• Auditory Evoked Potentials
• Behavioral Studies
• Cochlear Mechanics
• Electrophysiology
• Impaired Auditory Function
• Mathematical Modeling
• Multi-unit Recordings
• Neuroanatomy
• Otoacoustic Emissions
• Psychophysics

• Auditory Neurophysiology Laboratory - Biomedical Engineering, Charles River Campus
• Biomimetic Systems Laboratory - Biomedical Engineering, Charles River Campus
• Binaural Hearing Laboratory - Biomedical Engineering, Charles River Campus
• Biological Information Transmission Laboratory - Electrical, Computer and Systems Engineering/Biomedical Engineering, Charles River Campus
• CNS Psychoacoustics Laboratory - Biomedical Engineering, Charles River Campus
• Cochlear Biophysics Laboratory - Biomedical Engineering, Charles River Campus
• Knowledge-Based Signal Processing Group - Electrical and Computer Engineering, Charles River Campus
• Laboratory of Cellular and Molecular Hearing Research - Anatomy and Neurobiology, Medical Campus
• Psychoacoustics Laboratory - Communication Disorders, Charles River Campus
• Speech Perception Laboratory - Communication Disorders, Charles River Campus
• Vestibular Laboratory - Otolaryngology, Medical Campus

• H. Steven Colburn
• Modeling the Activity of Inferior Colliculus Neurons (H. Steven Colburn)
• Binaural Signal Processing in Complex Environments (H. Steven Colburn)
• Virtual Acoustic Environments (H. Steven Colburn)
• Models of Binaural Performance Based on Physiological Results (H. Steven Colburn)
• Douglas A. Cotanche
• Cell Form and Gene Expression in Hair Cell Regeneration (Douglas A. Cotanche)
• Regulation and Expression of Genes Involved in Hair Cell Regeneration (Douglas A. Cotanche)
• William F. Dolphin
• Measures of Auditory Function In Stranded Marine Mammals (William F. Dolphin)
• Spectral and Temporal Interactions within the cochlear Nuclei (William F. Dolphin)
• Development, Testing, and Implementation of Techniques for Optimizing the Acquisition of Auditory Evoked Potentials In Marine Mammals (William F. Dolphin)
• Gerald Kidd
• Peripheral and Central Processes in Auditory Masking (Gerald Kidd)
• Assessing informational masking in listeners with sensorineural hearing loss (Gerald Kidd)
• Melanie Mathies
• Constraints and Strategies in Speech Production (Melanie Matthies)
• David Mountain and Allyn Hubbard
• Active Filtering in the Cochlea (David Mountain and Allyn Hubbard)
• Forward and Reverse Traveling Waves in the Cochlea (David Mountain and Allyn Hubbard)
• Hardware and Software Model of Dolphin Auditory Processing (David Mountain and Allyn Hubbard)
• David Mountain
• Neural Circuits for Pitch Estimation (David Mountain)
• Software and Hardware Models of Event-Based Auditory Processing for Real World Application (Boston University component) (David Mountain)
• Barbara Shinn-Cunningham
• Cognitive factors in auditory localization (Barbara Shinn-Cunningham)
• Adaptation to remappings of auditory localization cues (Barbara Shinn-Cunningham)
• Herbert Voigt
• Structure/Function Relations of Cochlear Nucleus Neurons (Herbert Voigt)

RESEARCH AREAS

AUDITORY EVOKED POTENTIALS

William F. Dolphin - Biomedical Engineering

L. Clarke Cox - Otolaryngology

Herbert F. Voigt - Biomedical Engineering & Otolaryngology

BEHAVIORAL STUDIES

William F. Dolphin - Biomedical Engineering

COCHLEAR MECHANICS

Douglas A. Cotanche - Anatomy and Neurobiology

William F. Dolphin - Biomedical Engineering

Allyn E. Hubbard - Electrical and Computer Engineering/Biomedical Engineering

David C. Mountain - Biomedical Engineering

ELECTROPHYSIOLOGY

William F. Dolphin - Biomedical Engineering

J. Michael Harrison - Psychology

L. Clarke Cox - Otolaryngology

Allyn E. Hubbard - Electrical and Computer Engineering/Biomedical Engineering

David C. Mountain - Biomedical Engineering

John R. Stram - Otolaryngology

Herbert F. Voigt - Biomedical Engineering & Otolaryngology

IMPAIRED AUDITORY FUNCTION

H. Steven Colburn - Biomedical Engineering

Douglas A. Cotanche - Anatomy and Neurobiology

L. Clarke Cox - Otolaryngology

Gerald Kidd, Jr. - Communication Disorders

Melanie L. Matthies - Communication Disorders

David C. Mountain - Biomedical Engineering

MATHEMATICAL MODELING

H. Steven Colburn - Biomedical Engineering

William S. Hellman - Physics

Allyn E. Hubbard - Electrical and Computer Engineering/Biomedical Engineering

David C. Mountain - Biomedical Engineering

S. Hamid Nawab - Electricaland Computer Engineering/Biomedical Engineering

Malvin C. Teich - Electrical and Computer Engineering/Biomedical Engineering

Herbert F. Voigt - Biomedical Engineering, Otolaryngology

MULTI-UNIT RECORDINGS

Herbert F. Voigt - Biomedical Engineering, Otolaryngology

NEUROANATOMY

Douglas A. Cotanche - Anatomy and Neurobiology

OTOACOUSTIC EMISSIONS

L. Clarke Cox - Otolaryngology

Allyn E. Hubbard - Electrical and Computer Engineering/Biomedical Engineering

David C. Mountain - Biomedical Engineering

PSYCHOPHYSICS

H. Steven Colburn - Biomedical Engineering

John F. Culling - Experimental Psychology

J. Michael Harrison - Psychology

Gerald Kidd, Jr. - Communication Disorders

Melanie L. Matthies - Communication Disorders

Virginia Richards - Psychology

Malvin C. Teich - Electrical and Computer Engineering/Biomedical Engineering

LABORATORIES

Auditory Neurophysiology Laboratory - Herbert F. Voigt

Biomedical Engineering Dept. 44 Cummington Street Boston, MA 02215

The primary focus of this group is on understanding the neuronal circuitry of the cochlear nucleus, particularly the dorsal cochlear nucleus (DCN). There are two experimental thrusts - 1) studies of functional interactions of neurons by observing correlated activities of simultaneous-recorded pairs of units in both anesthetized and unanesthetized preparations, and 2) extracellular and intracellular recording and marking studies of neurons in cochlear nucleus. This experimental work provides the "reality checks" for the computational modeling effort within the group. They have been modeling an isofrequency patch of DCN and are beginning to explore cross-frequency interaction issues.

This laboratory contains an IAC sound-attenuation chamber that is fully instrumented to conduct both intracellular electrophysiological recording experiments of the auditory system in experimental animal. Experiments involving single and pair recordings are feasible under computer control. Dr. Voigt also has a fully functional histology laboratory for standard tissue staining and HRP histochemistry. An electrode room and a shared computer modeling and simulation laboratory are also available.

Biomimetic Systems Laboratory - Allyn E. Hubbard and David C. Mountain

Biomedical Engineering 44 Cummington Street Boston, MA 02215

The goal of this laboratory is to develop large-scale biophysically-based models of the auditory periphery and subcortical auditory pathways. The purpose of these models is to aid the interpretation of and the design of physiological and psychophysical experiments as well as to study auditory models for their usefulness as preprocessors for automated recognition of acoustic signals. This laboratory is also engaged in the study of natural acoustic signal sources and acoustic environments. The purpose of this second effort is to develop a better understanding of the evolutionary pressures which have shaped the auditory pathway as well as to develop computer simulations of natural environments for use as input to the auditory models. Other current projects include the use of auditory models for the acoustic transients and development of models for processing temporal sequences.

Click here to link to the Biomemetic Systems Laboratory page

Binaural Hearing Laboratory - H. Steven Colburn

Biomedical Engineering 44 Cummington Street Boston, MA 02215

The Binaural Hearing Laboratory is focused on studies of binaural interaction, including phenomena such as sound localization for which monaural processing also plays a major role. The goal of these studies is an integrated understanding of binaural interaction and its role in human sound perception including the interpretation of acoustic cues in complex sound environments (e.g., multiple sources in reverberant spaces). Specific projects range from signal processing models of physiological activity to empirical measurements of the hearing abilities of listeners with hearing losses and/or neurological lesions. In the neural modeling area, we are evaluating the abilities of simple neural models to generate firing patterns equivalent to those seen in binaural cells in brainstem nuclei such as the MSO, LSO, and IC. In psychophysical studies of normal listeners, current interests include interaural discrimination and binaural detection, especially detection with reproducible noise maskers. In studies of listeners with hearing impairments, we are trying to relate listeners' abilities on a variety of binaural tests to a primary set of psychophysical measures. In studies of sound localization and recognition, we are studying and simulating the cues that lead to externalization, localization, and separation of sources.

Click here for more information about the Binaural Hearing Laboratory

Biological Information Transmission Laboratory - Malvin C. Teich

Electrical and Computer Engineering/Biomedical Engineering 44 Cummington Street Boston, MA 02215

The Biological Information Transmission Laboratory is devoted to the computer analysis of sensory data, and to carrying out simulations and analytical calculations for biological models using engineering principles. Information processing mechanisms involving nonlinear dynamical models, and fractal characteristics, are of particular interest to members of the Laboratory. Special software has been developed to facilitate the dynamical display of theoretical results and processed data.

Click here for more information about research activities in the Biological Information Transmission Laboratory

CNS Psychoacoustics Laboratory - Barbara G. Shinn-Cunningham

Cognitive and Neural Systems 677 Beacon Street Boston, MA 02215

Research in the Psychoacoustics Laboratory focuses on studies of spatial hearing, learning and plasticity in spatial perception, and quantitative analysis and modeling of human perception and performance. One current project is focused on sound localization when sources are very close to the listener by measuring and analyzing the physical cues that occur for near field sound sources, and analysis of the effects of room acoustics on the acoustic signals reaching a listener, perception, and performance, for near-field sounds. In other work, we are investigating the relationship between different measures of human spatial auditory perception in order to develop quantitative models of performance. We also are investigating the degree to which spatial perception is affected by learning and training with unusual spatial cues. All of this work is aimed at the development of computational models capable of predicting how the physical cues available to a listener are integrated to form spatial percepts, and how these percepts govern responses on specific tasks.

The laboratory is equipped with special-purpose hardware for the generation and presentation of acoustic stimuli, personal computers, tracking devices to monitor subject positions, and a sound proof booth.

Click here for CNS Psychoacoustics Laboratory site

Cochlear Biophysics Laboratory - Allyn E. Hubbard, David C. Mountain, and Lisa Shatz Biomedical Engineering 44 Cummington Street Boston, MA 02215

This laboratory is focusing on two areas. The first is to identify, quantify, and model the mechanisms responsible for the mechanical sensitivity and frequency selectivity of the mammalian cochlea. The goal is to bridge the gap between those investigators who are principally interested in haircell biophysics and other investigators who are primarily interested in hearing mechanisms. The experimental approaches currently in use range from the direct measurement of acoustically and electrically-evoked basilar membrane motion and the direct measurement of the mechanical properties of the organ of Corti to the measurement of acoustically and electrically-evoked otacoustic emissions.

The second focus is on the measurement and interpretation of electric fields produced by electrically excitable tissues with the major focus being the auditory system. The goal is to develop methods of studying populations of cells both in vivo and in vitro. The measurement techniques under study range from conventional scalp-recorded evoked potentials to the use of electrode arrays fabricated using integrated-circuit technology. One current project involves testing the ability of planar electrode arrays to image current sources in vitro and in vivo. Another project focuses on the use of the envelope-following response for assessment of low-frequency hearing in marine mammals. Additional projects involve the use of single and multi-unit recording techniques to study processing of complex signals by the auditory brain-stem.

Laboratory facilities include a vibration-isolated sound attenuating booth with associated computer controlled instrumentation and two setups for video microscopy with associated computer equipment for stimulus control, data collection and image processing. Special purpose instrumentation includes a fiber-optic displacement probe capable of resolving motion at the angstrom level and a force-probe system capable of measuring forces at the nanonewton level.

Click here for the Cochlear Biophysics Laboratory page

Knowledge-Based Signal Processing Group - S. Hamid Nawab

Electrical and Computer Engineering 44 Cummington Street Boston, MA 02215

This group is concerned with the integration of signal processing and signal understanding technologies and their application to robotic hearing and assisting devices for the hearing impaired. This research utilizes electrical engineering principles for signal processing and computer science principles for signal understanding.

Laboratory facilities include workstations, A/D and D/A conversion, and special-purpose DSP hardware and software.

Laboratory of Cellular and Molecular Hearing Research - Douglas A. Cotanche

BUMC Anatomy and Neurobiology 80 E. Concord Street Boston, MA 02215

This laboratory is working on developmental factors which regulate the growth of hair cells in the cochlea during normal embryonic development and in regeneration following noise-induced trauma. They are primarily concerned with the structural development of the stereocilia: specialized microvilli which project from the apical surface of the hair cells. They have been studying the embryonic development of stereocilia and have searched for factors which regulate the establishment of morphological gradients in the sterociliary bundles. Also, they have examined structural changes in hair cells caused by acoustic overstimulation and have shown that the bird ear is able to completely regenerate the sensory epithelium during recovery from noise damage. Of interest to them is the fact that the regenerating hair cells go through a developmental sequence which is identical to that exhibited by hair cells in the embryonic cochlea. Thus, the regulatory mechanisms which are employed during embryogenesis are probably activated once again for hair cell regeneration. Recently, they have been examining events associated with regulating cell cycle events in the precursor cells during regeneration. This includes the expression of early response genes during the transition from G0 to G1 and timing of DNA synthesis during S phase of the cell cycle. The techniques utilized for these studies include scanning and transmission electron microscopy, confocal laser scanning microscopy, reverse transcription PCR amplification, fluorescence in situ DNA hybridization, and video- enhanced differential-interference-contrast (DIC) light microscopy.

The facilities in this laboratory include: video-enhanced DIC light microscopy, epifluorescence light microscopy, laser scanning microscopy, molecular biology facilities for RT-PCR, gene sequencing, in situ hybridization, photographic darkroom for color and black and white slide and paper processing and printing, dedicated computer digital darkroom, and image processing work station.

Click here to for site

Psychoacoustics Laboratory - Gerald Kidd, Jr.

Communication Disorders 635 Commonwealth Avenue Boston, MA 02215

The type of work done in this laboratory is human auditory perception, both experimental and theoretical. The empirical work consists of studies of auditory perception of listeners with normal hearing and listeners with sensorineural hearing loss. The theoretical work consists of evaluating decision theory based models in accounting for the empirical findings.

The Psychoacoustics Laboratory covers approximately 850 square feet. In addition, it has a shared 350 square foot shop/instrumentation room immediately adjacent. There are two sets of 3-room sound attenuating booths located in the lab. Associated with the booths are several microcomputers and racks of audiofrequency electronic equipment. The computers and electronic equipment are used to generate, control and measure stimuli used in psychophysical experiments and are wired to communicate with subject terminals located inside the booths. There is a local network for sharing peripheral devices for communication among laboratory computers, and for communication with the campus network. Also available within the department are a variety of audiometric instruments, including audiometers, evoked-potential units, and otoacoustics immitance and emission measurement devices.

Speech Perception Laboratory - Melanie L. Matthies

Communication Disorders 635 Commonwealth Avenue Boston, MA 02215

This laboratory is concerned with testing the perception of a wide range of speech stimuli. Work includes quantifying the changes of the speech of persons who have lost their hearing through Neurofibromatosis II. The work in this lab also includes analyzing the speech of subjects who were previously deafened and have received cochlear implants which allows them to improve their speech. Basic properties of speech produced by normal-hearing speakers under a variety of rate and clarity conditions is being studied.

This laboratory covers roughly 400 square feet and includes a double-walled sound booth (10' x 10) for speech perception studies. The laboratory contains the necessary instrumentation to generate and present speech stimuli, collect subject responses and analyze the data. Specific hardware includes a microcomputer, insert earphones, speakers, D/A, A/D, programmable attenuators and appropriate software.

Vestibular Laboratory - L. Clarke Cox

BUMC Otolaryngology 720 Harrison Avenue Boston, MA 02215

top of page

(H. Steven Colburn)

The overall goal of the following four inter-related projects is an integrated understanding of the mechanisms and function of binaural processing, incorporating both physiological and psychophysical results and both normal and impaired auditory systems. The distinctive feature of this package is the interplay between psychophysical measurements, psychophysical modeling, and physiological modeling. The physiological modeling is focused on the central nucleus of the inferior colliculus (IC), which is the binaural nucleus in which the majority of single-unit electrophysiological measurements have been taken and which is believed to be an obligatory nucleus in the ascending auditory pathway.

(H. Steven Colburn) We are developing two types of computational models of the activity of neurons in the central nucleus of the inferior colliculus. We refer to them as the Hodgkin-Huxley-Eccles (HHE) model and the signal-processing model. The HHE model is a refinement of the model that we developed during the last few years (Cai et al., 1997a,b). This model is exceptional in that it has been applied to a wide variety of IC experimental data for low frequencies and it is compatible with essentially all of these data. In this model, individual model neurons are described in terms of channel conductances and equilibrium potentials, and the input patterns are derived from models of more peripheral neurons. We will continue to develop this model by considering in detail the compatibility with new sets of observations that are becoming available from neurons in the IC and other brainstem nuclei, as well as with other types of data (e.g., in vitro recordings from slice preparations, and results from lesion studies). The signal-processing model will be a simplified version of the HHE model and described in signal processing terms. This signal-processing model, although abstracted from physiological mechanisms, will allow easier exploration of how the properties of the output patterns relate to the basic assumptions of the model. The stimulus waveforms from the psychophysical measurements will be used as stimuli for these models in order to understand how these stimuli are represented at the IC level.

(H. Steven Colburn) We are measuring psychophysical performance by normal-hearing and hearing-impaired listeners in several types of tasks in complex acoustical environments. Acoustical environments may be considered complex for several reasons, such as reverberation, echoes, multiple sources in different locations, and multiple frequency bands with high uncertainty. We believe that performance in complex environments is particularly important for listeners, particularly those with hearing impairments, and there have been few careful experiments in this area. Our proposed studies include four different types of experiments: non-speech identification (informational masking or central masking), non-speech discrimination, speech intelligibility, and source localization. All of these experiments are being done in the presence of stimuli at other locations and also in reverberant spaces. Initially, all experiments will be done without head motion. Although we believe that head motion is very important and should not be ignored, we start with the more controlled situation. Head motion will be considered explicitly later.

(H. Steven Colburn) We are developing simulation capabilities for reverberant acoustic spaces. In terms of scientific understanding, the simulations allow tests of hypotheses about what information is perceptually important for the performance measured in these environments. In terms of applications, our ultimate goal is to develop experimental results and simulation techniques that will lead to clinically useful tests and a relatively convenient simulation facility that could be used to test functional abilities of listeners wearing hearing aids in complex acoustical environments. The inclusion of listeners wearing hearing aids implies that the facility must include not only simulation capabilities using headphones but also the capability of controllable simulations using loudspeakers in a small but acoustically controllable environment.

(H. Steven Colburn) We are developing and testing a new psychophysical model that is based on the neural firing patterns in the IC. Among other things, the model will explore the consequences of post-onset suppression effects and dynamic phase effects which are observed in the IC but not at the olivary level. These effects are incorporated in the neural modeling, of course, but they must be formulated in simpler, signal-processing terms in order to be incorporated into a model for psychophysical performance. Further, in order to be able to apply this model to wideband stimuli, high frequencies are being included by describing responses to envelopes of filtered input waveforms for high frequency channels. The simplest psychophysical model assumes that the IC firing patterns are statistical inputs to an optimum processor. This model allows explicit quantitative predictions for essentially any objective psychophysical experiment. The model will be modified to include processors that would predict subjective variables such as lateral position. We plan to apply this model to a wide variety of tasks, including some that we expect to be limited by peripheral (up to and including the IC) processing. For these tasks, we expect the model to predict performance. For other tasks, particularly tasks with higher uncertainty (e.g., Kidd et al., 1994, 1995), we expect that the model will predict performance that is too good. The model should thus be capable of separating ``peripherally limited'' from ``centrally limited'' performance. We propose to apply the model to interaural time discrimination for single stimulus samples and for binaural transient pairs as are typically used in ``precedence effect'' experiments, and to the detection of reproducible noise tokens. In these experiments, except for simple time discrimination experiments, there is no successful model for available results and our proposed model could provide an physiologically based explanation for these results.

(Douglas A. Cotanche)

The long term objective of our research is to understand the mechanisms that induce hair cell loss and regulate regeneration in the chick cochlea. In the current grant we are addressing three major questions concerning hair cell regeneration in the chick cochlea following acoustic overstimulation and amino glycoside administration. The first question uses a combination of cell and molecular biological approaches to identify the factors that regulate proliferation of precursor cells and induce differentiation of the new hair cells. The second question addresses the interactions between the cytoskeleton of hair cells, supporting cells, and hyaline cells and the specific extracellular matrices that are associated with each cell type in normal and regenerating cochleae. The third question determines the nerve fiber distribution and innervation patterns of afferent and efferent fibers in the basilar papilla and hyaline cell region and assess the effects of noise exposure and amino glycoside treatment on these patterns. These three questions are being examined utilizing cell and molecular biology, electron microscopy, video-enhanced DIC light microscopy, immunofluorescence light microscopy, and confocal laser scanning microscopy. The information gained from these studies should provide us with a better understanding of how these events can influence the structural and functional recovery of cochlea.

(Douglas A. Cotanche)

The studies involved in this project are intended to identify and characterize the mechanisms which influence regeneration of damaged inner ear sensory epithelia in chickens.
Specific Aim 1: To determine in undamaged chick cochleae the expression and distribution of ECM components known to regulate cell behavior in other tissues.
Specific Aim 2: To develop protocols for investigating gene expression in the cochlea.
Specific Aim 3: To compare the distribution of the above ECM constituents in undamaged cochleas to that in cochleae damaged with amino glycoside antibiotics.
Specific Aim 4: To localize the sites of production of genes of interest, and identify the cell types responsible.

Development, Testing, and Implementation of Techniques for Optimizing the Acquisition of Auditory Evoked Potentials In Marine Mammals

(William F. Dolphin)

This project includes:
Testing of methods to improve estimation of the evoked response in non-stationary background noise by maximizing the signal-to-noise ratio (SNR). One method which is tested is based on a statistical approach called Bayesian inference, in which subaverages of the total signal acquired at the scalp surface are weighted inversely proportional to the level of noise activity during this acquisition period.

Testing of techniques for the rapid and accurate extraction of the evoked response from background noise, exploiting the relationship between the real and imaginary parts of Fourier components at specified frequencies. Results obtained using this technique are compared with the effectiveness of previously proposed techniques.

Several methods to minimize test time are being explored.

These methods and techniques are tested and compared using computer simulations and controlled experiments in the laboratory using gerbils. The most effective methods will be implemented for use in "field" conditions with marine mammals.

(William F. Dolphin)

The long-range goal of this project is to identify the mechanisms used by the auditory system to process complex sounds in the presence of background competing signals. The auditory system is constantly faced with the situation in which two or more competing sound sources are simultaneously active. A fundamental task and ability of the auditory system is to separate these sound sources from the many that exist in a complex acoustic environment. While this is generally an easy task for a normal hearing individual, in many situations this is extremely difficult for a hearing impaired listener. A more complete understanding of how the auditory system encodes complex sounds and extracts acoustic signals from noise is a necessary prerequisite for the development of new and better strategies for sound processing to aid the hearing impaired. The initial processing of a complex stimulus by the auditory system is to split the signal into its component frequencies via spectral analysis performed by the cochlea. In the presence of multiple sound sources, with each source consisting of multiple components, the sources overlap to different degrees along spectral and temporal dimensions. Therefore, this spectral processing results in the appearance of different frequency components from the same source in different neural channels, and similar frequency components from different sources in the same neural channels. The auditory system must decide which components belong to which source and must emphasize certain components at the expense of others. The overlap of signal components in time and frequency demands that the various components interact at the level of the discharge patterns of single neurons within the auditory system. The interaction of signal components results in variously weighted mixes of adaptation, excitation, inhibition, facilitation, and suppression in the responses of neural elements and thereby determines the perception of the acoustic event. The primary goal of this research is to examine the time course and spectro-temporal contexts of such interactions using extracellular recordings from single units within the dorsal and posterior-ventral cochlear nucleus.

(William F. Dolphin)

(Gerald Kidd)

This study is concerned with the processes that cause one sound to interfere with the reception of another sound, called masking. The work has theoretical significance in aiming to better understand normal auditory processes, and has significance for the study and remediation of hearing impairment by aiming to better understand the causes of the communication difficulties listeners with sensory hearing loss experience in noisy environments. Our work is based on a theory which identifies two types of masking that occur at different physiological levels. Peripheral or "energetic" masking occurs because of overlapping patterns of excitation in the cochlea and auditory nerve and has been studied in detail for many years. Central or "informational" masking involves cognitive processes related to the perceptual organization of sound images and the analysis of sound patterns. Informational masking is not well-understood, and occurs despite a robust representation of the "signal" in the auditory nerve. The purpose of the planned work is to learn more about informational masking: the conditions that cause it and the cues and strategies listeners use to overcome it. This will be accomplished through an empirical study of informational masking involving a series of psychophysical detection, discrimination and identification experiments. The methods of study are relatively new and are motivated by the desire to study processes at work in natural, nonlaboratory listening environments, but also to retain rigorous experimental design and stimulus control. This work is intended to examine the hypothesis that listeners with sensory hearing loss - particularly those experiencing great difficulty hearing in noise - experience large amounts of informational masking and/or make poor use of the cues that normally reduce informational masking. This hypothesis will be evaluated through a series of psychophysical experiments employing listeners with cochlear hearing loss. The experiments include pattern identification and discrimination tasks performed in masked conditions. It is believed that measures of informational masking and release from informational masking may ultimately prove useful clinically in the assessment of communication difficulty and the benefit provided by rehabilitative procedures such as hearing aid fittings.

(Gerald Kidd)

This study uses a new kind of hearing test to measure the difficulty hearing-impaired listeners experience in noise. The long-term goal is to design improved clinical procedures for assessing hearing impairment and determining effective rehabilitative strategies, especially those related to the selection of hearing aids. The purpose of this pilot study is to obtain data from hearing-impaired listeners using these new procedures for comparison with existing normative data, and to assist in determining the feasibility of undertaking a more extensive project on this topic.

(Melanie Matthies) The purpose of this project is to characterize how speech motor control strategies are organized given the constraints of the biomechanical properties of the articulators, the anatomy of the vocal tract and the requirements for acceptable message intelligibility. A very important issue that has not been previously studied is contact-related saturation effects. Available models and analysis techniques cannot account quantitatively for tissue contact influences but these are clearly influencing articulator trajectories. These issues will be investigated with a combination of experimentation and a control-model-driven physiological/biomechanical model. The articulatory data will be gathered at MIT using an Electro-Magnetic Midsagittal Articulometer (EMMA). Signal processing and algorithmic data extraction will be completed with Matlab software. Speech production of experimental subjects will be investigated with extensive acoustic analyses and perceptual judgments by an independent group of listeners.

Aim 1.Develop improved models of cochlear mechanics.

Aim 2.Study the impedances against which outer hair cell (OHC) forces must act by measuring the radial and longitudinal stiffness gradients of the basilar membrane (BM) and the reticular lamina (RL).

Aim 3.Study organ of Corti vibration modes by measuring electrically-evoked and acoustically-evoked motion as a function of radial position.

Aim 4.Study the effect of efferent stimulation on cochlear mechanics by measuring electrically-evoked emissions and electrically-evoked and acoustically-evoked BM motion.

(David Mountain) and (Allyn Hubbard) Considerable evidence has accumulated over the last 15 years that outer hair cells in the mammalian cochlea act as electromechanical elements which significantly increase cochlear sensitivity. As a byproduct of their normal function, outer hair cells produce acoustic energy which can be measured in the external ear canal (otoacoustic emissions). The otoacoustic emissions have provided scientists and clinicians with a unique noninvasive tool to study cochlear function. In spite of hundreds of studies on otoacoustic emissions, the details of their production and their propagation back to the ear canal are not well understood. Our experiments build on our extensive experience with otoacoustic emissions, cochlear electrophysiology, and computer simulation which will address the following 4 hypotheses:

1.Distortion-product emissions are generated in a region near the peak of the cochlear excitation pattern for the higher frequency primary.

2.Distortion-product generation is proportional to the overlap in the F1 and F2 excitation pattern but propagation back to the stapes acts as a low-pass filter due to phase cancellation from distributed generators.

3.The major source of nonlinearity underlying distortion-product emission generation is saturation of the outer hair cell receptor current.

4.Distortion products propagate back to the stapes via a traveling-wave mode which differs from the mode excited by acoustic stimulation.

The results from these experiments are being used to test current cochlear models and to refine our own traveling-wave amplifier model.

(David Mountain) Many natural acoustic sources produce signals which have similar spectra but differ in temporal properties. In addition, the separation of a signal from a single acoustic source from the background is very difficult in complex acoustic environments because the signals from other sources overlap the desired signal in time and frequency. Spectral components from a single source, however, are very likely to co-vary in time. A preprocessing scheme which can effectively enhance the representation of temporal information would be of use both for source separation and source classification. This project develops and tests models of neural circuits which enhance the temporal information. These circuits receive as input the output of onset cells from the auditory model. The circuits that are tested compute the equivalent of an interval histogram and are based on the echo-delay tuned neurons in the bat. These circuits are also closely related to the autocorrelation models which have been proposed to explain human pitch perception.

(David Mountain) We have established a multi-university, multi-disciplinary center of excellence in auditory and acoustics research, joining together the strong talent, track-record, and excellent research and educational infrastructure on the campuses of the University of Maryland, Boston University, the University of Washington, ETH/UZ in Zurich, and the University of California at Berkeley, in partnerships with collaborating teams in several DoD labs (Bremerton Detachment of the Carderock Division of the Naval Surface Warfare Center, Army Research Lab, and NASA Lewis), and partners from Industry (Apple Inc., Texas Instruments, Interval Research, Prometheus Inc., and Silvaco). The center carries out a high quality program of experimental and theoretical investigation of the auditory system and acoustic signal processing with the aim of:

(1) capturing the functionality of the auditory system in the form of mathematical models and signal processing algorithms;

(2) implementing these algorithms in software and hardware, and evaluating the algorithms by comparing their performance to human performance and against a range of robustness and flexibility requirements;

(3) evaluating the usefulness of these implementations for a wide range of applications including: acoustic diagnostic monitoring systems for machines and manufacturing processes; battlefield acoustic signal analysis, sound analysis and recognition systems; robust detection and recognition of multiple interacting faults; and detection and recognition of underwater transients and weak signals in low signal-to-noise ratio (SNR) in acoustically-cluttered environments.

(Barbara Shinn-Cunningham) The resolution with which subjects can identify individual source locations depends upon a number of factors, including their experience and expectation. In particular, the range of positions that a subject expects to hear (or is attending) has a large effect on the resolution they achieve in localization tests. Systematic investigations into the effects of experience and expectation on resolution of sound source location will be conducted using virtual stimuli presented by headphones.

(Barbara Shinn-Cunningham) Previous work has demonstrated that subjects can partially adapt to nonlinear remappings of auditory localization cues, given appropriate training. A quantitative, decision-theory model describing these earlier results suggests that subjects can adapt to linear remappings of localization cues. Preliminary results suggest that, for a linear cue remapping, adaptation is nearly instantaneous and virtually complete. These results will be further investigated, and the implications of these findings will be used to refine our existing model of adaptation.

(Herbert Voigt) Our studies involve the in vivo intracellular recording and labeling study of the dorsal (DCN) and posteroventral cochlear nuclei (PVCN). After conducting a series of extracellular single-unit and two-electrode studies in both the barbiturate anesthetized and unanesthetized decerebrate gerbil, a database of response types for gerbil has been established. Two new, substantial physiological classes based on responses to noise have been discovered. Intracellular responses of cochlear nucleus projection neurons and interneurons to both acoustic stimulation and current-pulse injection through the recording electrode have been recorded. Identification of these neurons using neurobiotin as the cellular marker has resulted in an increase of recovered neurons compared to HRP. We are planning to continue this work in DCN and expand it to the PVCN, where we will identify the neurons that project from the PVCN to the DCN and elucidate their acoustic response properties. In order to understand the relationships between the response properties of units recorded in the decerebrate versus the barbiturate anesthetized preparation, we will study the effects of an ultrashort acting barbiturate on the acoustic responses of units recorded in the DCN of decerebrate preparations. Our work on computational models of the DCN is also continuing. Using a model of the DCN's neuronal circuitry we simplified, yet detailed all comparing physiological responses to tones, noise, and match- noise stimuli to those obtained from the model.

top of page