 共1989兲. ‘‘Otoacoustic
evidence for nonlinear behavior in frog hearing: Suppression but no distortion products,’’ in Cochlear Mechanisms: Structure, Function and Models, edited by J. Wilson and D. T. Kemp 共Plenum, New York兲, pp. 349–
356.
Bowman, D. M., Brown, D. K., and Kimberley, B. P. 共2000兲. ‘‘An examination of gender differences in DPOAE phase delay measurements in
normal-hearing human adults,’’ Hear. Res. 142, 1–11.
Brown, A. M., Gaskill, S. A., and Williams, D. M. 共1992兲. ‘‘Mechanical
filtering of sound in the inner ear,’’ Proc. R. Soc. London, Ser. B 250,
29–34.
Brown, A. M., McDowell, B., and Forge, A. 共1989兲. ‘‘Effects of chronic
gentamicin treatment on hair cells can be monitored using acoustic distortion products,’’ Hear. Res. 42, 143–156.
Capranica, R. R. 共1976兲. ‘‘Morphology and physiology of the auditory system,’’ in Frog Neurobiology, edited by R. Llinas and W. Precht 共Springer
Verlag, Berlin兲, pp. 551–575.
Cassidy, J. W., and Ditty, K. M. 共2001兲. ‘‘Gender differences among newborns on a transient otoacoustic emissions test for hearing,’’ J. Music.
Ther. 38, 28 –35.
Crawford, A. C., and Fettiplace, R. 共1985兲. ‘‘The mechanical properties of
cilliary bundles of turtle cochlear hair cells,’’ J. Physiol. 共London兲 364,
359–379.
Ehret, G., and Capranica, R. R. 共1980兲. ‘‘Masking patterns and filter characteristics of auditory nerve fibers in the green treefrog 共Hyla cinerea兲,’’ J.
Comp. Physiol. 141, 1–12.
Emmerich, E., Richter, F., Meißner, W., and Dieroff, H. G. 共2000兲. ‘‘The
effect of impulse noise exposure on distortion product otoacoustic emissions in the awake guinea pig,’’ Eur. Arch. Otorhinolaryngol. 257128 –
132.
Fahey, P. F., Stagner, B. B., Lonsbury-Martin, B. L., and Martin, G. K.
共2000兲. ‘‘Nonlinear interactions that could explain distortion product interference response areas,’’ J. Acoust. Soc. Am. 108, 1786 –1802.
Faulstich, M., and Kössl, M. 共2000兲. ‘‘Evidence for multiple DPOAE components based upon group delay of the 2 f 1 - f 2 distortion in the gerbil,’’
Hear. Res. 140, 99–110.
Feng, A. S., Narins, P. M., and Capranica, R. R. 共1975兲. ‘‘Three populations
of primary auditory fibers in the bullfrog 共Rana catesbeiana兲: Their peripheral origins and frequency sensitivities,’’ J. Comp. Physiol. 100, 221–
229.
Geisler, C. D., Van Bergeijk, W. A., and Frishkopf, S. 共1964兲. ‘‘The inner ear
of the bullfrog,’’ J. Morphol. 114, 43–58.
Gennosa, T. J., Voigt, H. F., and Burkard, R. 共1989兲. ‘‘Spontaneous otoacoustic emissions in Rana catesbeiana, the American bullfrog,’’ J. Acoust.
Soc. Am. Suppl. 1 85, S35.
He, N., and Schmiedt, R. A. 共1993兲. ‘‘Fine structure of the 2f1-f2 acoustic
distortion product: Changes with primary level,’’ J. Acoust. Soc. Am. 94,
2659–2669.
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
Hillery, C. M., and Narins, P. M. 共1984兲. ‘‘Neurophysiological evidence for
a traveling wave in the amphibian inner ear,’’ Science 255, 1037–1039.
Hillery, C. M., and Narins, P. M. 共1987兲. ‘‘Frequency and time domain
comparison of low-frequency auditory fibre responses in two anuran amphibians,’’ Hear. Res. 25, 233–248.
Hudspeth, A. 共1997兲. ‘‘Mechanical amplification of stimuli by hair cells,’’
Curr. Opin. Neurobiol. 7, 490– 496.
Johansson, M., and Arlinger, S. 共2002兲. ‘‘Otoacoustic emissions in a general
adult population in Sweden,’’ Proceedings of Nordic Noise 2002. Stockholm, Sweden.
Kalluri, R., and Shera, C. A. 共2001兲. ‘‘Distortion-product source unmixing:
A test of the two-mechanism model for DPOAE generation,’’ J. Acoust.
Soc. Am. 109, 622– 637.
Kastanioudakis, I., Ziavra, N., Anastasopoulos, D., and Skevas, A. 共2003兲.
‘‘Measuring of distortion product otoacoustic emissions using multiple
tone pairs,’’ Eur. Arch. Otorhinolaryngol. 260, 395– 400.
Kemp, D. T. 共1979兲. ‘‘Evidence for cochlear nonlinearity and frequency
selective wave amplification in the cochlea,’’ Arch. Otolaryngol. 224, 37–
45.
Kennedy, H. J., Evans, M. G., Crawford, A. C., and Fettiplace, R. 共2003兲.
‘‘Fast adaptation of mechanoelectrical transducer channels in mammalian
cochlear hair cells,’’ Nat. Neurosci. 6, 832– 836.
Knight, R. D., and Kemp, D. T. 共2000兲. ‘‘Indications of different distortion
product otoacoustic emission mechanisms from a detailed f 1 , f 2 area
study,’’ J. Acoust. Soc. Am. 107, 457– 473.
Köppl, C., Manley, G. A., and Johnstone, B. M. 共1993兲. ‘‘Acoustic distortion
products in the ear canal of the bobtail lizard. I: General characteristics,’’
J. Acoust. Soc. Am. 93, 2820–2933.
Kössl, M., and Boyan, G. S. 共1998兲. ‘‘Acoustic distortion products from the
ear of a grasshopper,’’ J. Acoust. Soc. Am. 104, 326 –335.
Kössl, M., and Vater, M. 共1996兲. ‘‘Further studies on the mechanics of the
cochlear partition in the mustached bat II. A second cochlear frequency
map derived from acoustic distortion products,’’ Hear. Res. 94, 78 – 86.
Kummer, P., Janssen, T., Hulin, P., and Arnold, W. 共2000兲. ‘‘Optimal L1-L2
primary tone level separation remains independent of test frequency in
humans,’’ Hear. Res. 146, 47–56.
Lewis, E. R. 共1977兲. ‘‘Structural correlates of function in the anuran amphibian papilla,’’ Scan Electron Microsc. II, 429– 436.
Lewis, E. R., and Narins, P. M. 共1999兲. ‘‘The Acoustic Periphery of Amphibians: Anatomy and Physiology,’’ in Comparative Hearing: Fish and
Amphibians, edited by R. R. Fay and A. N. Popper 共Springer-Verlag, New
York兲, pp. 101–154.
Lewis, E. R., Leverenz, E. L., and Bialek, W. S. 共1985兲. The Vertebrate
Inner Ear 共CRC, Boca Raton, FL兲.
Lewis, E. R., Leverenz, E. L., and Koyama, H. 共1982兲. ‘‘The tonotopic
organization of the bullfrog amphibian papilla, an auditory organ lacking a
basilar membrane,’’ J. Comp. Physiol. 145, 437– 445.
Liberman, C. M., Gao, J., He, D. Z. Z., Wu, X., Jia, S., and Zuo, J. 共2002兲.
‘‘Prestin is required for electromotility of the outer hair cell and for the
cochlear amplifier,’’ Nature 共London兲 419, 300–304.
Lonsbury-Martin, B. L., Martin, G. K., Probst, R., and Coats, A. C. 共1988兲.
‘‘Spontaneous otoacoustic emissions in a nonhuman primate: II. Cochlear
anatomy,’’ Hear. Res. 33, 69–94.
Lonsbury-Martin, B. L., McCoy, M. J., Whitehead, M. L., and Martin, G. K.
共1993兲. ‘‘Clinical testing of distortion-product otoacoustic emissions,’’ Ear
Hear. 14, 11–22.
Lukashkin, A. N., and Russell, I. J. 共2003兲. ‘‘A second, low-frequency mode
of vibration in the intact mammalian cochlea,’’ J. Acoust. Soc. Am. 113,
1544 –1550.
Lukashkin, A. N., Lukashkina, V. A., and Russell, I. J. 共2002兲. ‘‘One source
for distortion product otoacoustic emissions generated by low- and highlevel primaries,’’ J. Acoust. Soc. Am. 111, 2740–2748.
Maison, S., Micheyl, C., and Collet, L. 共1997兲. ‘‘Medial olivocochlear efferent system in humans studied with amplitude-modulated tones,’’ J. Neurophysiol. 77, 1759–1768.
Manley, G. A., Kirk, D., Köppl, C., and Yates, G. K. 共2001兲. ‘‘In-vivo
evidence for a cochlear amplifier in the hair-cell bundle of lizards,’’ Proc.
Natl. Acad. Sci. U.S.A. 98, 2826 –2831.
Martin, G. K. 共2002兲. ‘‘DPOAE generation mechanisms,’’ OAEs Portal
Zone, University of Ferrara, Italy. http://www.otoemissions.org/
guest_editorials/editorial_May_June_2002.html. 关Last accessed: 10/2004.兴
Martin, P., and Hudspeth, A. 共1999兲. ‘‘Active hair-bundle movements can
amplify a hair cell’s response to oscillatory mechanical stimuli,’’ Proc.
Natl. Acad. Sci. U.S.A. 7, 14306 –14311.
Vassilakis et al.: Emissions and frog hearing mechanisms
3725
Mason, M. J., and Narins, P. M. 共2002a兲. ‘‘Vibrometric studies of the middle
ear of the bullfrog Rana catesbeiana. I. The extrastapes,’’ J. Exp. Biol.
205, 3153–3165.
Mason, M. J., and Narins, P. M. 共2002b兲. ‘‘Vibrometric studies of the middle
ear of the bullfrog Rana catesbeiana. II. The operculum,’’ J. Exp. Biol.
205, 3167–3176.
Mason, M. J., Lin, C. C., and Narins, P. M. 共2003兲. ‘‘Sex differences in the
middle ear of the bullfrog 共Rana catesbeiana兲,’’ Brain Behav. Evol. 61,
91–101.
McFadden, D., and Pasanen, E. G. 共1998兲. ‘‘Comparison of the auditory
systems of heterosexuals and homosexuals: Click-evoked otoacoustic
emissions,’’ Proc. Natl. Acad. Sci. U.S.A. 95, 2709–2713.
Meenderink, S. W. F., and Van Dijk, P. 共2004兲. ‘‘Level dependence of distortion product otoacoustic emissions in the leopard frog, Rana pipiens
pipiens,’’ Hear. Res. 192, 107–118.
Mills, D. M. 共2002兲. ‘‘Interpretation of standard distortion product otoacoustic emission measurements in light of the complete parametric response,’’
J. Acoust. Soc. Am. 112, 1545–1560.
Mills, D. M., and Shepherd, R. K. 共2001兲. ‘‘Distortion product otoacoustic
emission and auditory brainstem responses in the echidna 共Tachyglossus
aculeatus兲,’’ J. Assoc. Res. Otolaryngol. 2, 130–146.
Mom, T., Bonfils, P., Gilain, L., and Avan, P. 共2001兲. ‘‘Origin of cubic
difference tones generated by high-intensity stimuli: Effect of ischemia
and auditory fatigue on the gerbil cochlea,’’ J. Acoust. Soc. Am. 110,
1477–1488.
Narins, P. M., and Capranica, R. R. 共1976兲. ‘‘Sexual differences in the auditory system of the tree frog Eleutherodactylus coqui,’’ Science 192,
378 –380.
Narins, P. M., and Hillery, C. 共1983兲. ‘‘Frequency coding in the inner ear of
anuran amphibians,’’ in Hearing–Physiological Bases and Psychophysics,
edited by R. Klinke and R. Hartmann 共Springer-Verlag, New York兲, pp.
70–76.
Newmark, M., Merlob, P., Bresloff, I., Olsha, M., and Attias, J. 共1997兲.
‘‘Click evoked otoacoustic emissions: inter-aural and gender differences in
newborns,’’ J. Basic Clin. Physiol. Pharmacol. 8, 133–139.
Owens, J. J., McCoy, M. J., Lonsbury-Martin, B. L., and Martin, G. K.
共1992兲. ‘‘Influence of otitis media on evoked otoacoustic emission in children,’’ Semin. Hear. 13, 53– 65.
Owens, J. J., McCoy, M. J., Lonsbury-Martin, B. L., and Martin, G. K.
共1993兲. ‘‘Otoacoustic emissions in children with normal ears, middle ear
dysfunction, and ventilating tubes,’’ Am. J. Otol. 14, 34 – 40.
Palmer, A. R., and Wilson, J. P. 共1982兲. ‘‘Spontaneous and evoked acoustic
emissions in the frog Rana esculenta,’’ J. Phonetics 324, 66.
Pitchford, S., and Ashmore, J. F. 共1987兲. ‘‘An electrical resonance in hair
cells of the amphibian papilla of the frog Rana temporaria,’’ Hear. Res.
27, 75– 83.
Probst, R., Lonsbury-Martin, B. L., and Martin, G. K. 共1991兲. ‘‘A review of
otoacoustic emissions,’’ J. Acoust. Soc. Am. 89, 2027–2067.
Puria, S. 共2003兲. ‘‘Measurements of human middle ear forward and reverse
acoustics: Implications for otoacoustic emissions,’’ J. Acoust. Soc. Am.
113, 2773–2789.
Robbins, R. G., Bauknight, B. S., and Honrubia, M. D. 共1967兲. ‘‘Anatomical
distribution of efferent fibers in the VIIIth cranial nerve of the bullfrog
共Rana catesbeiana兲,’’ Acta Oto-Laryngol. 64, 436 – 448.
Robles, L., and Ruggero, M. A. 共2001兲. ‘‘Mechanics of the mammalian
cochlea,’’ Physiol. Rev. 81, 1305–1352.
Ronken, D. A. 共1991兲. ‘‘Spike discharge properties that are related to the
characteristic frequency of single units in the frog auditory nerve,’’ J.
Acoust. Soc. Am. 90, 2428 –2440.
Rosowski, J. J., Peake, W. T., and White, J. R. 共1984兲. ‘‘Cochlear nonlinearities inferred from two-tone distortion products in the ear canal of the
alligator lizard,’’ Hear. Res. 13, 141–158.
Schneider, S., Prijs, V., and Schoonhoven, R. 共2003兲. ‘‘Amplitude and phase
of distortion product otoacoustic emissions in the guinea pig in an ( f 1 , f 2 )
area study,’’ J. Acoust. Soc. Am. 113, 3285–3296.
Shera, C. A., and Guina, Jr., J. J. 共1999兲. ‘‘Evoked otoacoustic emissions
arise by two fundamentally different mechanisms: A taxonomy for mammalian OAEs,’’ J. Acoust. Soc. Am. 105, 782–798.
3726
J. Acoust. Soc. Am., Vol. 116, No. 6, December 2004
Shera, C. A., and Guinan, Jr., J. J. 共2003兲. ‘‘Stimulus-frequency-emission
group delay: A test of coherent reflection filtering and a window on cochlear tuning,’’ J. Acoust. Soc. Am. 113, 2762–2772.
Shofner, W. P., and Feng, A. S. 共1984兲. ‘‘Quantitative light and scanning
electron microscopic study of the developing auditory organs of the bullfrog: Implications on their functional characteristics,’’ J. Comp. Neurol.
224, 141–154.
Simmons, D. D., Bertolotto, C., and Narins, P. M. 共1992兲. ‘‘Innervation of
the amphibian and basilar papillae in the leopard frog: Reconstructions of
single labeled fibers,’’ J. Comp. Neurol. 322, 191–200.
Stover, L. J., Gorga, M. P., Neely, S. T., and Montoya, D. 共1996兲. ‘‘Toward
optimizing the clinical utility of distortion product otoacoustic emission
measurements,’’ J. Acoust. Soc. Am. 100, 956 –967.
Talmadge, C. L., Long, G. R., Tubis, A., and Dhar, S. 共1999兲. ‘‘Experimental
confirmation of the two-source interference model for the fine structure of
distortion product otoacoustic emissions,’’ J. Acoust. Soc. Am. 105, 275–
292.
Talmadge, C. L., Tubis, A., Long, G. R., and Piskorski, P. 共1998兲. ‘‘Modeling otoacoustic emission and hearing threshold fine structures,’’ J. Acoust.
Soc. Am. 104, 1517–1541.
Tubis, A., Talmadge, C. L., Long, G. R., and Dhar, S. 共2000兲. ‘‘On the
relationships between the fixed-f 1 , fixed-f 2 , and fixed-ratio phase derivatives of the 2 f 1 - f 2 distortion product otoacoustic emission,’’ J. Acoust.
Soc. Am. 108, 1772–1784.
Van Dijk, P., and Manley, G. A. 共2001兲. ‘‘Distortion product otoacoustic
emissions in the tree frog Hyla cinerea,’’ Hear. Res. 153, 14 –22.
Van Dijk, P., and Wit, H. P. 共1987兲. ‘‘Temperature dependence of frog spontaneous otoacoustic emissions,’’ J. Acoust. Soc. Am. 82, 2147–2150.
Van Dijk, P., Mason, M. J., and Narins, P. M. 共2002兲. ‘‘Distortion product
otoacoustic emissions in frogs: correlation with middle and inner ear properties,’’ Hear. Res. 173, 100–108.
Van Dijk, P., Narins, P. M., and Mason, M. J. 共2003兲. ‘‘Physiological vulnerability of distortion product otoacoustic emissions from the amphibian
ear,’’ J. Acoust. Soc. Am. 114, 2044 –2048.
Van Dijk, P., Narins, P. M., and Wang, J., 共1996兲. ‘‘Spontaneous otoacoustic
emissions in seven frog species,’’ Hear. Res. 101, 102–112.
Van Dijk, P., Wit, H. P., and Segenhout, J. M. 共1989兲. ‘‘Spontaneous otoacoustic emissions in the European edible frog 共Rana esculenta兲: Spectral
details and temperature dependence,’’ Hear. Res. 42, 273–282.
Wagner, W., and Plinkert, P. K. 共1999兲. ‘‘The relationship between auditory
threshold and evoked otoacoustic emissions,’’ Eur. Arch. Otorhinolaryngol. 256, 177–188.
Whitehead, M. L. 共1998兲. ‘‘Species differences of distortion-product otoacoustic emissions: Comment on Interpretation of distortion-product otoacoustic emission measurements. I. Two stimulus tones 关J. Acoust. Soc.
Am. 102, 413– 429 共1997兲兴,’’ J. Acoust. Soc. Am. 103, 2740–2742.
Whitehead, M. L., McCoy, M. J., Lonsbury-Martin, B. L., and Martin, G. K.
共1995a兲. ‘‘Dependence of distortion-product otoacoustic emissions on primary levels in normal and impaired ears: I. Effects of decreasing L2 below
L1 ,’’ J. Acoust. Soc. Am. 97, 2346 –2358.
Whitehead, M. L., Stagner, B. B., McCoy, M. J., Lonsbury-Martin, B. L.,
and Martin, G. K. 共1995b兲. ‘‘Dependence of distortion-product otoacoustic
emissions on primary levels in normal and impaired ears: II. Asymmetry
in L1 , L2 space,’’ J. Acoust. Soc. Am. 97, 2359–2377.
Wit, H. P., Van Dijk, P., and Segenhout, J. M. 共1989兲. ‘‘DC injection alters
otoacoustic emission frequency in the frog,’’ Hear. Res. 41, 199–204.
Zelick, R. D., and Narins, P. M. 共1985兲. ‘‘Temporary threshold shifts, adaptation, and characteristics of frog auditory nerve fibers,’’ Hear. Res. 17,
161–176.
Zhang, M., and Abbas, P. J. 共1997兲. ‘‘Effects of middle ear pressure on
otoacoustic emission measures,’’ J. Acoust. Soc. Am. 102, 1032–1037.
Zheng, J., Shen, W., He, D. Z. Z., Long, K., Madison, L. D., and Dallos, P.
共2000兲. ‘‘Prestin is the motor protein of cochlear outer hair cells,’’ Nature
共London兲 405, 149–155.
Vassilakis et al.: Emissions and frog hearing mechanisms