Noise-induced changes in cochlear compression in the rat as indexed by forward masking of the auditory brainstem response

Noise-induced changes in cochlear compression in the rat as indexed by forward masking of the auditory brainstem response

Hearing Research 294 (2012) 64e72 Contents lists available at SciVerse ScienceDirect Hearing Research journal homepage: www.elsevier.com/locate/hear...

1MB Sizes 0 Downloads 21 Views

Hearing Research 294 (2012) 64e72

Contents lists available at SciVerse ScienceDirect

Hearing Research journal homepage: www.elsevier.com/locate/heares

Research paper

Noise-induced changes in cochlear compression in the rat as indexed by forward masking of the auditory brainstem response Eric C. Bielefeld*, Evelyn M. Hoglund, Lawrence L. Feth Department of Speech and Hearing Science, The Ohio State University, 110 Pressey Hall, 1070 Carmack Road, Columbus, OH 43220, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 June 2012 Received in revised form 4 September 2012 Accepted 18 October 2012 Available online 30 October 2012

The current study was undertaken to investigate changes in forward masking patterns using onfrequency and off-frequency maskers of 7 and 10 kHz probes in the SpragueeDawley rat. Offfrequency forward masking growth functions have been shown in humans to be non-linear, while on-frequency functions behave linearly. The non-linear nature of the off-frequency functions is attributable to active processing from the outer hair cells, and was therefore expected to be sensitive to noiseinduced cochlear damage. For the study, nine SpragueeDawley rats’ auditory brainstem responses (ABRs) were recorded with and without forward maskers. Forward masker-induced changes in latency and amplitude of the initial positive peak of the rats’ auditory brainstem responses were assessed with both off-frequency and on-frequency maskers. The rats were then exposed to a noise designed to induce 20e 40 dB of permanent threshold shift. Twenty-one days after the noise exposure, the forward masking growth functions were measured to assess noise-induced changes in the off-frequency and on-frequency forward masking patterns. Pre-exposure results showed compressive non-linear masking effects of the off-frequency conditions on both latency and amplitude of the auditory brainstem response. The noise rendered the off-frequency forward masking patterns more linear, consistent with human behavioral findings. On- and off-frequency forward masking growth functions were calculated, and they displayed patterns consistent with human behavioral functions, both prior to noise and after the noise exposure. Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction Noise-induced hearing loss (NIHL) is estimated to affect 26 million Americans between the ages of 20 and 69 years old (NIDCD, 2008), as well as w5.2 million children between the ages of 6 and 19 years old (Niskar et al., 2001). NIOSH estimates that 30 million workers are exposed to potentially hazardous noise, and that another 9 million to potentially ototoxic solvents and chemicals (NIOSH, 2001). NIHL is known to result from a broad spectrum of changes in the cochlea (reviewed in Henderson et al., 2006), resulting from a combination of oxidative stress (Yamane et al., 1995; Ohlemiller et al., 1999; Ohinata et al., 2000) and direct mechanical trauma to the structures of the organ of Corti (Hamernik et al., 1984, 1986; Ahmad et al., 2003). While most structures in the organ of Corti are vulnerable to noise damage to

Abbreviations: ABR, auditory brainstem response; FWM, forward masking; GoM, growth of masking; IeO, inputeoutput; NIHL, noise-induced hearing loss; Off-Freq, off-frequency; On-Freq, on-frequency; OHC, outer hair cell; PTS, permanent threshold shift; TDT, Tucker Davis Technologies. * Corresponding author. Tel.: þ1 614 292 9436; fax: þ1 614 292 7504. E-mail address: [email protected] (E.C. Bielefeld). 0378-5955/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.heares.2012.10.007

some degree, a key pathology of NIHL is damage/death of the outer hair cells (OHCs). OHC pathology has been commonly linked with permanent threshold shift (PTS), though damage to the OHCs may also underlie much noise-induced temporary threshold shift as well (Patuzzi, 2002; Howgate and Plack, 2011). OHCs, through their electromotility (Brownell et al., 1985), contribute active processing to drive the cochlear amplifier. The cochlear amplifier contributes the cochlea’s compressive non-linearity of basilar membrane motion, which allows for the large dynamic range that the mammalian cochlea possesses. The cochlear amplifier also contributes to very precise frequency resolution at the level of the basilar membrane (Neely and Kim, 1983). Loss of the OHCs in a particular frequency region of the cochlea will result in a 40e 60 dB loss of hearing sensitivity (Henderson et al., 2006), due to the loss of the cochlear amplification needed to hear the lowerintensity sounds. In addition, tuning curves are broader in individuals with OHC loss (Ryan and Dallos, 1975). The current study examined forward masking (FWM) patterns as an index of compressive non-linearity in the cochlea, and the effects of noise exposure on those FWM patterns. Oxenham and Plack (1997) studied behavioral measures of cochlear compression using FWM growth-of-masking (GoM) functions in human

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72

listeners. In normal hearing listeners, in conditions in which the probe and masker frequencies were the same (the on-frequency (On-Freq) condition), the GoM function was linear. But when the masker was one octave below the probe (the off-frequency (OffFreq) condition), the GoM function was highly compressive. In the Off-Freq condition, a high masker level was required to adequately mask even lower-intensity probe levels. This compressive nonlinear function was largely absent in the study participants with sensorineural hearing losses. In those participants, the Off-Freq GoM was linear, and closely mirrored the On-Freq condition. The results indicated that this development of a linear Off-Freq FWM function reflected a loss of cochlear compression in the participants with sensorineural hearing loss (Oxenham and Plack, 1997). The purpose of the current study was to examine the FWM patterns in rats using masking of the P1 wave of the ABR, as well as to examine the effects of noise exposure on FWM functions. Electrophysiology responses represent unique measures of FWM in that there are multiple dependent variables to be examined. The current study examined both P1 latency and P1 peak-to-peak amplitude. P1 latency serves as an index of speed of neural transmission in the cochlea (Møller, 1981; Don et al., 1998) and P1 amplitude indicates the amount and synchrony of neural activity generated by the afferent auditory nerve fibers (Jewett and Williston, 1971; Møller et al., 1988) firing in response to cochlear stimulation. GoM functions using latency and amplitude measures were also calculated to determine the effects of the noise on FWM patterns. Study of FWM GoM functions in an animal model allows for controlled noise exposures that result in NIHL that is predictable and consistent in both magnitude of PTS and frequency distribution. Numerous studies have established that FWM of the ABR is a viable test procedure in animal models (Gorga et al., 1983; Walton et al., 1995; Duan and Canlon, 1996a,b,c). Those earlier studies utilized probe levels that were fixed in a range 10e25 dB above threshold. The use of multiple probe levels and masker levels in the current study (detailed in Materials and Methods) allowed for the development of FWM GoM functions to determine whether masking of the ABR, as indexed by latency or amplitude changes, was a valid means of assessing cochlear compressive non-linearity, and whether the FWM GoM functions changed with NIHL in a manner similar to that measured by Oxenham and Plack (1997) in their study participants with sensorineural hearing loss. 2. Materials and methods 2.1. Subjects Nine adult male SpragueeDawley rats were used in the study. They were obtained from Harlan Laboratories at ages 2e3 months. The animals were housed in a quiet colony (<45 dBA). All procedures involving use and care of the animals were reviewed and approved by The Ohio State University’s Institutional Animal Care and Use Committee. 2.2. Auditory brainstem response (ABR) testing In order to assess hearing thresholds and FWM patterns, the rats were tested using free-field ABR. For all ABR test procedures, the animals were anesthetized with inhalant isoflurane (4% for induction, 1.5% for maintenance, 1 L/min O2 flow rate). Induction was achieved in plastic induction box. Once the anesthetic state was reached, the animal was transferred to the single-walled sound-attenuating booth (Industrial Acoustics Company, Bronx, NY) for testing, during which time anesthesia was delivered through a nose cone. Needle recording electrodes (30-gage, 12 mm in length; Grass Technologies, West Warwick, RI) were

65

placed at the vertex (non-inverting), below the left pinna (inverting) and behind the shoulder blade (ground). All stimuli were generated using commercial software (SigGenRz version 5.1, Tucker Davis Technologies (TDT), Gainesville, FL). Each tone burst was 1 ms in duration, and had a 0.5 ms rise/fall time with no plateau. Stimuli were presented at a rate of 19/s. Signals were routed to a speaker (TDT Model MF1) positioned at zero degrees azimuth, 10 cm from the vertex of each rat’s head. Acoustic stimuli were calibrated prior to each testing session, by recording the output of the speaker with a microphone placed at the animals’ head level. The rats’ evoked responses were amplified with a gain of 50,000, using a headstage (TDT RA4LI) connected to a preamplifier (TDT RA4PA), and bandpass filtered from 100 to 3000 Hz. For threshold testing, test stimuli consisted of alternating phase tone bursts at frequencies of 5, 7, 10, 12, and 15 kHz. 250 sweeps were averaged at each stimulus level using commercial software (TDT BioSigRz version 5.1). The level of the signal was decreased in 5 dB steps from 90 dB pSPL to a level 15 dB below that of the lowest level that evoked a detectable and repeatable response. Threshold was recorded as the lowest level at which a detectable response was elicited and could be repeated. For FWM, the 1 ms probe tone burst was preceded by a 40 ms masker tone that was either the same frequency as the probe tone burst (for On-Freq masking conditions) or one octave below the probe tone burst (for Off-Freq masking conditions). The silent gap between the masker and the probe was fixed at 5 ms. For each probe level, a series of masker levels was presented: 0, 20, 30, 40, 50, 60, 70, 80, and 90 dB SPL. Probe level descended in 10 dB steps from 90 to 40 dB SPL. Two probe frequencies were tested: 7 (OffFreq masker of 3.5 kHz; On-Freq masker of 7 kHz) and 10 kHz (OffFreq masker of 5 kHz; On-Freq masker of 10 kHz). Thus, a matrix of 216 test conditions (2 probe frequencies  2 masker frequencies  6 probe levels  9 masker levels) was utilized. Latency and amplitude of the P1 wave were measured, with amplitude measured as the peak-to-peak amplitude of the positive peak to the following negative peak. Latency and amplitude changes at different masking levels were assembled into latency and amplitude inputeoutput (IeO) functions. Two sets of GoM functions were also calculated, one set using latency shift as the dependent variable, and the other using amplitude reduction as the dependent variable. For calculation of the latency GoM functions, masker threshold was defined as the lowest masker level that induced a 0.07 ms shift in P1 latency. The criterion of a 0.07 ms latency shift was selected because it was the lowest level found to be outside the range of normal latency variance, and consistently represented the first significant shift that appeared to be induced by the masker. In any individual animal, once a 0.07 ms latency shift occurred, each successive increase in masker level induced a latency shift no less than 0.07 ms. For the amplitude GoM functions, the masker threshold was defined as the lowest masker level that reduced the amplitude of the P1 wave of the ABR by a minimum of 50% compared to the 0 dB SPL masker condition. A representative ABR FWM waveform series is presented in Fig. 1, displaying Off- and On-Freq conditions pre- and post-noise for a 70 dB SPL probe at 7 kHz. A small unpublished pilot study was performed prior to the current study in which the ABR FWM was tested weekly in three rats for four weeks to assess repeatability of the waveforms. The waveforms were consistent, and the assigned GoM thresholds did not vary from test to test. 2.3. Noise exposure One week following baseline ABR threshold and FWM testing, each animal was exposed to a noise of 110 dB SPL octave band (5e 10 kHz) continuous noise combined with 120 dB pSPL impacts.

66

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72

cages. Anesthesia was not used for the noise exposure in order to more closely model human noise exposure, and to minimize the number of anesthetic inductions each animal experienced over the duration of the study. Food and water were available throughout the duration of the noise exposures. 2.4. Post-exposure measures Twenty-one days following the noise exposure, threshold shift was assessed by re-testing ABR thresholds at 5e15 kHz. FWM IeO functions and GoM functions were also measured in order to assess noise-induced changes on FWM patterns. 2.5. Statistical analysis

Fig. 1. Representative FWM series for a 70 dB SPL 7 kHz probe tone. A) Off-Freq prenoise; B) On-Freq pre-noise; C) Off-Freq post-noise; D) On-Freq post-noise. The white boxes are cursors used to measure the latency and amplitude of the waves. The point of a minimum 50% reduction in amplitude occurs with the 90 dB SPL masker in A, the 70 dB SPL masker in B, the 70 dB SPL masker in C, and the 70 dB SPL masker in D.

For all analyses, the 7 kHz and 10 kHz probe conditions were analyzed separately. For latency and amplitude masker level IeO functions, the curves were analyzed with three-factor repeated measures ANOVAs (On/Off-Freq  masker level  pre/post noise). Differences in FWM GoM functions were also analyzed with repeated measures ANOVAs. Differences in pre-exposure FWM GoM functions were analyzed with two-factor ANOVAs (variables On/Off-Freq masking  probe stimulus level). Differences in the masking functions before and after the noise exposures were analyzed with three-factor ANOVAs (variables On/Off-Freq masking  pre/post-noise  probe stimulus level). In all analyses, On- vs. Off-Freq, test time, and probe stimulus level were all treated as repeated measures. Two-way interactions involving the On- vs. Off-Freq masking groups were analyzed with paired samples t-tests with Bonferroni corrections to determine if the On-Freq functions were different from the Off-Freq functions. Two-way interactions of test time and probe stimulus level were analyzed with repeated measures ANOVAs analyzing each stimulus level separately to determine which stimulus levels showed differences when comparing pre- and post-noise. Pre-exposure amplitude and latency GoM functions were also analyzed with linear regression analyses to allow for comparison with human behavioral data. A p-value of <0.05 was considered significant in all analyses with the exception of the paired samples t-tests to which the Bonferroni correction was applied. 3. Results 3.1. Pre-noise latency and amplitude FWM IeO functions

Duration of the impacts was 30 ms with 1 ms of rise and 29 ms of fall time. The rate of the impacts was 1/s. The duration of the combined continuous and impact noise was 120 min. The noise was created on commercial visual design software (TDT RPvdsEX version 71) and then generated using a real time signal processor (TDT RP2), amplified by a power amplifier (Marathon DJ-5000, Marathon Professional, New York, NY). The noise signal was then delivered to a set of speakers (Vifa D25AG35 100 Dome Tweeter, Madisound Speaker Components, Inc., Middleton, WI) that were individually mounted on the sides of wire mesh cages (1200  1500  2800 ) in which the animals were held for the noise exposure. Each animal was housed separately in one of the wire cages for the duration of the exposure. The noise levels were calibrated at the level of the animals’ heads utilizing a calibrated sound level meter (LxT1, Larson Davis Inc., Depew, NY) and a 1/200 condenser microphone (Model 377B02, PCB Piezotronics Inc., Depew, NY). Additional measurements were made at various locations within each cage to assure that the noise level was distributed evenly throughout each cage. No points within any cage differed by more than 0.6 dB SPL. Animals were awake throughout the noise exposure and were able to move freely throughout their

Fig. 2 displays the pre-noise ABR P1 latency IeO functions for Off-Freq (Panels A and C) and On-Freq FWM conditions (Panels B and D). For On-Freq conditions for both the 7 kHz and 10 kHz probes, there are missing data points in the 40 and 50 dB SPL probe curves. Those missing data points denote conditions in which no P1 wave was consistently detectable across subjects due to the effects of the masker. This also occurred in the 7 kHz OffFreq 40 dB SPL probe condition with the masker level of 90 dB SPL. The statistical analyses revealed significant differences between the On- and Off-Freq IeO functions for the 40 and 50 dB SPL probe levels in both the 7 and 10 kHz conditions at the masker levels of 30 dB SPL and above. In the 60 and 70 dB SPL probe conditions, On- and Off-Freq differed at the 70 and 80 dB SPL masker levels. Fig. 3 displays the pre-noise ABR P1 amplitude IeO functions for Off-Freq (Panels A and C) and On-Freq FWM conditions (Panels B and D). The amplitudes are displayed as percentages of the amplitude of the 0 dB SPL masker condition in order to prove a visual indicator of the extent of masking that occurred in all conditions. Like in Fig. 2, the missing data points denote conditions

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72

67

Fig. 2. Latency IeO functions for the pre-noise FWM conditions. Each plot displays six curves, one for each probe level from 40 to 90 dB SPL with mean latency plotted as a function of masker level. A) 7 kHz Off-Freq; B) 7 kHz On-Freq; C) 10 kHz Off-Freq; D) 10 kHz On-Freq. Error bars are 1 SEM.

Fig. 3. Amplitude IeO functions for the pre-noise FWM conditions. Amplitudes are expressed as a proportion of the amplitude of the 0 dB SPL masker condition. Each plot displays six curves, one for each probe level from 40 to 90 dB SPL with mean amplitude percentages plotted as a function of masker level. A) 7 kHz Off-Freq; B) 7 kHz On-Freq; C) 10 kHz OffFreq; D) 10 kHz On-Freq. Error bars are 1 SEM.

68

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72

in which no P1 wave was consistently detectable across subjects due to the effects of the masker. The significant differences detected between the Off-Freq and On-Freq conditions occurred at the 40e 70 dB SPL masker levels for the 40 and 50 dB SPL probes; the 50e70 dB SPL masker levels for the 60 and 70 dB SPL probes; and the 80 and 90 dB SPL masker levels for the 80 and 90 dB SPL probes. 3.2. Pre-noise latency and amplitude GoM functions Pre-noise GoM functions are displayed in Fig. 4. GoM functions derived from latency shifts for the 7 kHz probe (Panel A) and the 10 kHz probe (Panel B) and GoM functions derived from amplitude changes (Panels C and D) are displayed. Data are plotted as ABR probe stimulus level on the x-axis against the masker level that induced a 0.07 ms shift in latency (Panels A and B) or a reduction in P1 wave amplitude to 50% or less of the 0 dB SPL masker condition (Panels C and D). Qualitatively, the latency GoM and amplitude GoM functions are similar to human behavioral data presented in Oxenham and Plack (1997). The On-Freq conditions both map very close to a linear function. The Off-Freq functions show a compressive non-linearity with higher masker levels required to mask even the lowest stimulus presentation levels. A two-way ANOVA comparing the On-Freq and Off-Freq amplitude GoM functions at 7 kHz revealed a significant two-way interaction. Paired samples ttests with Bonferroni corrections were used to compare On-Freq to Off-Freq at each probe stimulus level. At probe levels 40e80 dB SPL, the On-Freq condition showed significantly lower masker thresholds than the Off-Freq condition. There was no difference between On-Freq and Off-Freq at the 90 dB SPL probe level. For the 7 kHz latency GoM functions, the On-Freq condition had lower masker

thresholds than the On-Freq condition at the 40, 60, 70, and 80 dB SPL probe levels, with the 50 and 90 dB SPL probe levels showing no differences in masker threshold. The ANOVAs of the 10 kHz amplitude and latency GoM functions also found significant twoway interactions. The paired samples t-tests with Bonferroni corrections found that the On-Freq condition had lower masker thresholds than the Off-Freq condition at all probe levels except 90 dB SPL for both the amplitude and latency GoM functions. Results from linear regression analyses are displayed in Table 1. For the On-Freq functions, the y-intercepts ranged from 11.57 to 7.69, while the Off-Freq y-intercepts ranged from 58.57 to 71.51. Those differences indicate more compressive functions for the OffFreq conditions compared to the On-Freq. The On-Freq regression slopes were also closer to a linear value of 1 (range of 0.795e1.138) than the Off-Freq slopes (range of 0.161e0.487). 3.3. FWM latency and amplitude IeO function changes following noise exposure Fig. 5 displays the ABR PTS in the 5e15 kHz range after the noise. Mean threshold shifts were between 20 and 35 dB. Following ABR threshold shift assessment, latency and amplitude masker level Ie O functions were measured, and are displayed in Figs. 6 and 7, respectively. In both data sets, only the 60e90 dB SPL probe conditions are displayed. Few subjects had any detectable responses in the 40 and 50 dB SPL probe conditions even with the 0 dB SPL masker. Like in the pre-noise IeO functions, the lower probe levels’ (60 and 70 dB SPL) curves are missing data points due to the absence of consistently detectable responses across subjects in conditions with higher masker levels. For the 7 kHz latency IeO

Fig. 4. Pre-noise exposure FWM GoM functions for the On-Freq and Off-Freq conditions for probe frequencies of 7 kHz and 10 kHz. Panels A and B display 7 kHz and 10 kHz GoM functions derived using the masker level required to induce a 0.07 ms shift in P1 latency as the masker threshold. Panels C and D display the 7 kHz and 10 kHz GoM functions derived from the masker level required to induce a minimum 50% reduction in P1 amplitude from the 0 dB SPL masker condition. Error bars are 1 SEM.

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72 Table 1 Regression data for latency and amplitude FWM GoM functions for the 7 and 10 kHz probes prior to noise exposure. Condition

y-intercept

Slope

R2

7 kHz Off-Freq latency GoM function 7 kHz On-Freq latency GoM function 10 kHz Off-Freq latency GoM function 10 kHz On-Freq latency GoM function 7 kHz Off-Freq amplitude GoM function 7 kHz On-Freq amplitude GoM function 10 kHz Off-Freq amplitude GoM function 10 kHz On-Freq amplitude GoM function

71.51 2.36 68.42 7.69 58.57 11.57 70.27 4.22

0.161 0.963 0.181 0.795 0.487 1.138 0.349 0.978

0.042 0.758 0.110 0.756 0.534 0.818 0.444 0.830

functions, statistical analysis revealed that the On-Freq and OffFreq conditions post-noise differed significantly in the 60 dB SPL probe condition with the 30e50 dB SPL maskers; the 70 dB SPL probe at the 70 dB SPL masker level, and the 90 dB SPL probe at the 90 dB SPL masker condition. For the 10 kHz latency I/O functions, differences between On-Freq and Off-Freq occurred in the 60 dB SPL probe at the 30 and 40 dB SPL masker levels; and in the 90 dB SPL probe condition with the 90 dB SPL masker. For the amplitude IeO functions, variability in the responses post-noise was higher than the pre-noise condition. Statistical analyses of the 7 kHz IeO functions revealed that the On-Freq and Off-Freq conditions differed in the 70 dB SPL probe condition at the 70 dB SPL masker level, and in the 80 and 90 dB SPL probe conditions at the 90 dB SPL masker. In the 10 kHz functions, differences occurred only in the 70 dB SPL probe condition at the 40e70 dB SPL masker levels. 3.4. FWM latency and amplitude GoM functions following noise exposure

69

levels from 60 dB SPL to 90 dB SPL were included, because the postexposure animals’ threshold shifts eliminated responses at 40 and 50 dB SPL. The three-factor ANOVA assessing the 7 kHz latencyderived GoM functions before and after noise revealed significant two-way interactions of probe stimulus level and On/Off-Freq and probe stimulus level and pre/post noise. Post hoc testing revealed significant differences between the pre-noise and post-noise OffFreq GoM functions at the 60 and 80 dB SPL probe levels. The OnFreq function did not change from pre- to post-noise. Comparing the post-noise Off-Freq function to the post-noise On-Freq function, the two functions were not statistically different from one another. In the 7 kHz amplitude GoM, the post-noise Off-Freq function was significantly different from the pre-noise function at the 60e 80 dB SPL probe levels. The On-Freq functions did not change after the noise exposure. The post-noise Off-Freq function differed from the post-noise On-Freq function only at the 70 dB SPL probe level. Similar results were found in the 10 kHz condition GoM functions. In the latency-derived GoM functions, the pre-noise Off-Freq GoM function differed from the post-noise function at all levels, 60e90 dB SPL. As in the 7 kHz condition, the pre-noise and postnoise On-Freq functions were not significantly different from one another. The comparison of the post-noise Off-Freq function to the post-noise On-Freq function revealed that the two functions were not statistically different from one another at any probe level. For the 10 kHz amplitude-derived GoM functions, the same pattern of results as the 10 kHz latency-derived GoM functions was found. The pre-noise Off-Freq function differed from the post-noise function at 60e90 dB SPL. The post-noise On-Freq function did not differ from the pre-noise On-Freq or post-noise Off-Freq functions. 4. Discussion

Post-noise On- and Off-Freq FWM GoM functions were calculated using both latency and amplitude changes. Results are displayed in Fig. 8 (latency-based GoM functions in Panels A and B, amplitude-based in Panels C and D) with the pre-noise GoM functions also displayed for visual comparison of the changes induced by the noise damage. The post-noise Off-Freq masker thresholds were lowered in both the 7 and 10 kHz conditions, rendering the GoM functions more linear and qualitatively similar to the On-Freq conditions. For statistical analyses, only the probe

4.1. Cochlear compression as measured by FWM of the ABR The current study utilized ABR FWM GoM functions to assess cochlear compression. The current physiological work showed FWM GoM functions qualitatively similar to those obtained in behavioral studies in humans (Oxenham and Plack, 1997) when using either latency or amplitude of the ABR P1 wave. The pre-noise On-Freq condition GoM functions were approximately linear for both the 7 and 10 kHz probe frequencies using both latency and amplitude changes to generate the GoM functions. As displayed in Table 1, the regression lines had y-intercepts ranging from 11.57 to 7.69, and slopes ranging from 0.795 to 1.138. Those slopes are generally in the range of those obtained from human behavioral data by Oxenham and Plack (1997) in which the slopes ranged from 0.96 to 1.08. The only slope value that represents an outlier from that range is the 0.795 obtained in the 10 kHz On-Freq latency GoM function. The Off-Freq condition in the current study displayed the compressive non-linearity that was observed in the FWM growth functions collected by Oxenham and Plack (1997). The regression lines had y-intercepts ranging from 58.57 to 71.51, and slopes ranging from 0.161 to 0.487. The slopes are lower than the Off-Freq data from individual listeners reported by Oxenham and Plack (1997) that showed regression lines of 0.58e0.91. 4.2. Noise-induced changes in FWM GoM functions

Fig. 5. Permanent threshold shift of the auditory brainstem response induced by the 2-h, 110 dB SPL 5e10 kHz octave band noise combined with the 120 dB pSPL impacts. Error bars are þ1 SEM.

The rats in the current study were exposed to a noise that was intended to induce significant PTS across the 5e15 kHz frequency range. The PTS-inducing noise had no significant effects on the shapes of the On-Freq GoM functions, as they were not significantly different pre-noise versus post-noise. The Off-Freq GoM functions

70

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72

Fig. 6. Latency IeO functions for the post-noise FWM conditions. Each plot displays four curves, one for each probe level from 60 to 90 dB SPL with mean latency plotted as a function of masker level. A) 7 kHz Off-Freq; B) 7 kHz On-Freq; C) 10 kHz Off-Freq; D) 10 kHz On-Freq. Error bars are 1 SEM.

Fig. 7. Amplitude IeO functions for the post-noise FWM conditions. Amplitudes are expressed as a proportion of the amplitude of the 0 dB SPL masker condition. Each plot displays six curves, one for each probe level from 40 to 90 dB SPL with mean amplitude percentages plotted as a function of masker level. A) 7 kHz Off-Freq; B) 7 kHz On-Freq; C) 10 kHz OffFreq; D) 10 kHz On-Freq. Error bars are 1 SEM.

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72

71

Fig. 8. On-Freq and Off-Freq FWM GoM functions following the noise exposure. Pre-exposure GoM functions are also included to allow comparison with post-exposure. Panels A and B display 7 kHz and 10 kHz GoM functions derived using changes in latency to determine masker threshold. Panels C and D display the 7 kHz and 10 kHz GoM functions derived using changes in amplitude to determine masker threshold. Error bars are 1 SEM.

were significantly altered by the noise. They became more linear, and largely statistically indistinguishable from the post-noise OnFreq FWM GoM functions in the same animals. The changes reflect a loss of cochlear compression. These results are consistent with the human behavioral data collected by Oxenham and Plack (1997) from subjects with sensorineural hearing losses. The underlying pathology for the NIHL in the rats from the current study can be attributed heavily to pathology of the OHCs (Chen and Henderson, 2009) in the form of a combination of dead OHCs and living OHCs that are dysfunctional (Chen and Fechter, 2003). No specific underlying pathology was identified in the human subjects in whom FWM was tested (Oxenham and Plack, 1997), but with an identified cochlear origin, some OHC pathology can be considered likely. The current study measured amplitude and latency changes in the P1 wave of the rat ABR. P1 was chosen because the noiseinduced changes were expected to be primarily cochlear in origin, and the P1 wave reflects cochlear activity and excitation of the auditory nerve (Jewett and Williston, 1971; Møller et al., 1988). Future studies investigating later waves in the ABR response may reveal different patterns in the FWM GoM functions both pre-noise and post-noise.

4.3. Practical implications of the findings The findings from the current study demonstrate that human behavioral FWM GoM functions can be replicated physiologically in animal models. This corroborates other FWM findings that manipulated FWM variables using the ABR in animal models (Gorga et al., 1983; Walton et al., 1995; Duan and Canlon, 1996a,b,c). The key distinction in the current study is the use of multiple probe

and masker levels in combination with one another to give a full GoM function. The findings of significant changes in the FWM GoM functions with noise-induced PTS are consistent with human behavioral data (Oxenham and Plack, 1997) reflecting a loss of cochlear compression. Ongoing investigations of noise-induced changes in Off-Freq GoM functions will target assessing the time course of the changes to determine if loss of the compressive non-linearity of the Off-Freq GoM functions occurs early in the development of NIHL. The use of animal modeling to test this phenomenon is valuable in that it allows for controlled noise doses to gradually create subclinical damage associated with small lesions of damaged/dead OHCs, and then eventually growing into measurable threshold shift as the OHC lesion expands. The non-linear Off-Freq FWM GoM function is the result of a non-linear response to a probe stimulus (Oxenham and Moore, 1995) being masked by a tone that induces a linear response at that characteristic frequency (Yates et al., 1990; Ruggero et al., 1997). By contrast, the linear On-Freq GoM functions are the product of a non-linear response to the probe being masked by a non-linear response to the masker. The current study’s use of maskers that were exactly one octave below the probe may not fully have created a linear response to the masker (Lopez-Proveda and Alves-Pinto, 2008) at the probe characteristic frequencies of 7 and 10 kHz. For an optimal non-linear Off-Freq GoM function, the masker must be inducing a purely linear response at the probe characteristic frequency. If the masker is inducing a slightly nonlinear response at the probe’s characteristic frequency, then the resulting Off-Freq GoM function will be slightly more linear than it would be with that optimal masker frequency. Thus, the current study may have created Off-Freq GoM functions that were slightly more linear than if maskers below the 3.5 and 5 kHz maskers

72

E.C. Bielefeld et al. / Hearing Research 294 (2012) 64e72

(exactly one octave below the 7 and 10 kHz probes) had been used. Additionally, the use of short-duration tone bursts reduces the frequency specificity of the probe. Therefore, to enhance frequency specificity and to have improved certainty that the responses observed do not reflect activation of basal frequency regions of the cochlear outside the frequency region under investigation, future studies on FWM of the ABR will employ high-pass masking noise to mask out the basal portion of the cochlea. Acknowledgements The authors thank Megan Kobel, Anna Kiener, Marie Neel, and Joseph Hribar for their assistance with the development of the test conditions and with data collection. Research was supported by a grant from the Office of Naval Research #N000140911. References Ahmad, M., Bohne, B.A., Harding, G.W., 2003. An in vivo tracer study of noiseinduced damage to the reticular lamina. Hear. Res. 175, 82e100. Brownell, W.E., Bader, C.R., Bertrand, D., de Ribaupierre, Y., 1985. Evoked mechanical responses of isolated cochlear outer hair cells. Science 227, 194e196. Chen, G.D., Fechter, L.D., 2003. The relationship between noise-induced hearing loss and hair cell loss in rats. Hear. Res. 177, 81e90. Chen, G.D., Henderson, D., 2009. Cochlear injuries induced by the combined exposure to noise and styrene. Hear. Res. 254, 25e33. Don, M., Ponton, C.W., Eggermont, J.J., Kwong, B., 1998. The effects of sensory hearing loss on cochlear filter times estimated from auditory brainstem response latencies. J. Acoust. Soc. Am. 104, 2280e2289. Duan, M.L., Canlon, B., 1996a. Outer hair cell activity is not required for the generation of the forward masking curve. Audiol. Neurootol. 1, 309e319. Duan, M.L., Canlon, B., 1996b. Forward masking is dependent on inner hair cell activity. Audiol. Neurootol. 1, 320e327. Duan, M.L., Canlon, B., 1996c. Differences in forward masking after a temporary and a permanent noise-induced hearing loss. Audiol. Neurootol. 1, 328e338. Gorga, M.P., McGee, J., Walsh, E.J., Javel, E., Farley, G.R., 1983. ABR measurements in the cat using a forward-masking paradigm. J. Acoust. Soc. Am. 73, 256e261. Hamernik, R.P., Turrentine, G., Roberto, M., 1986. Mechanically induced morphological changes in organ of Corti. In: Henderson, D., Hamernik, R.P., Colletti, V. (Eds.), Basic and Applied Mechanisms of Noise Induced Hearing Loss. Raven Press, New York, pp. 69e83. Hamernik, R.P., Turrentine, G., Roberto, M., Salvi, R., Henderson, D., 1984. Anatomical correlates of impulse noise-induced mechanical damage in the cochlea. Hear. Res. 13, 229e247.

Henderson, D., Bielefeld, E.C., Harris, K.C., Hu, B.H., 2006. The role of oxidative stress in noise-induced hearing loss. Ear. Hear. 27, 1e19. Howgate, S., Plack, C.J., 2011. A behavioral measure of the cochlear changes underlying temporary threshold shifts. Hear. Res. 277, 78e87. Jewett, D.L., Williston, J.S., 1971. Auditory-evoked far fields averaged from the scalp of humans. Brain 94, 681e696. Lopez-Proveda, E.A., Alves-Pinto, A., 2008. A variant temporal-masking-curve method for inferring peripheral auditory compression. J. Acoust. Soc. Am. 123, 1544e1554. Møller, A.R., 1981. Latency in the ascending auditory pathway determined using continuous sounds: comparison between transient and envelope latency. Brain Res. 207, 184e188. Møller, A.R., Jannetta, P.J., Sekhar, L.N., 1988. Contributions from the auditory nerve to the brain-stem auditory evoked potentials (BAEPs): results of intracranial recording in man. Electroencephalogr. Clin. Neurophysiol. 71, 198e211. National Institute for Occupational Safety and Health, 2001. Fact Sheet Publication No. 2001e2003. URL. http://www.cdc.gov/niosh/docs/2001-103/. National Institute on Deafness and Other Communication Disorders, 2008. Quick Statistics. U.S. Department of Health and Human Services, Bethesda, MD. Neely, S.T., Kim, D.O., 1983. An active cochlear model showing sharp tuning and high sensitivity. Hear. Res. 9, 123e130. Niskar, A.S., Kieszak, S.M., Holmes, A.E., Esteban, E., Rubin, C., Brody, D.J., 2001. Estimated prevalence of noise-induced hearing threshold shifts among children 6 to 19 years of age: the Third National Health and Nutrition Examination Survey, 1988e1994, United States. Pediatrics 108, 40e43. Ohinata, Y., Miller, J.M., Altschuler, R.A., Schacht, J., 2000. Intense noise induces formation of vasoactive lipid peroxidation products in the cochlea. Brain Res. 878, 163e173. Ohlemiller, K.K., Wright, J.S., Dugan, L.L., 1999. Early elevation of cochlear reactive oxygen species following noise exposure. Audiol. Neurootol. 4, 229e236. Oxenham, A.J., Moore, B.C.J., 1995. Additivity of masking in normally hearing and hearing-impaired subjects. J. Acoust. Soc. Am. 98, 1921e1934. Oxenham, A.J., Plack, C.J., 1997. A behavioral measure of basilar-membrane nonlinearity in listeners with normal and impaired hearing. J. Acoust. Soc. Am. 101, 3666e3675. Patuzzi, R., 2002. Non-linear aspects of outer hair cell transduction and the temporary threshold shifts after acoustic trauma. Audiol. Neurootol. 7, 17e20. Ruggero, M.A., Rich, N.C., Recio, A., Narayan, S.S., Robles, L., 1997. Basilar membrane responses to tones at the base of the chinchilla cochlea. J. Acoust. Soc. Am. 101, 2151e2163. Ryan, A., Dallos, P., 1975. Effect of absence of cochlear outer hair cells on behavioural auditory threshold. Nature 253, 44e46. Walton, J.P., Frisina, R.D., Meierhans, L.R., 1995. Sensorineural hearing loss alters recovery from short-term adaptation in the C57BL/6 mouse. Hear. Res. 88, 19e26. Yamane, H., Nakai, Y., Takayama, M., Konishi, K., Iguchi, H., Nakagawa, T., Shibata, S., Kato, A., Sunami, K., Kawakatsu, C., 1995. The emergence of free radicals after acoustic trauma and strial blood flow. Acta Otolaryngol. Suppl. 519, 87e92. Yates, G.K., Winter, I.M., Robertson, D., 1990. Basilar membrane nonlinearity determines auditory nerve rate-intensity functions and cochlear dynamic range. Hear. Res. 45, 203e219.