Amercian Auditory Society
homecontact
about aaslibraryresourcesjoin aas
top selections

search



 Search the Web

Search AAS Site
quick links
•Membership Renewal
•New Membership
•Graduate Student
 Application
•President's Message
highlights
2009 Meeting
AAS Meeting Info
& Abstracts (When Available)
Meeting Program
Meeting Registration
Call for Papers
Call for Papers (PDF)
Mentored Student
Poster Sessions
General Poster Sessions

 
 >> Home > Library > Abstracts > 2004 Mentored Poster Presentationa

MENTORED POSTER
PRESENTATION ABSTRACTS

March 8, 2004, 8:30 - 10:30 a.m.

NIH/AAS Funded Mentored Doctoral Research Poster Presentations
and Conventional Poster Presentations
GENERAL POSTER PRESENTATIONS
The Role of Experience in Elicitation of Perceptual Overshoot
Radhika Aravamudhan M.S.
John W. Hawks, Ph.D., Mentor
Kent State University, Kent, OH

Perceptual overshoot (PO), a phenomenon where the boundary of a vowel category shifts as a result of formant movement has been demonstrated previously with synthetic vowels in CV contexts. In this study, sinewave analogs were constructed to mimic the acoustic properties of a continuum of synthetic vowels known to elicit perceptual overshoot. Subjects failed to demonstrate perceptual overshoot with these sinewave analogs. Further investigation sought to determine if experience or further exposure to the sinewave analogs brought about any difference in their perception. To this end, one group of subjects was trained to categorize the sinewave analogs while the other group received no training. Post-training boundary estimates showed that the group, which received training, now demonstrated PO for the sinewave analogs while the control group did not. The results support the 'Auditory Enhancement Hypothesis' (Kluender, et. al., 1988), whereby experience might be a major factor influencing the perception of nonspeech sounds. The results will be discussed in detail in the poster.

Clinical Measures of Temporal Resolution Using Gaps-In-Noise (GIN)
Jennifer Brooke Shinn, M.S.
University of Connecticut, Storrs, CT
Robert E. Jirsa, Ph.D., Mentor
Southern Connecticut State University, New Haven, CT
Jane A. Baran, Ph.D.
University of Massachusetts, Amherst, MA
Frank E. Musiek, Ph.D.
University of Connecticut, Storrs, CT

A variety of psychoacoustic procedures are available to assess temporal resolution. However, the clinical use of these procedures is minimal at best. The purpose of the present study was to investigate the value of a new gap detection procedure called Gaps-In-Noise (GIN) for assessment of temporal resolution in a clinical population. The test consists of 1-to 3 silent intervals ranging from 2-20 msec. embedded in 6 second bursts of white noise. The location, number and duration of the gaps per noise burst vary throughout the test for a total of 60 gaps presented in each of the four available lists. The GIN procedure was administered to 50 normal listeners (Group 1), and 18 participants with confirmed neurological involvement of the central auditory nervous system (Group 2). Results showed a mean gap detection threshold of 4.9 (+ 1) msec. bilaterally for Group 1. This is consistent with previous findings for similar procedures reported in the literature. In comparison, results from Group 2 demonstrated a statistically significant increase in gap detection thresholds. Based on the results of the present study, the GIN test presents promise as a clinically useful tool in assessment of temporal resolution in the clinical arena.

Recovery From Forward Masking in a Continuous Broadband Noise
René H. Gifford, Ph.D.
Sid P. Bacon, Ph.D., Mentor
Arizona State University, Tempe, AZ

Individuals with hearing loss perform considerably worse than those with normal hearing on a speech recognition task in the presence of a temporally complex background. The role of reduced audibility has been examined by elevating the thresholds of normal-hearing listeners with a continuous, frequency-shaped noise to match those of the impaired listeners (e.g., Bacon et al., 1998). Audibility appears to account for some of the speech recognition deficit exhibited by hearing-impaired listeners. It is unclear, however, whether the shaped noise only elevates threshold, or whether it also affects suprathreshold processing. This study examined the effects of such a noise on forward masking. Hearing-impaired, normal-hearing, and noise-masked normal-hearing listeners were employed. The signal was a 20-ms, 2-kHz tone masked by a 300-ms octave-wide noise centered at the signal frequency or one-octave below it. Signal delays ranged from 5 to 160 ms. Results revealed that although the frequency-shaped continuous noise influenced the recovery function for the normal-hearing listeners, forward-masked thresholds were much higher for the hearing-impaired listeners than for the noise-masked normal-hearing listeners at short delays.

Asymmetric Suppression of Distortion Product Otoacoustic Emissions (DPOAE)
Cassie A. Garner, B.A.
T. Newell Decker, Ph.D., Mentor
University of Nebraska-Lincoln, Lincoln, NE

Virtually all of what is known about suppression of Otoacoustic Emissions (OAE) is based upon band limited (narrow band, white) noise stimuli. Under normal conditions of life the human OAE suppression system is not activated exclusively by band-limited noise but rather by competing speech signals. The use of a meaningful speech stimulus as a suppressor of DPOAEs has been studied little. In addition, little information is available regarding any asymmetric effects produced by suppression stimuli at each ear. Our aim in this study was to determine if speech and or noise caused suppression of OAEs is different at the two ears.

The DPOAE's of 18 participants were measured with and without speech noise and running speech suppressor stimuli. DPOAE's were measured using unequal primaries (L1 = 60 dB SPL, L2 = 45 dB SPL) and the suppressor stimuli were presented contralateral at 60 dB SPL.

Results reveal significant suppressive effects of the speech noise at both ears with greater suppression occurring in the left ear. Suppressive effects of the running speech were only observable in the right ear.

"Correcting" Sound for the Hearing-Impaired:
A Physiologically-Based Signal-Processing Strategy
Lufeng Shi, M.S.
Laurel H. Carney, Ph.D.
Karen A. Doherty, Ph.D., Mentor
Syracuse University, Syracuse, NY

Typical hearing aids amplify to compensate for loss of gain and/or sensitivity in the cochlea, but they do not purposefully manipulate the phase spectrum. In this study, the phase spectrum of complex sounds was manipulated based on knowledge of the level-dependent temporal response properties of auditory-nerve (AN) fibers in normal and impaired ears. This approach attempts to "correct" a response patterns by introducing time-varying phase delays that differ across frequency. Sentences from the Hearing in Noise Test (HINT) and vowel-consonant (VC) syllables from the nonsense syllable test (NST) were used as stimuli. Stimuli were processed at different "corrections", i.e. maximum phase delays introduced to the input signal. In the first half of the study, hearing-impaired (HI) and normal-hearing (NH) listeners judged the quality of HINT sentences. Different HI listeners preferred stimuli processed at different corrections, whereas NH listeners preferred less corrected stimuli. In the second half, VC syllables were presented to HI listeners. Listeners' speech intelligibility and clarity rating were measured. In general, correction improved HI listeners' speech intelligibility and clarity rating for some VCs. Research supported by grant #R21 DC006057.

Performance Differences Among CII, 3G, and Tempo+ Cochlear Implant Users
Anthony J. Spahr, M.A.
Michael Dorman, Ph.D., Mentor
Arizona State University, Tempe, AZ

The perception of speech, voice and music was assessed in subjects fit with (a) the Advanced Bionics Corporation's CII BTE with HiResolution, (b) the Cochlear Corporation's 3G BTE and (c) the Med El Tempo+ BTE. The groups were matched on CNC word scores in quiet. The groups did not differ in mean age at testing, mean length of deafness, or mean length of experience with electrical stimulation. At issue was whether the groups differed in performance on tests of detailed spectral resolution, on tests of detailed temporal resolution, speech presented in noise, or speech presented at low signal levels.

Significant differences in performance were found for sentence material presented at 74 dB SPL at +5 dB SNR (CII mean = 41.2%, 3G mean = 20.5%, Tempo+ mean = 46.5%) and a measure called the robustness index (CII mean = 82.6, 3G mean = 61.5, Tempo+ mean = 81.1). The robustness index is a measure used to describe drops in performance in difficult listening situations (74 dB in +10 SNR and 54 dB in quiet), relative to performance in an ideal listening situation (74 dB in quiet). Higher robustness scores indicate smaller differences between performance in ideal and difficult listening situations.

Multitaper Spectral Analysis of ASSR Data
J. Kip Kelly, M.S.
Wayne M. King, Ph.D., Mentor
Ohio State University, Columbus, OH

In this paper we introduce a new statistical test for the detection of the ASSR. This represents a novel application of multitaper spectral methods to the analysis of ASSR data. Specifically, we model the ASSR data in the frequency domain as a complex-valued regression onto optimally band limited prolate spheroidal (Slepian) wave functions and employ the usual goodness of fit F-statistic based on the theory of complex least squares to detect the ASSR at the modulation frequency. We compare the multitaper F-test against the conventional F-test based on the single-tapered Discrete Fourier Transform (DFT), which is the default analysis algorithm for the equipment used in this study (IHS SmartEP ASSR System). Detector performance was assessed in both Monte Carlo simulations and in normal-hearing human subjects. Results indicate that the multitaper F-test statistic offers some advantage over the conventional F-test in terms of detection, while both estimators are comparable in the expected number of false positives.

Human Efferent Adaptation of DPOAEs in the L1,L2 Space
Deanna Meinke, M.A.
Barden Stagner
Glen Martin, Ph.D., Mentor
University of Colorado Health Sciences Center, Denver, CO

The adaptation properties of the 3f1-2f2, 2f1-f2, and the 2f2-f1 distortion product otoacoustic emissions (DPOAEs) in response to 1-s stimulus on-times were investigated in 12 ears of normally hearing humans. DPOAEs were elicited under both monaural and binaural stimulus-presentation conditions at f2=1.55 kHz (f2/f1=1.21). Primary-tones were varied in a matrix of L1 and L2 levels in 5-dB steps between 45 and 80 dB SPL. Only the 2f1-f2 emission revealed consistent regions of suppression or enhancement within the L1,L2 matrix for both the monaural and binaural presentation conditions. While the 2f1-f2 binaural condition revealed broader regions of enhancement than the monaural condition, adaptation contours were generally consistent across both the monaural and binaural conditions. For subjects with measurable emissions at 2f2-f1, adaptation was evident only in the binaural condition for equilevel high-intensity primaries >70 dB SPL. The 3f1-2f2 emission was present in a subset of subjects, however the narrow L1,L2 range that evoked this emission limited analysis. Overall, the findings suggest that the optimal L1,L2 levels to measure DPOAE adaptation may be different than those typically used to maximize DPOAE magnitude.

Effects of Noise-excited-vocoder Processing on F0 Discriminability and Utility
Michael K. Qin
Andrew J. Oxenham, Ph.D., Mentor
Massachusetts Institute of Technology, Cambridge, MA

Although enormous progress has been made in the development of cochlear implants, the speech reception of implant users is still not comparable to that of normal-hearing listeners. The difference in performance is especially pronounced in complex auditory situations, such as in the presence of competing talkers. The poorer speech reception performance of implant users in complex environments may, at least in part, be due to the poorer fundamental frequency (F0) representation.

The present study examined the effects of implant-like processing (i.e. noise-excited vocoding) on F0 discriminability and utility in normal-hearing listeners. F0 difference limens (F0DLs) were measured as a function of the number of vocoder channels. In addition, vowel identification was measured as a function of F0 difference between competing vowels in a double-vowel paradigm, with number of vocoder channels as a parameter. Our findings show that despite the reasonable F0DLs (< 1 semitone) with 24- and 8-channel vocoder processing, listeners were unable to benefit from F0 differences between the competing vowels in a double-vowel paradigm. The implications of the findings will be discussed in the context of previous pitch and double-vowel studies, models of pitch perception and double-vowel identification, as well as cochlear implant design.

The Effect of Release Time on Speech Acoustics and Intelligibility
Lorienne M. Jenstad, MCLSC
Pamela E. Souza, Ph.D., Mentor
University of Washington, Seattle, WA

Compression hearing aids have the inherent and adjustable feature of release time. Research to date does not provide a consensus on how to choose or set release time. The current study focused on two purposes: 1) evaluating the acoustic effects of release time (12 ms, 100 ms, and 800 ms), and 2) evaluating the relationship between the acoustic changes and speech intelligibility. The stimuli were vowel-consonant nonsense syllables spoken by a female talker at three input levels to the hearing aid. Two acoustic measures were made on individual speech segments: the Envelope Difference Index (EDI) and consonant-vowel ratio (CVR). These measurements allow for quantification of the short-term amplitude characteristics of speech and the changes caused by compression. The acoustic analyses revealed statistically significant effects among the three release times. The size of the effect depended upon the input level and characteristics of the phoneme. The acoustic measurements were significant predictors of phoneme recognition. The recognition of some phonemes was affected more than others by the acoustic changes. Depending on the phoneme, support was found for two competing hypotheses: that shorter release times increase phoneme audibility and thus benefit, and that shorter release times increase temporal distortion to unacceptable levels.

Age Effects on Speech Recognition in Temporally Complex Background Noise
Kumiko T. Boike, M.S.
Pamela E. Souza, Ph.D., Mentor
University of Washington, Seattle, WA

In general, younger normal-hearing listeners are able to take advantage of background noise modulation to improve speech understanding. However, some research has suggested that certain modulated noises may have an added masking effect not present in an unmodulated noise. The current study was completed to examine this effect by comparing the speech recognition of younger and older listeners with both normal hearing and hearing impairment when the speech was presented in background noises varying in temporal complexity. Sentence recognition was tested in three different noises (i.e., single-talker modulated, multitalker modulated, and unmodulated) and at a range of signal-to-noise ratios. The results suggest that both age and hearing impairment affected the listeners' abilities to understand speech in noise. However, age was the only factor influencing a person's ability to understand speech in a modulated noise. That is, the younger listeners had more difficulty recognizing speech in the multitalker modulated noise than in the unmodulated and the single-talker modulated noise; whereas, the older listeners' speech recognition was relatively insensitive to the amount of modulation in the background noise.

Effects of High-pass Filtering and Masking Noise on Word Recognition
Saravanan Elangovan, M.Sc.
Andrew Stuart, Ph.D., Mentor
East Carolina University, Greenville, NC

We examined the word recognition performance in noise of individuals with a simulated low-frequency hearing loss. The ulterior motive was to understand how a low-frequency hearing impairment affects tasks that challenge temporal processing skills. Twenty-one normal hearing adults participated. Monosyllabic words were presented in spectrally identical continuous and interrupted noise at three signal-to-noise ratios (S/Ns) of -10, 0, +10 dB. High-pass filtering with three different cutoff frequencies simulated the low-frequency hearing impairment (i.e., 1000, 1250, and 1500 Hz). In general, performance decreased with increasing cutoff frequency, was higher for more favorable S/Ns, and was superior in the interrupted noise condition. The most important revelation was that the magnitude of the performance superiority observed in the interrupted noise did not diminish with high-pass filtering (i.e., the release from masking in interrupted noise was preserved). Previous investigations utilizing this paradigm have demonstrated that the loss of high-frequency spectral information with simulated hearing loss and/or pathologies diminishes listener superiority in the interrupted noise condition. The results of the present study complement these findings viz. low-frequency hearing channels are inherently poorer than high-frequency channels for temporal resolution.

Infant Auditory Filter Widths Obtained Using Auditory Brainstem Response
Carolyn Tooley-Young, M.S.
Richard C. Folsom, Ph.D., Mentor
Lynne A. Werner, Ph.D.
University of Washington, Seattle, WA

Auditory filters measure frequency resolution, which is described as the ability to distinguish individual sounds in the presence of other sounds differing in frequency. Frequency resolution is necessary for the clear understanding of speech, whether in quiet or in background noise. A common psychophysical measurement of frequency resolution is to create auditory filter widths (Patterson, 1974). Due to the need for subjects in this type of an experiment to participate in a sophisticated task, auditory filter widths have not been measured in infants since they are often unable to perform the necessary complex behavioral tasks. A method using auditory brainstem response (ABR) to obtain auditory filter widths as a measure of frequency resolution has been developed (Tooley-Young, Folsom and Werner, 1998). The use of the ABR is a non-invasive and straightforward way to measure frequency resolution, which is particularly helpful for testing infants. Minimal cooperation is needed during testing. The purpose of this study is to see if previously documented developmental stages of frequency resolution will also be observed using the ABR auditory filter width method. Comparisons of ABR auditory filter width data obtained in three- and six-month old infants will be made to published behavioral and physiologic frequency resolution data in the literature. Application of the ABR to study auditory filter widths will provide insight into the development of frequency resolution in infants and, in the future, may also be applicable for testing certain populations of adults as well,

Neural Representation of Amplified Speech-Sounds
Curtis J. Billings, M.S.
Kelly Tremblay, Ph.D., Mentor
Pamela E. Souza, Ph.D.
University of Washington, Seattle, WA

The Acoustic Change Complex (ACC), an auditory cortical evoked potential consisting of multiple N1-P2 responses, reflects time-varying acoustic changes in speech (Martin & Boothroyd, 1999). For this reason, there is interest in using this response to monitor changes in neural representation of speech in conjunction with various forms of rehabilitation (e.g. amplification, auditory training). However, if the ACC is to be used to examine longitudinal changes in neural activity following rehabilitation, it is first critical to determine: 1) ACC test-retest reliability when evoked by amplified speech tokens, 2) whether different speech-sounds evoke distinct neural patterns, and 3) if amplification alters the neural response patterns.

The ACC was recorded from seven normal-hearing young adults in response to tokens 'see' and 'shee' from the Nonsense Syllable Test (NST). Using a repeated measures design, subjects were tested and then retested in sound field within an eight-day period. In both aided and unaided conditions, we found: 1) the ACC can be reliably recorded in individuals with intraclass correlations comparing test-retest neural responses ranging from .60 to .99, 2) naturally-produced speech syllables 'see' and 'shee' evoked distinct ACC patterns, and 3) amplification alters neural representation of vowel onset.

Frequency Discrimination of Electrical Stimulation Using Three Pulse Patterns
Hongbin Chen, M.S.
Fan-Gang Zeng, Ph.D., Mentor
University of California, Irvine, CA

One significant limitation for current cochlear-implant users is poor pitch perception. Here we hypothesize that the highly synchronized neural firing and the lack of statistically independent neurons in electric hearing are the cause for the poor pitch perception in cochlear-implant users. Three different stimulation patterns were used in the frequency discrimination test on Nucleus-24 cochlear-implant users in an attempt to restore normal neural responses. The output of the auditory model, which had more normal-like neuron firing pattern, was utilized to modulate pulse trains. For comparison, stochastic pulses (probability = 1, 0.8, 0.5,0.3) and pulses with Gaussian distributed rate (standard deviation = 0,0.1, 0.2,0.3) were also used in the experiment. Multiple electrodes were tested up to four electrodes with electrodes spacing as adjacent, one electrode apart and two electrodes apart. Each electrode carried same temporal information but they were independent from each other. The pilot data showed frequency discrimination using auditory model were much worse than that of regular fixed rate. Better frequency discrimination was found at high frequency for stochastic pulse condition. The data also showed that frequency discrimination was not affected by adding little frequency jitter in the stimuli at low frequency. At high frequency (1000 Hz), adding appropriate frequency jitter may improve frequency discrimination. No improvement was observed between single electrode and multiple electrodes, suggesting additional independent channels did not provide more information.

GENERAL POSTER PRESENTATIONS

New Norms for Abbreviated Profile of Hearing Aid Benefit (APHAB)?
Yu-Hsiang Wu, M.D.
Ruth A. Bentler, Ph.D.
The University of Iowa, Iowa City, IA

The Abbreviated Profile of Hearing aid Benefit (APHAB) (Cox & Alexander, 1995) has been used extensively in clinic and research endeavors over the past eight years. The original norms were obtained on 55 subjects deemed "successful hearing aid users" of linear technologies. Yet, the norms are applied to many other groups, including private pay (PP), VA patients (VA) and research subjects (RS) in clinical trials and other evaluations of newer technologies. It is unclear whether the norms are appropriate for current hearing aid users, or whether the norms can be applied across different populations (e.g., PP, VA and RS). The purpose of this ongoing study is to create a new set of norms for each group. We currently have RS norms from two sites (The University of Iowa Hearing Aid Laboratory for Basic and Applied Research, and University of Memphis Hearing Aid Research Laboratory). A total of 250 completed APHAB inventories were collected from the sites. Similar data are being collected from two independent private practice centers and two VA centers. Data will be presented here only for the RS group from The University of Iowa.

Risk Factor Abnormalities In Patients Who Have Undergone Recent Falls
Lynn S. Alvord, Ph.D.
Henry Ford Hospital, Detroit, MI

Patients attending a Risk of Falls clinic who had recently fallen were evaluated on various risk factors and test performance scores with the ultimate goal of attributing causes of falls. Assessment of the following factors was performed: previous falls, multiple medications, major illness, metabolic, cardiogenic, CNS, muscular weakness or degeneration, vestibular, visual, osculoskeletal, pulmonary, environmental, depression, footwear and alcohol ingestion. Detailed evaluation of vestibular, oculomotor control and postural maintenance was made using computerized dynamic posturography, VNG and rotary chair testing. Abnormalities on these factors are reported descriptively for the group and comparatively within patients. It is hoped that this analysis will aid in the determination of which tests to include in a comprehensive falls assessment clinic.

Novel Acoustic Test Fixture to Evaluate Viscoelastic Foams
Vasant V. Kolpe, Ph.D.
Gregory Hoeker
Hearing Components Inc., Oakdale, MN

The utility of a novel acoustic test fixture (ATF) was presented at the AAS last year. The main emphasis was to evaluate acoustic attenuation at audio frequencies for cylinders of earmold materials at body temperature. Compliant viscoelastic foam at body temperature was found to have superior attenuation properties.

We have continued testing the ATF in detail to understand its primary acoustic characteristics. This ability, when integrated with the FONIX(R)6300-CX, to rapidly screen viscoelastic foams will be documented. The ATF also provides performance characteristics of hearing aids coupled to the compressible viscoelastic foam. The screening process is carried out over several temperatures employing the pure tone as well the as composite mode for testing. These acoustic characterization data, especially viscoelastic foams, strongly support the existence of energy loss process in foams. The mechanism of viscoelastic loss in foams will be further documented in the future.

Hearing Aid Satisfaction and its' Relationship to
Expectation, Performance, Benefit, and Disconfirmation
Lena L N Wong, M.S.
University of Hong Kong, Hong Kong, China
Louise Hickson, Ph.D.
University of Queensland, Brisbane, Australia
Bradley McPherson, Ph.D.
University of Hong Kong, Hong Kong, China

The relationships between satisfaction and expectation, performance, benefit, and disconfirmation were examined. Expectations about the amount of help provided by hearing aids in specific situations, about problems with hearing aid use, cost and dispenser service were obtained prior to fitting. Performance and benefit in the same aspects were measured post-fitting. Disconfirmation, which has not been assessed in previous hearing aid studies, is defined as the deviation of perceived benefit from expectation. A detailed questionnaire examining these issues and several conventional measures (e.g., IOI-HA, PHAB) were administered to elderly participants with moderately-severe hearing loss prior to and/or at 3 weeks, 6 weeks, 3 months, 6 months and 12 months post-fitting.

Preliminary results on 36 participants showed no significant change in performance, disconfirmation and satisfaction over the study intervals. Results obtained at 3 weeks post-fitting show high satisfaction with hearing aids. Satisfaction with specific aspects was significantly related to general satisfaction. Pre-fitting expectations did not affect post-fitting satisfaction. Better performance, and positive disconfirmation, defined as a perception that the hearing aid is performing better than expectation, was associated with greater satisfaction. Benefit did not always predict satisfaction. While findings are consistent with reports in the literature, this study also suggests the importance of considering the effect of disconfirmation in hearing aid satisfaction.

Calcium and Vitamin D Deficiency Impair Auditory Function in Mice
Sohyun Park, M.S.
Albert R. De Chicchis, Ph.D.
Houssam Marseli, D.V.M.
Mary Ann Johnson, Ph.D.
University of Georgia, Athens, GA
James F. Willott, Ph.D.
University of South Florida and The Jackson Laboratory, Tampa, FL

Previously, we have reported on the effects of vitamin D status and hearing loss in several strains of mice carrying a gene for age-related hearing loss. Small differences in hearing thresholds between BALB/c mice fed vitamin D adequate and vitamin D deficient diets were found. This study examined the effects of vitamin D and calcium deficient diets on hearing sensitivity in three groups of BALB/c mice. Starting at the 14th day of pregnancy, female BALB/c mice were fed one of three different purified diets: 1) adequate in vitamin D, 2) deficient in vitamin D and CA (0.2% Ca), and 3) deficient in vitamin D and Ca (0.1% Ca). This diet regimen continued throughout the remaining days of pregnancy, lactation and weaning, and for each of the three groups of pups after weaning. Auditory brainstem response measurements were obtained for five test frequencies 4, 8, 16, 24 and 32 kHz. Four weeks following weaning, the mice being fed the adequate diet had better hearing thresholds than the vitamin D and 0.1% Ca deficient group at 4, 24, & 32 kHz (p < 0.05). In summary, deficiencies of vitamin D and Ca during pregnancy, lactation, and weaning may cause auditory dysfunction in offspring. Research supported by USDA-NRICGP #2001-35200-10677.

Real-Time Statistical Analysis to Optimize Sound Within the Dynamic Range
Lois Martin, Ph.D.
Peter Blamey, Ph.D.
Dynamic Hearing Pty Ltd., Richmond, Australia
Todd Fortune, Ph.D.
Interton-USA, Casselbury, FL

An aim of non-linear amplification is to place a wide range of inputs into a narrower output dynamic range for hearing-impaired listeners. A new approach to solve this problem uses real-time statistical analysis of the amplified output signal. Gain is increased if the analysis shows the 30th percentile of the signal falls below audible levels for the listener. Gain is decreased if the 90th percentile of the signal reaches levels above a specified comfort target, which are consequently too loud for the listener. Sound is optimized within the dynamic range by slow gain adaptation in 64 independent frequency channels. Results for comparisons with linear amplification, with 15 subjects with moderate-to-profound hearing losses, showed a significant 36% improvement in sentence scores at 55 dB in quiet. A second study (N=10) showed a 6.7% improvement for sentence scores in babble. These advantages were replicated with 19 subjects in a field trial and evaluation of an open-platform digital BTE hearing aid when compared to a 3-channel WDRC scheme. A further study, with 22 subjects with mild-to-moderate hearing losses, compared this statistical approach with a 9-channel compression hearing aid and showed similar advantages for speech perception in quiet and in noise.

footer.gif

 
 

American Auditory Society

352 Sundial Ridge Circle • Dammeron Valley, UT 84783
Phone (435) 574-0062 FAX(435) 574-0063