Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
17 results
Search Results
Item EFFECTS OF AGING ON VOICE-PITCH PROCESSING: THE ROLE OF SPECTRAL AND TEMPORAL CUES(2010) Schvartz, Kara Chantal; Chatterjee, Monita; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Declines in auditory temporal processing are a common consequence of natural aging. Interactions between aging and spectro-temporal pitch processing have yet to be thoroughly investigated in humans, though recent neurophysiologic and electrophysiologic data lend support to the notion that periodicity coding using only unresolved harmonics (i.e., those available via the temporal envelope) is negatively affected as a consequence of age. Individuals with cochlear implants (CIs) must rely on the temporal envelope of speech to glean information about voice pitch [coded through the fundamental frequency (f0)], as spectral f0 cues are not available. While cochlear implants have been shown to be efficacious in older adults, it is hypothesized that they would experience difficulty perceiving spectrally-degraded voice-pitch information. The current experiments were aimed at quantifying the ability of younger and older listeners to utilize spectro-temporal cues to obtain voice pitch information when performing simple and complex auditory tasks. Experiment 1 measured the ability of younger and older NH listeners to perceive a difference in the frequency of amplitude modulated broad-band noise, thereby exploiting only temporal envelope cues to perform the task. Experiment 2 measured age-related differences in f0 difference limens as the degree of spectral degradation was manipulated to approximate CI processing. Results from Experiments 1 and 2 demonstrated that spectro-temporal processing of f0 information in non-speech stimuli is affected in older adults. Experiment 3 showed that age-related performances observed in Experiments 1 and 2 translated to voice gender identification using a natural speech stimulus. Experiment 4 attempted to estimate how younger and older NH listeners are able to utilize differences in voice pitch information in everyday listening environments (i.e., speech in noise) and how such abilities are affected by spectral degradation. Comprehensive results provide further insight on pitch coding in both normal and impaired auditory systems, and demonstrate that spectro-temporal pitch processing is dependent upon the age of the listener. Results could have important implications for elderly cochlear implant recipients.Item Spectrotemporal Modulation Sensitivity in Hearing-Impaired Listeners(2009) Mehraei, Golbarg; Shamma, Shihab; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Speech is characterized by temporal and spectral modulations. Hearing-impaired (HI) listeners may have reduced spectrotemporal modulation (STM) sensitivity, which could affect their speech understanding. This study examined effects of hearing loss and absolute frequency on STM sensitivity and their relationship to speech intelligibility, frequency selectivity and temporal fine-structure (TFS) sensitivity. Sensitivity to STM applied to four-octave or one-octave noise carriers were measured for normal-hearing and HI listeners as a function of spectral modulation, temporal modulation and absolute frequency. Across-frequency variation in STM sensitivity suggests that broadband measurements do not sufficiently characterize performance. Results were simulated with a cortical STM-sensitivity model. No correlation was found between the reduced frequency selectivity required in the model to explain the HI STM data and more direct notched-noise estimates. Correlations between low-frequency and broadband STM performance, speech intelligibility and frequency-modulation sensitivity suggest that speech and STM processing may depend on the ability to use TFS.Item Closed Loop System Identification of Postural Control with Bilateral Vestibular Loss(2009) Amenabar, Katharine Elizabeth; Jeka, John J; Kinesiology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Human upright stance can be characterized as a combination of feedback and plant. Feedback consists of integrated sensory signals, producing estimates of position and velocity of the body segments while plant includes both musculotendon dynamics and body dynamics. Separating plant and feedback is possible mathematically through closed loop system identification. By studying bilateral vestibular loss (BVL) patients it is hoped knowledge regarding human posture and the role of the vestibular system will be gained. Two BVL patients and two age, height and gender matched controls had visual and mechanical perturbations applied simultaneously to determine these properties. Both leg and trunk kinematics and EMG data were collected. Using frequency response functions plant and feedback properties were calculated. Plant and feedback dynamics differ. BVL patients show more variable weighted hip EMG data, supporting the idea that this population can not properly use hip movement with their lack of vestibular system.Item Development of an Evidence Based Referral Protocol for Early Diagnosis of Vestibular Schwannomas(2008-09-03) Barrett, Jessica Ann; Gordon-Salant, Sandra; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The purpose of this investigation was to identify the presenting symptoms and testing outcomes that were most suggestive of a potential vestibular schwannoma and to propose an audiological referral protocol for MRIs. To that end, a retrospective chart review was conducted to examine radiologic, audiometric, and case history information from patients at Walter Reed Army Medical Center who were referred to the Department of Radiology to rule out retrocochlear pathology. Charts of 628 patients were reviewed from their electronic medical records, although the final patient sample was 328 patients who had complete audiologic data. Analyses were conducted to compare the unaffected and affected ears of the positive MRI group to the better and poorer ears of the negative MRI group. Results were significant between the affected ear of the positive group and the poorer ear of the negative group for pure tone thresholds, speech discrimination scores, and acoustic reflex thresholds. Significant differences between the groups were not generally seen for the comparison of the unaffected ear to the better ear, with the exception of acoustic reflex thresholds. The interaural difference between ears was significant between the two groups for pure tone thresholds and speech discrimination scores; however, the difference was not significant for acoustic reflex thresholds. For all significant differences between the groups, the positive MRI group evidenced poorer audiological results. Additionally, three symptoms/outcomes that led to the patients' referral were significantly different between the two groups: unilateral tinnitus, asymmetrical word recognition, and positive rollover in speech recognition scores. Logistic regression was applied to the audiological tests and symptoms to determine the most predictive set of variables that differentiated between the patients with a positive and negative MRI. The most predictive model yielded a sensitivity of 81.25% and a specificity of 82.59% when applied to the current patient sample. The audiological profile identified may be useful for clinicians in deciding whether their patient should be referred for an MRI to rule out the presence of a vestibular schwannoma.Item Predicting the loudness discomfort level from the acoustic reflex threshold and growth function(2008-08-18) Cannavo, Justine Marie; Gordon-Salant, Sandra; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The purpose of the present study was to evaluate the relationship between several measures of the acoustic reflex [acoustic reflex threshold (ART), dynamic range of the acoustic reflex growth function, the 50% point along the acoustic reflex growth function, and the maximum intensity value of the acoustic reflex growth function] and behavioral measurements of loudness [loudness discomfort level (LDL) and the loudness contour (LC)]. The underlying objective was to determine if any of these measures can be used to predict the LDL. A finding of a strong relationship between these measures could potentially assist in the creation of an objective method to measure LDLs, which may have implications for hearing aid fittings. Prior research in this area has yielded conflicting results. However, very few studies examined measures of loudness growth and the dynamic range of the acoustic reflex. Twenty young adults ranging from 22-35 years of age (Mean age = 25.85, s.d. 3.07) with normal hearing participated in this study. Participants were required to provide a subjective loudness rating to warbled-tone stimuli in accordance with a categorical loudness scaling procedure adapted from Cox et al. (1997), as well as an LDL rating. Additionally, an ART was obtained from each participant, as defined by a 0.02 mmho change in admittance. Following identification of the ART, the acoustic reflex growth function was obtained by increasing the stimulus until the termination point. Experimental measures were obtained over two test sessions. Results revealed no significant relationship between measures of the acoustic reflex and loudness. Analysis of test-retest measures revealed moderate to very high positive (0.70 - 0.92) correlations for the acoustic reflex and LDL measures over a period of 1 day to 2 weeks. Test-retest performance on the majority of loudness categories on the LC did not reveal stable results. Implications for these findings are that the ART cannot be used to reliably predict the LDL. Additionally, the LC may not be a reliable clinical measurement to assess loudness.Item The Effect of Body Position on Distortion Product Otoacoustic Emissions Testing in Neonates(2008-08-15) Heinlen, Krista; Fitzgerald, Tracy; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The current study investigated the effects of body position on the measurement of distortion product otoacoustic emissions (DPOAEs) in newborns. DPOAE measurements are commonly used to screen for hearing loss in newborn hearing screening programs conducted in hospitals nationwide. To measure DPOAEs, a small probe is placed in the external ear canal and a series of tone pairs is presented to the ear. The ear's acoustic response to these tones is measured to determine if the infant is at risk for a hearing loss. Research in adults has indicated effects of body position on DPOAE levels and noise floor levels (Driscoll et al., 2004). However, no information is available on the effects of body position on DPOAE testing in infants, despite the fact that newborn screening is one of the primary clinical applications of DPOAEs. Participants were 47 full-term newborns recruited from the well-baby nursery. DPOAEs were measured from the right ear while the infants were in each of three body positions: lying on the left side, supine, and head raised 45 degrees from supine. DPOAE levels, noise floor levels, DPOAE/noise levels, test time, and pass/fail rate were compared across body positions to determine whether there is an optimal body position for newborn hearing screenings that would minimize test time and/or increase specificity. No statistically significant differences were found in the various DPOAE measures or screening results across body positions or between genders. Significant effects of frequency on DPOAE levels and noise floor levels were similar to those expected based on the literature (e.g., Gorga et al., 1993). The results suggest that newborn hearing screenings on infants in the well-baby nursery can be conducted in different body positions without significantly influencing the screening outcome or measurements obtained.Item The benefits of acoustic input to combined electric and contralateral acoustic hearing(2008-08-01) Zhang, Ting; Gordan-Salant, Sandra; Dorman, Michael F.; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)With the extension of cochlear implant candidacy, more and more cochlear-implant listeners fitted with a traditional-long electrode array or a partial-insertion electrode array have residual acoustic hearing either in the nonimplanted ear or in both ears and have shown to receive significant speech-perception benefits from the low-frequency acoustic information provided by residual acoustic hearing. The aim of Experiment 1 was to assess the minimum amount of low-frequency acoustic information that was required to achieve speech-perception benefits both in quiet and in noise from combined electric and contralateral acoustic stimulation (EAS). Speech-recognition performance of consonant-nucleus vowel-consonant (CNC) words in quiet and AzBio sentences in a competing babble noise at +10 dB SNR was evaluated in nine cochlear-implant subjects with residual acoustic hearing in the nonimplanted ear in three listening conditions: acoustic stimulation alone, electric stimulation alone, and combined contralateral EAS. The results showed that adding low-frequency acoustic information to electrically stimulated information led to an overall improvement in speech-recognition performance for both words in quiet and sentences in noise. This improvement was observed even when the acoustic information was limited down to 125 Hz, suggesting that the benefits were primarily due to the voice-pitch information provided by residual acoustic hearing. A further improvement in speech-recognition performance was also observed for sentences in noise, suggesting that part of the improvement in performance was likely due to the improved spectral representation of the first formant. The aims of Experiments 2 and 3 were to investigate the underlying psychophysical mechanisms of the contribution of the acoustic input to electric hearing. Temporal Modulation Transfer Functions (TMTFs) and Spectral Modulation Transfer Functions (SMTFs) were measured in three stimulation conditions: acoustic stimulation alone, electric stimulation alone, and combined contralateral EAS. The results showed that the temporal resolution of acoustic hearing was as good as that of electric hearing and the spectral resolution of acoustic hearing was better than that of electric hearing, suggesting that the speech-perception benefits were attributable to the normal temporal resolution and the better spectral resolution of residual acoustic hearing. The present dissertation research provided important information about the benefits of low-frequency acoustic input added to electric hearing in cochlear-implant listeners with some residual hearing. The overall results reinforced the importance of preserving residual acoustic hearing in cochlear-implant listeners.Item Age-related Effects on the Threshold Equalizing Noise (TEN) Test(2008-05-30) Gmitter, Christine; Gordon-Salant, Sandra; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Some individuals with sensorineural hearing loss have certain places along the basilar membrane where inner hair cells and/or neurons are damaged or destroyed and consequently have ceased to function. These regions have been referred to as "dead regions" in the literature. The TEN (HL) test is a relatively quick behavioral test designed to identify cochlear dead regions. The test relies on the detection of pure-tone signals in the presence of a specially designed broadband noise (threshold equalizing noise) masker. The TEN (HL) test was validated on young to middle aged adult listeners, an age group which does not represent that of all adults with hearing loss. The goal of this study was to evaluate the effects of age on the TEN (HL) test. The TEN (HL) test was administered to 18 younger and 18 older adults with normal to near-normal hearing sensitivity at seven different frequencies in three different levels of TEN noise. These measures were conducted twice to assess test re-test reliability. The older group demonstrated significantly poorer (higher) SNRs compared to the younger group at all three TEN noise levels and for all seven test frequencies. The greatest difference between groups was observed for the highest level of TEN noise. The greatest difference in SNRs was at 4000 Hz compared to other test frequencies for both groups. Both groups performed best (lowest SNRs) at 4000 Hz compared to the other test frequencies. Finally, a main effect of trial was found, revealing that both groups performed statistically better (lower SNRs) on the second trial; however the small magnitude of this improvement (0.37 dB), suggests that the TEN (HL) test has good repeatability for clinical use, at least within the time period assessed. Although there were significant differences between the two groups, overall the TEN (HL) test yielded accurate results in classifying all normal to near-normal hearing participants as not having a dead region. The significantly higher (poorer) SNRs associated with age, combined with the expected difference in SNRs associated with hearing loss, may allow for older hearing-impaired individuals to demonstrate abnormally high SNRs on the TEN (HL) test in the absence of a cochlear dead region. Future studies that include younger and older participants with normal hearing and hearing loss are needed to assess these differences and examine whether different norms are needed for this older population.Item Speech Recognition in Noise and Intonation Recognition in Primary-School-Aged Children, and Preliminary Results in Children with Cochlear Implants(2008-05-13) Wawroski, Lauren Ruth; Chatterjee, Monita; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Fundamental frequency (F0), or voice pitch, is an important acoustic cue for speech intonation and is perceived most accurately through the fine spectral resolution of the normal human auditory system. However, relatively little is known about how young children process F0-based speech intonation cues. The fine spectral resolution required for F0 information has also been shown to be beneficial for listening in noise, a skill that normally-hearing children are required to use on a daily basis. While it is known that hearing-impaired adults with cochlear implants are at a disadvantage for intonation recognition and listening in noise following loss of fine spectral structure cues, relatively little is known about how young children with unilateral cochlear implants perform in these situations. The goal of the current study was to quantify how a group of twenty normally-hearing children (6-8 years of age) perform in a listening-in-noise task and in a speech intonation recognition task. These skills were also measured in a small group of 5 children of similar age with unilateral cochlear implants (all implanted prior to the age of five). The cochlear implant participants in this study presumably had reduced spectral information, and it was hypothesized that this would be manifested as performance differences between groups. In the listening-in-noise task, sentence recognition was measured in the presence of a single-talker masker at different signal-to-noise ratios. Results indicated that the participants with cochlear implants achieved significantly lower scores than the normally-hearing participants. In the intonation recognition task, listeners heard re-synthesized versions of a single bisyllabic word ("popcorn") with systematically varying F0 contours, and indicated whether the speaker was "asking" or "telling" (i.e., question-like or statement-like). Both groups of children were able to use the F0 cue to perform the task, and no significant differences between the groups were observed. Although limited in scope, the results suggest that children who receive their cochlear implant before the age of five have significantly more difficulty understanding speech in noise than their normally-hearing peers. However, the two populations appear to be equally able to use F0 cues to determine speech intonation patterns.Item The effect of the menstrual cycle on evoked otoacoustic efferent suppression(2008-04-28) Mahmood, Sally Louay; Fitzgerald, Tracy; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The purpose of the study was to investigate the effects of the menstrual cycle on suppression of transient-evoked otoacoustic emissions. Otoacoustic emissions (OAEs) are soft sounds produced by the inner ear that can be measured in the ear canal by a sensitive microphone. OAEs may be present spontaneously or may be evoked by presenting sound(s) to the ear. Presenting a noise (in addition to the eliciting stimulus) to one or both ears during testing typically causes a change in the measured OAE levels. Because the change is most often a decrease in OAE levels, this effect has been termed "suppression." Although OAE suppression is not used routinely in audiometric evaluations, research has indicated potential clinical value for diagnosis of certain pathologies, such as auditory neuropathy (e.g., Starr et al., 1996). However, more information on sources of normal variation in OAE suppression is needed. Little information is available on how the menstrual cycle affects OAE suppression. In the present study, suppression of transient-evoked OAEs (TEOAEs) was investigated. TEOAEs are measured following the presentation of clicks to the ear. Repeated measures of TEOAE suppression were completed on 30 participants divided into three groups: (1) 10 normally-menstruating females who were not taking oral contraceptives, (2) 10 normally-menstruating females who were taking oral contraceptives, and (3) 10 males. Participants were tested on three separate days. Female participants were tested during menstruation, pre-ovulation/mid-cycle and pre-menstruation. An ovulation prediction kit was used by female participants not taking oral contraceptives to aid in estimating the time of ovulation. Male participants were tested at intervals that corresponded in time to those for the female groups. TEOAE suppression did not differ significantly between the three groups or across the three sessions (one menstrual cycle) for any of the groups. Unsuppressed TEOAE levels were also similar between groups and stable across sessions for all groups. The findings suggest that female sex hormones do not affect TEOAE suppression. From a clinical standpoint, these results are fortuitous in that phases of the menstrual cycle would not need to be taken into account when interpreting unsuppressed TEOAE levels or TEOAE suppression results.