Bryan E. Pfingst - US grants

Two mechanisms are believed to underlie the perception and discrimination of pitch: spectrally-based place mechanisms and temporally-based periodically mechanisms. Nonspectral pitch can be studied by using special acoustic or electrical signals which lack spectral cues such as sinusoidally amplitude modulated (SAM) broadband noise, or periodic electrical signals delivered to a deaf ear lacking hair cells. A basic assumption underlying the use of these signals to study nonspectral pitch discrimination is that the signal periodically dominates the discrimination, and other aspects of the signal are relatively unimportant. We will test this hypothesis by comparing, in the same subject, frequency difference limens (measured with careful control to eliminate loudness cues) for different stimulus types which are expected to produce different but predictable neural responses, and by measuring the effects of stimulus levels and the effects of nerve survival patterns on these difference limens. The subjects will be psychophysically trained nonhuman primates with one normal ear and one ear deafened by local perfusion of neomycin and implanted with a scala tympani electrode array. After comparisons between the normal and the implanted ear are completed, effects of a partial lesion of the auditory nerve will be studied on the unimplanted side. We will also compare the performance of the nonhuman primates on the SAM noise discrimination task with performance of human subjects tested under the same conditions to determine if human-animal differences seen with pure tone frequency discrimination also apply to nonspectral frequency discrimination. These studies are directly relevant to the design of processors for cochlear implants and to audiological assessment of auditory nerve fiber damage. They will also lead to a better understanding of stimuli used in psychophysical and neurophysiological investigations related to nonspectral pitch and the applicability of these studies to humans.

Implantable auditory prostheses comprise one of the most exciting and rapidly changing developments in Otolaryngology and in sensory neuroscience. They are in fact the only form of rehabilitation for profound deafness which restore some degree of hearing to the patient. Further development of these prostheses requires combined efforts in a variety od disciplines ranging from anatomy, physiology and biophysics to psychophysics and speech science. The 1989 Engineering Foundation on this conference on implantable Auditory Prostheses is the next in a series of biennial international research meetings in which scientists from this field, as well as those from related fields, come together for 4 1/2 days of lectures and discussions. The meeting format is designed to maximize the interaction of scientists from the many disciplines which influence the development of these devices, to encourage exchange of the most recent data and research ideas, and to stimulate and enhance the future research in this field. This application seeks part of the funds to reimburse speakers and discussants for travel and registration expenses. Without these funds, we cannot be assured of obtaining the best participants from the United States and abroad

DESCRIPTION(adapted from applicant's abstract): Deafness resulting from destruction of the inner ear is accompanied by partial degeneration of the auditory nerve as well as anatomical, physiological and molecular changes throughout the central auditory nervous system. These changes vary considerably among deaf individuals. By hypothesis, these individual differences are associated with the large differences across subjects in psychophysical and speech recognition performance with cochlear implants. In the set of experiments proposed in this application, we will use neurophysiological measures to characterize the large psychophysical changes that occur following deafening and implantation of the cochlea. We will then examine the functional effects of post-deafening treatments, such as chronic electrical stimulation of the cochlea, that are intended to retard the long-term deleterious effects of deafness on the auditory system. The functional responses to electrical stimulation through cochlear implants will be assessed at the psychophysical level in humans and guinea pigs by measuring detection threshold functions and discrimination functions, and at the neurophysiological level in guinea pigs by measuring spatial and temporal responses of neurons in the inferior colliculus to electrical stimulation and by measuring gross potentials that reflect summed neural activity at peripheral (ECAP and EABR) and central (EMLR) sites along the auditory pathway. Three aims will be addressed. In Aim 1 we will compare psychophysical responses obtained from humans and guinea pigs to determine the relevance of guinea pig data to humans. In Aim 2 we will characterize the pronounced short-term effects of deafening and implantation procedures on the psychophysical and neurophysiological responses to electrical stimulation using the guinea-pig animal model. In Aim 3, we will test hypotheses about the functional benefits of post-deafening treatment of the ear. The results of these experiments will lead to better use of animal (guinea pig) models for auditory prosthesis experiments, they will provide a better understanding of the functional (diagnostic) implications of commonly used psychophysical measures such as detection threshold level, and they will have direct application to treatment protocols for deaf and severely hearing-impaired patients.

DESCRIPTION (provided by applicant): This interdepartmental training program provides a broadly based predoctoral and postdoctoral education and research experience in hearing, balance and chemical senses. Predoctoral training is based in one of the 14 basic science departments and degree-granting graduate programs with which the 25 training faculty are associated. Doctoral thesis research and postdoctoral research training reside primarily within the laboratory of the mentor. In addition to research training, the program includes a) introductory and advanced courses in sensory systems; b) seminars in hearing, balance and chemical senses with experts from within and outside the University of Michigan; c) student and faculty seminars, journal clubs and research seminars; and d) training in research standards and ethics. Support for 5 predoctoral and 4 postdoctoral trainees is requested. Predoctoral trainees will be selected from the most highly qualified graduate students in the training faculty's affiliated departments and programs. Interest and motivation for research in sensory systems will be an important criterion. Postdoctoral trainees must have a doctoral degree such as a Ph.D. or M.D. and a strong commitment to sensory biology. The program's Admissions Committee and Executive Committee will evaluate professional promise on the strength of prior academic performance, letters of recommendation and publication record. Support is also requested to recruit the deaf minority into biomedical research. The deaf are the largest underrepresented handicapped population in our society, and they are underrepresented in scientific and medical professions. We propose to continue our very successful program for deaf undergraduates, begun in 1990, by training 8 deaf students per year in summer research internships. Faculty members have current research support and well-equipped laboratories offering state-of-the-art methodology in a variety of specialties, including immunological, biochemical, molecular and genetic approaches to sensory systems, intra- and extra-cellular electrophysiological recording, modern morphological and physiological techniques including computerized image analysis, and human and animal psychophysics. Modern AAALAC-accredited animal care facilities are maintained by the Unit of Laboratory Animal Medicine, and the U. of M. computing facilities and libraries are outstanding.

Previous studies have shown that electrode configuration has a large effect on speech recognition scores of deaf patients with cochlear implants (auditory prostheses). Different hypotheses about the mechanisms of these effects all refer to the spatial pattern of evoked neural activity. PET imaging at the level of human auditory cortex will be used to determine the spatial pattern of activation for the different configurations and to compare these patterns with those found in response to acoustic stimulation in normal hearing subjects.

DESCRIPTION: (as per applicant abstract) One of the most exciting advances in auditory prosthesis research in the last five years is the demonstration by our laboratory and others that speech recognition with cochlear implants can be significantly affected by the configuration of the electrical stimuli delivered to individual channels of the implant. The objectives of this proposal are to define the functional effects of specific spatial configurations of the electrical stimulus using speech recognition and psychophysical measures. We will expand on previous studies to define the processing strategies and types of stimuli for which various configurations of the electrical signal are most effective. Human subjects with Nucleus 22, Nucleus 24M and Clarion prostheses will be used. Psychophysical measures will include difference limens for pulse-rate, modulation-frequency, electrode-place, and intensity. Stimuli for speech-recognition testing will include phonemes, words, and complete sentences tested in quiet and in noise. The studies will be guided by models of the spatial extent and pattern of central neural excitation generated by specific electrical- current configurations. Psychophysical threshold, loudness, and gap detection data that have been proposed to reflect spread of neural activation will be obtained for the various configurations. This series of studies is expected to lead to improved performance in patients with cochlear implants by (a) influencing the procedures for fitting today's processors to individual patients and (b) guiding the design of electrode arrays and processors for the implants of the future.

We have recently demonstrated that new inner hair cells (IHCs) are generated in the mammalian organ of Corti following gene therapy. Specifically, the experiments involve in vivo inoculation of an adenovirus vector with the Math 1 gene insert into the mature predeafened guinea pig cochlea. We have also demonstrated that the new hair cells can attract neurons. Predeafend ears treated with Math 1 exhibit partial restoration of function as determined by Preyer's refex and significant improvement in acoustically-evoked auditory brainstem response (ABR) threshods. In the current application we propose to use animal-psychophysics techniques to define the nature and extent of the restored acoustic hearing. We will also conduct experiments relevant to the use of this gene therapy in conjunction with cochlear implants. Specifically, in Aims 1 and 2 we will deafen guinea pigs with ototoxic drugs, treat them with Math 1 and measure psychophysical detection and discrimination thresholds as well as ABRs. In Aim 3, we will assess the interaction between Ad.Math 1 inoculation, cochlear implantation, and electrical stimulation of the cochlear implant. In Aim 4 we will place a cochlear implant in the inoculated ear and assess psychophysical responses to electrical stimulation with and without restored IHCs. The experiments we propose involve the use of techniques that are already in place in our laboratories. The work is backed by strong preliminary data, and represents an ideal combination of relatively low-risk with very high impact. The proposed experiments will bring this gene therapy application closer to clinical use.

We have recently demonstrated that new inner hair cells (IHCs) are generated in the mammalian organ of Corti following gene therapy. Specifically, the experiments involve in vivo inoculation of an adenovirus vector with the Atohl gene insert (M.Atohl) into the mature predeafened guinea pig cochlea. We have also demonstrated that the new hair cells can attract neurons. Predeafened ears treated with Atohl exhibit partial restoration of function as determined by Preyer's reflex and acoustically-evoked auditory brainstem response (ABR) thresholds. In the current application we propose to use animal- psychophysics techniques and physiological measures to define the nature and extent of the restored acoustic hearing. We will also conduct experiments relevant to the use of this Atohl therapy in conjunction with cochlear implants. For the first two specific aims we will deafen guinea pigs with ototoxic drugs, treat them with Atohl and measure psychophysical detection (Aim 1) and discrimination (Aim 2) as well as ABRs and distortion product otoacoustic emissions (DPOAEs). In Aim 3, we will assess the interaction between M.Atohl inoculation, cochlear implantation, and electrical stimulation of the cochlear implant. In Aim 4 we will place a cochlear implant in the inoculated ear and assess psychophysical responses to electrical stimulation and electrically evoked compound action potentials (ECAPs) with and without restored IHCs. Controls with some surviving hair cells (i.e., animals that are implanted but not predeafened) will be used for comparison to animals with restored hair cells. The experiments we propose involve the use of techniques that are already in place in our laboratories. The work is backed by strong preliminary data, and represents an ideal combination of relatively low-risk with very high impact. The proposed experiments will enhance our understanding of the functional significance of hair cell regeneration in the mammalian cochlea and contribute to the development of a clinical therapy for sensorineural deafness involving the combined use of cochlear implantation with cell replacement therapy.

DESCRIPTION (provided by applicant): The long-term objective of our research is to increase the benefit that hearing impaired patients receive from their cochlear implant auditory prostheses. The approach is to improve the custom-fitting strategy by which the auditory prosthesis speech processor is adjusted to meet the characteristics of the individual patient. To do this, we need to understand the root causes of performance deficits. In our preliminary studies we have shown that most subjects can discriminate the fundamental spatial and temporal features of prosthetic stimulation needed for speech recognition, but that the ability to discriminate these features depends on the parameters of stimulation, including the location of the stimulating electrodes within their electrode array and input-output function as reflected in loudness growth. Performance varies from one stimulation site to another within an individual's cochlear implant and the pattern of this variation across stimulation sites is patient specific. These results suggest two strategies for optimizing patient performance. One strategy is to create a processor map that utilizes only those stimulation sites where performance on basic perceptual tasks is best. In support of this approach, previous studies have shown that patients tolerate a reduction in the number of stimulation sites well and that there is sometimes an improvement in performance when sites where perception is poor are removed from the processor map. A second strategy is to individually optimize the stimulation parameters on a site-by-site basis. It is known for example that stimulus level can be adjusted to improve temporal acuity. Implementation of these two strategies requires a better understanding of the basic principles underlying across-site patterns of perception in cochlear implants. Specifically, the following questions must be addressed. (1) To what extent are across-site patterns of perception similar for all perceptual measures? For example, do the underlying conditions that cause poor temporal perception also affect spatial resolution? (2) What are the relative contributions of the various psychophysical acuities to speech recognition? For example, would it be better to enhance temporal acuity at the expense, if necessary, of spatial acuity? (3) Do spatial and temporal acuity show similar dependency on stimulus level? These issues will be addressed using psychophysical and speech recognition studies in implanted human subjects. The work will deepen our understanding of the mechanisms underlying across-subject variation in speech recognition performance with cochlear implants and serve as a guide for establishing and testing clinical procedures to improve performance in individual patients.

? DESCRIPTION (provided by applicant): The objectives of our research are (1) to determine how the health of the implanted cochlea, i.e. the biological conditions near the individual cochlear-implant electrodes, affects specific psychophysical and electrophysiological measures of electrical hearing; (2) to determine the relationships of these specific measures to speech recognition with the cochlear prosthesis; and (3) to use this information to increase the benefit that hearing impaired patients receive from their prostheses. The data from these studies can be used in two ways to improve speech recognition in cochlear implant users. First, based on animal work that will correlate the pattern of pathology with functional measures, we will provide audiologists with simple clinically applicable measures they can use to gain insight into the characteristics of the individual patient's cochlea and better identify and select the best stimulation sites for an individual patient's speech processor MAP. Second the data can help the biologist and the surgeon to determine the best anatomical targets for improving implant function through tissue-preservation and tissue-engineering strategies to make the impaired cochlea more receptive to cochlear implant stimulation. Our approach involves psychophysical and electrophysiological experiments in guinea pigs as well as psychophysical, electrophysiological and speech recognition studies in humans. We measure psychophysical performance, such as perceptual integration of pulse trains, and electrophysiological performance such as the rate at which evoked neural responses grow as a function of stimulus level. These measurements are made at individual stimulation sites in guinea pigs and humans. In guinea pigs we determine the specific anatomical features in the deaf or hearing-impaired cochlea that are correlated with these measures. In humans we determine the correlation of these same measures with speech recognition in quiet and in noisy backgrounds. We can then use these measures in humans to select the best stimulation sites for an individual subject's speech processor. This approach is supported by our previous studies showing that subjects usually perform better using a processor MAP with a subset of stimulation sites, carefully selected using appropriate functional measures, than they do with a processor that uses all available sites. The work proposed in this application will deepen our understanding of the mechanisms underlying variation in speech recognition performance across users of cochlear implants and serve as a guide for establishing and testing clinical procedures that will improve performance in individual patients.

Project Summary/Abstract The objectives of our research are (1) to determine how conditions in the cochlea near the individual cochlear-implant electrodes affect specific psychophysical and electrophysiological measures of electrical hearing; (2) to determine the relationships of these specific measures to speech recognition with the cochlear prosthesis; and (3) to use this information to increase the benefit that cochlear implant patients receive from their prostheses. The data from these studies can be used in two ways to improve speech recognition in cochlear implant users. First, based on animal work that will correlate the pattern of pathology with functional measures, we will provide audiologists with simple, clinically-applicable measures they can use to assess individual stimulation sites in a patient's cochlea and guide selection of the best stimulation sites for an individual patient's speech processor MAP. Second, the data can help research scientists and surgeons determine the best anatomical targets for improving implant function through tissue-preservation and tissue- engineering strategies that make the impaired cochlea a better recipient of prosthetic stimulation. Our approach involves psychophysical and electrophysiological experiments in guinea pigs as well as psychophysical, electrophysiological and speech recognition studies in humans. We will measure psychophysical performance, such as perceptual integration of pulse trains or phase duration, and electrophysiological performance such as the rate at which evoked neural responses grow as a function of stimulus level. These measurements will be made at individual stimulation sites in guinea pigs and humans. In guinea pigs, we will determine the specific anatomical features in the hearing-impaired cochlea that are correlated with these measures. In humans we will determine the correlation of these same measures with speech recognition in quiet and in noisy backgrounds. We will then use these measures in humans to select the best stimulation sites to include in individual subjects' speech processor MAPs. This approach is supported by our previous studies showing that subjects usually perform better using a processor MAP that includes a subset of stimulation sites, carefully selected based on appropriate functional measures, than they do with a MAP that uses all available sites. The work proposed in this application will deepen our understanding of the mechanisms underlying variation in speech recognition performance across users of cochlear implants and serve as a guide for establishing and testing biological and clinical procedures that will improve performance in individual patients.