Auditory BiophysicsArnold Tubis, Professor Emeritus of Physics
Professor Arnold Tubis, who spent the first twenty years of his professional career working in the areas of theoretical nuclear and elementary particle physics, began co-teaching (with Professor Orland E. Johnson) in the late 1970's an advanced undergraduate course on the physics of musical sounds for science and engineering majors. As a result of his teaching experiences, he developed a strong interest in the problems of musical sound production and human sound perception, wrote an important paper on the effects of air loading on the spectral characteristics of the orchestral tympani, and began a research program in auditory biophysics in 1981. His initial efforts in hearing science involved collaborations with Professor Laurence Feth (now at Ohio State University) and Edward Burns (now at the University of Washington, Seattle) of the Purdue Department of Audiology and Speech Sciences. Professor Glenis Long replaced Professors Burns and Feth in the Department of Audiology and Speech Sciences in 1984, and became a member of the program because her previous research was closely related to that being conducted at Purdue. Dr. Carrick Talmadge (now at the National Center for Physical Acoustics, University of Mississippi) joined the program in 1989. Seven Purdue Ph.D. degrees have been connected with the program, and currently four of the recipients are active scientists at leading hearing research institutes (National Institute for Occupational Safety and Health, Boys Town (NE) National Research Hospital, and the Northwestern University Department of Communicative Science & Disorders). The research has been supported by the Deafness Research Foundation, the National Science Foundation, the Showalter Trust, and the NIDCD division of the National Institutes of Health (the source of current funding). Professor Burns, Feth, Long and Tubis have all been elected to Fellowship in the Acoustical Society of America in recognition of their contributions to hearing research.
The year, 1981 was a fortuitous one for entering the field of auditory biophysics because of the discovery in 1977 of the phenomena of otoacoustic emissions (OAEs), and the earlier discovery of the associated microstructure of the hearing threshold (reproducible patterns of maxima and minima as functions of frequency). The OAEs are sounds which can be detected with a sensitive microphone in the ear canal, whose origin is now generally attributed to nonlinear-active (undamping) mechanisms occurring in the cochlea or inner ear.
The historical background of OAEs is very interesting. From observations on cadaver ears in the 1930's, Nobel Laureate G. von Bekesy (who was trained as an electrical engineer) discovered relatively slow (about 1 m/s) traveling waves on the basilar membrane of the cochlea which neatly delivered sound energy of different frequencies to different places along the membrane, where the wave amplitude peaked. This wavelet type of analysis in the cochlea is sometimes referred to as that of an "acoustic prism." The place of peak activity for a given frequency is called the tonotopic place for that frequency. The observed frequency resolution found for normal ears of living beings is much too high to be accounted for by Bekesy's observations. Clearly the problem was associated with energy losses from viscous damping of the structures of a passive ("dead") cochlea. The function of these cochlear structures at the time was attributed to the simple mechanical properties of a partition containing the sensory elements (the Organ of Corti), which separated the two main fluid canals or scalae). Nevertheless, from the 1940's to the early 1980's, almost all auditory physiologists accepted the observations of von Bekesy as being directly relevant to the sensitivity and frequency resolution of normal hearing, and invented various so-calledsecond-filter mechanisms to account for the observed resolution.
In 1948 Thomas Gold (now a distinguished astrophysicist at Cornell University) argued that there must be an active (undamping) mechanism in the cochlea, and he proposed that the cochlea had the same positive feedback mechanism that radio engineers applied in the 1920s and 1930s to enhance the selectivity of radio receivers. Presumably this type of feedback amplification was not operative for the dead specimens of von Bekesy, and therefore they did not exhibit true cochlear function. Gold predicted that in some cases, "abnormal" ears might emit continuous tones due to self-sustained oscillations in a manner roughly similar to the way a defective sound amplification system sometimes begins to squeal because of too much feedback. He searched without success for such sounds from the ears of persons who experienced strong tonal subjective tinnitus (ringing in the ears). Gold did not realize that these sounds (now called spontaneous otoacoustic emissions or SOAEs) as well as possible associated stimulated sounds of cochlear origin (stimulated otoacoustic emissions, which he did not predict) might be associated with the normal (as opposed to the malfunctioning) cochlea. The various types of otoacoustic emissions were finally discovered in 1977 by a British engineer, David Kemp. Also, in the early 1980's, physiological investigations began to show that the mechanical response of the cochlea in living animals was much more sharply tuned than that observed by von Bekesy in cadavers ears, presumably because of the type of feedback mechanism suggested by Gold. Thus auditory theory was unfortunately diverted for over 40 years because insights such as those of Gold were largely ignored.
As had already been mentioned, otoacoustic emissions are sounds that originate in the physiologically vital and vulnerable activity of the cochlea. As such, they are now being seriously proposed as objective indicators of the health of the cochlea. In addition to SOAEs, there are otoacoustic emissions which may be evoked by steady tones (stimulus frequency otoacoustic emissions or SFOAEs), as intermodulation distortion product otoacoustic emissions (DPOAEs) in response to tonal combinations, or transient otoacoustic emissions evoked by clicks, tone bursts, or chirps.
There is growing evidence that the undamping mechanism which underlies the OAEs is rooted in the properly phased electromotility(contraction/expansion controlled by the cell membrane potential) of theouter hair cells (OHCs) of the Organ of Corti in the cochlea. The compressive mechanical nonlinearity observed in cochlear function appears to make the cochlea operate as a sensitive high-Q detector of low level sounds and a less sensitive lower-Q detector of moderate to high levels of acoustic stimulation. The nonlinear mechanical response of the cochlea constitutes nature's way of compressing (without significantly degrading the signal) the 120 dB dynamic range of audible acoustic pressure variation into an approximate 50 dB range of basilar membrane or associated inner hair cell (IHC) cilia displacement, with the latter corresponding to the dynamic range of the mechano-electrical-neural transduction in the IHC's.
In order to provide a firm foundation for clinical applications of OAE's to the objective screening for hearing impairments, it is very important to extensively investigate the properties of spontaneous and evoked otoacoustic emissions under a variety of stimulation conditions. The data obtained, in combination with the results of psychophysical and neurophysiological studies, then provides the basis for a detailed physical/mathematical model of what is going on in the cochlea in relation to OAE measurements obtained in the ear canal. This dual experimental/theoretical approach to OAE's has been the main focus of the Purdue effort during the last decade.
Some of the major accomplishments of the Purdue Auditory Biophysics Group are briefly described below.
- The development of new off-line techniques for the measurement of spontaneous otoacoustic emissions. Some ears have been found to have over 35 narrow-band SOAEs with bandwidths ranging from 0.5 Hz to 200 Hz.
- A comprehensive study of the perceptual consequences of otoacoustic emissions.
- The discovery of novel interactions between SOAEs includingintermodulation distortion products mutual suppression, and interactions giving rise to linked alternate-state spontaneous emissions.
- Experimental and theoretical studies that firmly connect large spontaneous otoacoustic emissions to noise-perturbed limit-cycle oscillations in the cochlea. These involve the acquisition and model interpretation of data on the statistical properties of SOAEs, suppression and frequency pushing/pulling/locking of SOAEs by external tones in steady state or pulsed modes, the reversible effects of aspirin administration on these measures, the modulation of SOAE properties by cochlear blood flow, and ear canal reflectance in the neighborhood of a SOAE.
- Experimental and theoretical studies in support of a comprehensive model of the cochlea that relates the various types of OAEs and the microstructure of the hearing threshold.
The basic ingredients of the model of OAEs and threshold microstructure used are the following:
- An underlying linear-active cochlear model which gives, for a single frequency acoustic stimulus, a tall and broad traveling wave envelope along the basilar membrane (with "broad" implying that at least two wavelengths of traveling wave are contained in the peak region).
- A low level (of the order of 0.5 to l%) of inhomogeneity of cochlear properties along the cochlear partition, which will give rise to the reflection of cochlear waves (which would otherwise travel only from the cochlear base to the apex), with the main reflection region being near that of peak basilar membrane activity. This reflection mechanism was originally proposed as a physical basis for stimulus frequency emissions by George Zweig (a MacArthur Fellow and one of the originators of the quark model of elementary particles, who has been working in hearing research for the past twenty years) and Christopher Shera.
- Cochlear nonlinearity associated with the "active gain control" of the basilar membrane.
The interference phenomena associated with cochlear reflections can be shown to give a good description of most of the data on all types of OAEs and the microstructure of the hearing threshold, except for distortion product OAEs, which obviously require cochlear nonlinearity as a primary generation mechanism. Spontaneous emissions arise as discrete instability modes of the linear active cochlear mechanics, and are stabilized by the cochlear nonlinearity so as to become limit cycle oscillations. The model gives a natural explanation of the fact that the frequency spacing of neighboring spontaneous emissions and the fine structures (separation between successive maxima or minima in the frequency dependence) of evoked OAEs and the hearing threshold are all approximately equal to the change in the local resonant (tonotopic) frequency associated with a displacement of about 0.4 mm along the basilar membrane.
Distortion product OAEs may be of most importance for the objective clinical screening for hearing disorders. They are fairly easy to measure, and by using a variable frequency two-tone complex of frequencies f1 and f2(>f1), with the ratio f2/f1 fixed at 1.2, it is possible to systematically vary the primary generation region of distortion product production from one end of the cochlea to the other. By monitoring the level of the DP components in the ear canal (the 2f1 - f2 one is normally the largest), the integrity of the cochlea in the local generation region may be probed. However, the observed DPOAE may be significantly altered as a result of interference between the DP signal reaching the ear canal directly from the generation region and another one reaching the ear canal after being reflected in the cochlea from around the tonotopic location associated with the DP frequency. This alteration must be taken into account if the DPOAEs are to be useful indicators in hearing screenings.
Several schemes have been successfully developed for obtaining the DPOAE signal from the primary generation region only, even when the above mentioned interference is operative. One procedure is based on suppressing the reflected wave in the cochlea. The other is a dynamic scheme in which one of the primary tones is pulsed on and off, and the intially generated DPOAE component can be separated in time from the reflected one. These schemes have the potential of greatly improving the reliability of hearing screenings based on DPOAEs.
It is hoped that the work of the Purdue hearing research group will help make the objective asssessment of hearing based on DPOAEs and other types of otoacoustic emissions both reliable and routine, especially in the case of infants for whom psychophysical measures of hearing are extremely difficult to obtain.
In addition to its value to society, the investigations of sensory systems from a physical point of view, in conjunction with knowledge obtained from neurophysiological and psychophysical studies, provides many interesting challenges in the areas of signal processing, experimental design, and the mathematical modeling of complex systems with mechanical, acoustical, electrical, and biological elements.
Professor Tubis was born on 28 March 1932 in Pottstown, Pennsylvania. He received the B.S. at MIT in 1954 and the Ph.D. at the same institution in 1959. Before coming to Purdue as a Research Associate in 1960, he was Assistant Professor at Worcester Polytechnic Institute for two years. At Purdue he was promoted to Assistant Professor in 1962, to Associate Professor in 1964 and to Professor in 1969. He was Assistant Head of the department 1966-73, Acting Head of the department during 1988-89 and Head of the department during 1989-1997. Prior to 1981, his field of research was theoretical nuclear and high energy physics and since then he has worked in the field of acoustics and hearing research. He has guided the Ph.D. thesis research of ten students. He is the author of more than 100 refereed publications and is a fellow of the American Physical Society and a Fellow of the Acoustical Society of America.