Thesis (M.Sc.Eng.) PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at open-help@bu.edu. Thank you. / Binaural hearing studies focus on how binaural processing improves the extraction of information from one source in the presence of competing sources. The most extensively studied condition is the detection of an out-of-phase tonal signal in an interaurally identical, Gaussian masking noise, called the N0Spi condition. Recently, attention turned to the dependence of detection performance on individual waveforms in the context of random noise waveforms from trial to trial. This thesis addresses this dependence, as measured in experiments (Isabelle 1991, 1995) that estimated probabilities of detection (Pd) and false alarm (Pf) for each of 30, narrowband-noise waveforms in the N0Spi condition. In previous work, models were shown to describe average performance and much of the variation over Pd, but the variation of Pf across noise samples was not explained. The current study explores two approaches to understanding the variation of Pd and Pf with noise waveform. First, a metric based on Shannon entropy is evaluated with the entropy computed from a combination of Pd and Pf. Second, internal noise in the form of temporal jitter is incorporated into existing interaural differences models. Results show that the correlation of the variation of interaural differences with the entropy is slightly stronger than that correlation with Pd alone. Models based on variations in the interaural differences with temporal jitter included are neither better nor worse than those without temporal jitter. Overall, these results suggest that the variation with Pf as captured by the entropy can be explained by interaural difference models. / 2031-01-01

No description available.

No description available.

The most accepted model of mammalian binaural sound localization postulates two neural/perceptual channels with hemifield tuning and overlapping medial borders; the extent to which the two channels are co-activated by the source is the neural "code" for the source's azimuthal location. This model does not take into account physiological data on the existence of a population of cells with spatial receptive fields centered on the azimuthal midline. The following work sought to test the hypothesis that the mammalian binaural sound localization apparatus includes a third, midline-tuned channel. Ten experiments used a selective adaptation paradigm in human listeners to probe for the existence of a midline channel. Psychometric functions were obtained for lateral position based on ITD or ILD both before and after adaptation with high-frequency (2800 and 4200 Hz) or low-frequency (260 and 570 Hz) tones. Listeners experienced highly lateralized adaptor stimuli with different frequencies at each ear (asymmetrical adaptation), highly lateralized adaptor stimuli of the same frequency at each ear (symmetrical adaptation), and single frequency adaptation at the midline (central adaptation). At both high and low frequencies in the domains of both interaural time difference (ITD) and interaural level difference (ILD), location judgements after asymmetrical adaptation shifted away from the fatigued side. These shifts occurred across each adapted hemifield and extended slightly over the midline, as is consistent with the two-channel model. The two-channel model would predict no effect of symmetrical or central adaptation because fatiguing both lateral channels equally would not change their relative activation by a given source. In practice, the result of symmetrical adaptation was a shift in location judgements towards the midline as would be expected if adaptation of the lateral channels resulted in a greater relative contribution of a third, midline channel. Likewise, central adaptation tended to result in shifts in perceived location towards the sides. The evidence for the midline channel was strong for high and low frequencies localized by ILD, and was present for low frequencies, but not for high frequencies, localized by ITD.

A new objective measurement system to predict speech intelligibility in binaural listening conditions is proposed for use with nonlinear hearing devices. Digital processing inside such devices often involves nonlinear operations such as clipping, compression, and noise reduction algorithms. Standard objective measures such as the Articulation Indeix (AI), the Speech Intelligibility Index (SII) and the Speech Transmission Index (STI) have been developed for monaural listening. Binaural extensions of these measures have been proposed in the literature, essentially consisting of a binaural pre-processing stage followed by monaural intelligibility prediction using the better ear or the binaurally enhanced signal. In this work, a three-stage extension of the binaural SII approach is proposed that deals with nonlinear acoustic input signals. The reference-based model operates as follows: (1) a stage to deal with nonlinear processing based on a signal-separation model to recover estimates of speech, noise and distortion signals at the output of hearing devices; (2) a binaural processing stage using the Equalization-Cancellation (EC) model; and (3) a stage for intelligibility prediction using the SII or the short-time Extended SII (ESII). Multiple versions of the model have been developed and tested for use with hearing devices. A software simulator is used to perform hearing-device processing under various binaural listening conditions. Details of the modeling procedure are discussed along with an experimental framework for collecting subjective intelligibility data. In the absence of hearing-device processing, the model successfully predicts speech intelligibility in all spatial configurations considered. Varying levels of success were obtained using two simple distortion modeling approaches with different distortion mechanisms. Future refinements to the model are proposed based on the results discussed in this work.

The central theme of this thesis is a description of information processing in the sound localization circuit of the auditory pathway. The focus is on principal neurons of the medial superior olive (MSO), the first major convergence point for binaural information. Selected properties and relations of MSO neurons are derived and expressed through models. In the thesis we present three modeling studies. The first one clarifies a relation- ship between biophysical parameters of the MSO neuron and its ability to detect coincidental spikes from the left and the right ear. The second study describes the statistical behavior of spike trains on the input and output of the MSO neuron. In the third work, we studied how interaural coherence could guide localization of sound sources in complex listening situations with multiple sound sources in reverberant environments. The main results are analytical and numerical models describing the aforemen- tioned relations and behaviors. Secondary results include that inhibitory input to the MSO neuron narrows and shifts the time range of coincidence detection, that ergodic assumption from statistical physics and circular statistics are beneficial in the description of spike trains in the auditory pathway, and that interaural level difference of parts of the signal with...

The purpose of the study was to determine whether hearing aids with binaural processing improve performance during a localization and a hearing in noise task. The study included 16 participants, ages 29 – 67, with bilateral, essentially symmetrical, sensorineural hearing loss who had no prior hearing aid experience. Participants were fit with Oticon Epoq XW hearing aids bilaterally and completed the localization and the hearing in noise task with three listening conditions: (1) without hearing aids (NO), (2) with hearing aids that were not linked (BIL), and (3) with hearing aids that were linked (BIN). For the localization task, 1.5 second pink noise bursts at 75 dB SPL were used as the stimulus. A 180° 11-speaker array was set up to the right or left side of the participants. A twelfth speaker on the contralateral side of the array introduced constant background pink noise at 65 dB SPL. Results revealed that participants performed the best with the NO condition, followed by BIL, then BIN. There was a significant difference between NO and BIL and NO and BIN. For the hearing in noise (HIN) task, sentences from the Hearing in Noise Test (HINT) were used as target stimuli. Continuous discourse by one male and two female talkers were used as maskers. There were four masker conditions for this task: (1) signal at 0°, masker at 90° (S0-N90), (2) signal at 0°, masker at 180° (S0-N180), (3) signal at 0°, masker at 270° (S0-N270), and (4) signal at 0°, maskers at 90°, 180°, and 270° (S0-N90, 180, 270). Results revealed that there was no significant difference between listening conditions when all masker conditions were considered. When the one-masker conditions were included, there was a significant difference between the NO and BIL and the NO and BIN conditions with the best performance for BIL, followed by BIN, then NO. Results also revealed a significant difference between masker conditions with the best performance for S0-N270, next best for S0-N90, followed by S0-N180, then S0-N90, 180, 270. / text

Children with central auditory processing disorders (CAPD) have a normal pure-tone audiogram, however, they have difficulty understanding speech in the presence of background noise. The present study examined binaural hearing in normal children and those with possible CAPD. Each subject was administered the SCAN or SCAN-A, screening tests for CAPD, to determine whether they were at risk for CAPD. Participants were then subjected to several monaural and binaural speech tasks, in quiet and noise. Spondee words were utilized in each task, under headphone and soundfield conditions.

Three experiments investigated the ability and means by which Florida manatees determine sound source directionality. An eight-choice discrimination paradigm determined the sound localization abilities of two manatees within a 360° array of speakers. Five conditions were tested including a 3,000 and 200 ms, 95 dB, 0.2-24 kHz signal, a 3,000 ms, 80 dB, 18-24 kHz signal, a 3000 ms, 110 dB, 0.2-1.5 kHz signal and a 200 ms, 101 dB, 4 kHz tonal signal. A sixth condition attenuated the level of the 3,000 ms, 95 dB, 0.2-24 kHz signal in 3 dB increments until accuracy reached 75%. Subjects performed above the 12.5% chance level for all broadband frequencies and were able to localize over a large level range. Errors were typically located to either side of the signal source location when presented in the front 180° but were more dispersed when presented from the 135°, 180° and 225° locations. Front-to-back confusions were few and accuracy was greater when signals originated from the front 180°. Head/body related transfer functions determined how different frequencies were filtered by the manatees' head/torso to create frequency-specific interaural level differences (ILDs). Hydrophones were suspended next to each manatee ear and Fast Fourier transform (FFT) ratios compared received signals with and without the subject's presence. ILD magnitudes were derived for all frequencies, as well as specific 0.2-1.5, 0.2-5, and 18-30 kHz bands of frequencies. ILDs were found for all frequencies as a function of source location, although they were largest with frequencies above 18 kHz and when signals originated at 90° and 270°. Larger ILDs were found when the signals originated behind the subjects as compared to in front of them. Auditory evoked potential (AEP) techniques were used to map manatee sound conduction pathways in-water and in-air using 15 and 24 kHz carriers. All subjects produced AEPs at each position the transducer was placed, however specific sound conduction pathway(s) were not identified. AEP amplitudes were usually greater with the 24 kHz carrier, however patterns between carriers at identical body positions were highly variable between subjects.

abstract: Two groups of cochlear implant (CI) listeners were tested for sound source localization and for speech recognition in complex listening environments. One group (n=11) wore bilateral CIs and, potentially, had access to interaural level difference (ILD) cues, but not interaural timing difference (ITD) cues. The second group (n=12) wore a single CI and had low-frequency, acoustic hearing in both the ear contralateral to the CI and in the implanted ear. These `hearing preservation' listeners, potentially, had access to ITD cues but not to ILD cues. At issue in this dissertation was the value of the two types of information about sound sources, ITDs and ILDs, for localization and for speech perception when speech and noise sources were separated in space. For Experiment 1, normal hearing (NH) listeners and the two groups of CI listeners were tested for sound source localization using a 13 loudspeaker array. For the NH listeners, the mean RMS error for localization was 7 degrees, for the bilateral CI listeners, 20 degrees, and for the hearing preservation listeners, 23 degrees. The scores for the two CI groups did not differ significantly. Thus, both CI groups showed equivalent, but poorer than normal, localization. This outcome using the filtered noise bands for the normal hearing listeners, suggests ILD and ITD cues can support equivalent levels of localization. For Experiment 2, the two groups of CI listeners were tested for speech recognition in noise when the noise sources and targets were spatially separated in a simulated `restaurant' environment and in two versions of a `cocktail party' environment. At issue was whether either CI group would show benefits from binaural hearing, i.e., better performance when the noise and targets were separated in space. Neither of the CI groups showed spatial release from masking. However, both groups showed a significant binaural advantage (a combination of squelch and summation), which also maintained separation of the target and noise, indicating the presence of some binaural processing or `unmasking' of speech in noise. Finally, localization ability in Experiment 1 was not correlated with binaural advantage in Experiment 2. / Dissertation/Thesis / Ph.D. Speech and Hearing Science 2013