The Search for the Mechanisms of Hearing
Peter Dallos, PhD
The remarkably effective ear has an intricate intercoordination of parts which scientists still cannot explain.
Human beings are able to process sounds-from the quietest whisper of the forest to the roar of a close up engine-with remarkable precision. The extremes of loudness and softness that we encounter possess energies whose ratio is some hundred-million-million to one. At the lowest limit, the softest sound we can hear, the eardrum may move a thousand-millionth of an inch; at the top of the range, the movement is so great that the hearing sensation gives way to pain. And across this astounding span, we are also able to detect minute changes in intensity.
We can, moreover, hear very small changes in both the frequency of sound (corresponding to its pitch) and its duration. A normal young adult human can hear sounds whose frequencies extend from a rumble to a squeak, from approximately 20 to 20,000 Hz (the name of the scientific unit for cycles-per-second is Hertz, Hz for short). And humans can reliably tell a difference in frequency as small as 0.1 percent. That is, we recognize that a sound of 1,000 Hz is not the same as one whose frequency is 1,001 Hz.
To get a hint of the auditory system's remarkable capacities, consider a single microscopic hair, or cilium, on one of the sensory-receptor cells in the ear. The movement of cilia is important, since they act as microlevers to transmit sound-related movements to their sensory-receptor cells-which in turn convey the implied information to the brain. If you scale the dimension of that one cilium to the height of Chicago's Sears Tower, the movement of the tip of the cilium at the threshold of hearing is equal to a two-inch displacement at the top of the Tower!
Such auditory abilities pose a formidable challenge to our understanding. What sort of mechanism did Nature devise that can perform these feats? What happens when parts of the mechanism go awry? How can we remedy such defects?
The cross-section shows the external ear and ear canal, which funnel sound waves arriving from the environment onto the eardrum. This flexible membrane is set into vibration by sound and, in turn, couples its vibrations to three interconnected bones in the middle ear. These bones form a link between the drum and the inner ear. The inner ear, or labyrinth, is a maze of interconnected fluid-filled channels hollowed out of bone. Suspended within the channels is a closed system of membranous specialized regions that serve as sense organs. The subdivision called the cochlea contains the actual auditory sensory structure, known as the organ of Corti.
Cross section of the human ear. Click on the image for a larger and more detailed diagram.
Cross section of the cochlear duct. Click on the image for more details.
The cochlea is a fluid-filled cavity in the form of a snail shell. The middle ear delivers sound-related vibrations to the inner-ear fluids at the base of the cochlea, where its channel is widest. The diameter of the channel diminishes from base to apex. The fibers of the auditory nerve enter through the cochlea's hollow central core, fan out through its whole length, and connect to structures within the organ of Corti. The other ends of the nerve fibers are in the brain stem. The fibers receive information about the auditory world from the organ of Corti and send it to the brain, where it is interpreted as a cat's meow or a Bach partita. Other nerve fibers carry command signals back from the central nervous system to the hearing organ.
Membranes subdivide the cochlea into three compartments, all of which run throughout its length. On the membrane forming the bottom of the central compartment is the organ of Corti, the actual sense organ. This complex structure is made up of an ordered array of supporting cells-these provide structural stability-and among the supporting cells two types of sensory-receptor cells: inner and outer hair cells. Inner hair cells are arranged in a single row running the length of the organ of Corti from base to apex, and there are about 3,500 of them. There are almost four times as many outer hair cells, which run in three parallel rows, also the length of the cochlea.
The fibers of the auditory nerve terminate on both types of hair cell, but their distribution is very uneven. About 95 percent of the nerve fibers contact the inner hair cells; the more numerous outer hair cells receive the remaining approximate 5 percent.
As sensory receptors, the hair cells are the link in the auditory system where mechanical vibrations, derived from an incoming sound, are changed into signals that are compatible with the language of the nervous system. Considering that the inner hair cells provide some 95 percent of the information that travels to the brain, we may safely assume that they constitute the main avenue for information about the sounds around us. The role of outer hair cells is less clear, and some intriguing possibilities are considered below.
However, to understand the most current theories of the function of the hair cells, it is necessary first to have a clear understanding of their surroundings.
Light microscope photo of the cochlea in cross section. Click on the image for a larger picture.
The organ of Corti, site of the hair cells, rests upon the basilar membrane and is overlain by the gel-like tectorial membrane. Relative movement between these two parallel surfaces generates fluid currents in the channel between them. It is these currents and the relative movement of the two surfaces which convey to the hair cells the auditory vibrations.
Detail of the hair cells from the organ of corti. Click on the image for more details.
The surface of the organ of Corti. Click on the image for more details.
The movement between the two surfaces arises because of their arrangement in space. They have different points of suspension, and so when they are both deflected, they slide slightly, relative to one another. Their movement is something like that between the pages of a book which is suspended from its spine while its covers are deflected together.
The receptor hair cells, both inner and outer, have fine, membrane-bound extrusions, called hairs or cilia, protruding from their top segments. As these cilia are deflected, their minute local deformations modify an electrical current that flows into their receptor hair cells. In the hair cells, this electrical current is further processed to initiate chemical messengers to the auditory nerve fibers, which, in turn, generate the electrical impulses that carry information to the brain. The tectorial membrane provides the mechanical stimulation to the cilia and thus to the sensory cells. The cilia of outer hair cells are firmly attached to the underside of the tectorial membrane so that they are deflected by the relative sideways sliding motion between the tectorial membrane and the organ of Corti. However, the inner hair cell cilia have no, or at most tenuous, contact with the tectorial membrane and are moved by the force of the fluid streaming in the narrow channel between the tectorial membrane and the organ of Corti, as previously noted.
The cilia of the single row of inner hair cells on the surface of the organ of Corti. Click on the image for more details.
A side view of the organ of Corti. Click on the image for more details.
The central compartment of the cochlea, containing the organ of Corti, is called the cochlear partition. The movement of the partition is a result of a complicated set of hydraulic phenomena that culminates in a wave motion.
The schematic drawing partially unrolls and unwraps the normal snail-shell configuration of the cochlea, and it employs two other simplifications: the entire cochlear partition is represented as a single surface, and the wave displacement of this partition is grossly exaggerated.
Traveling wave superimposed on the cochlear partition. Click on the image for more details.
When a pure musical tone, whose physical representation is a periodic sound-pressure variation, falls on the ear, the eardrum and the linked bones of the middle ear vibrate in a similar to-and-fro motion. This motion creates a periodic pressure change across the cochlear partition within the cochlear fluid. As a result, alternating up-and-down forces are exerted on the partition, and it too is set in motion. These vibrations, which originate near the base of the cochlea, do not remain confined to that locale. In fact, the energy exchange between fluid and partition results in a wave. The discovery and description of this traveling wave earned Georg von Bekesy the Nobel Prize in 1961.
Alternating crests and troughs travel from the base of the cochlea toward its apex. Of course, any one point on the cochlear partition simply moves up and down at the same frequency as the sound; what travels is the wave effect. Notice that from point to point along the length of the wave, the displacement of the partition changes. At the beginning, it is small, then it grows to a maximum, then subsides. Far towards the apex, the partition is at rest. For any frequency, there is a unique point of maximum displacement, or peak, and this it is said that the location of the traveling-wave peak is frequency-dependent. In other words, the traveling wave represents a process of mechanical frequency analysis or discrimination. How good, how discriminating, is this analysis?
Early observations suggested that, in fact, the traveling wave's frequency analysis is not especially refined, and there is an enormous discrepancy between the known frequency-analyzing capabilities of a human being-recall our ability to discriminate between tone whose frequencies are only one Hz apart-and that of the traveling-wave mechanism. In order to explain the discrepancy, some researchers argued that the neutral interactions must sharpen the poor tuning of the cochlear mechanics. Such explanations relegated the difficult task of producing the exquisite frequency selectivity of the hearing organ to the central nervous system.
But by the 1960's, extensive recordings of the electrical activity in fibers of the auditory nerve showed that the necessary frequency selectivity was already established at the preneural level of the auditory system. It also became evident that auditory-nerve fibers make very simple one-to-one connections with hair cells; no complex network required by a scheme of neural interactions was in evidence. So the neural-sharpening idea died a quiet death and left behind a void which is just now being filled. The current view is that the phenomenal frequency-selective properties of the auditory system are entirely established in mechanical processes within the cochlea.
To reach this conclusion, improved measuring techniques were a necessary but not a sufficient condition. More important was the acceptance of two ideas (neither was new but both had been largely disregarded when first proposed). In the first place, the vibration of the cochlear partition is highly nonlinear. The amount of displacement does not grow in proportion to a rising amount of sound, but a much lesser rate. Moreover, this nonlinearity is evident only in the vicinity of the frequency to which a given location responds optimally; at lower and higher frequencies, the amount of displacement is proportional to the level of the sound. A consequence of this peculiar nonlinearity is that the quieter the sound that drives the cochlea, the sharper the tuning represented in the traveling wave. This idea of nonlinearity goes a long way toward explaining the discrepancies between earlier observations and today's measurements. What it implies is that in order to get maximum sharpness of tuning, measurements need to be taken at relatively soft sound levels. Earlier investigators had to rely on visual observations and, in order to make the partition movements visible, had to use sound levels at or above the highest limit of auditory experience. Contemporary measurements, using methods borrowed from nuclear physics, allow the use of sounds at about the level of quiet conversation.
The second idea is more interesting. Several investigators, even though they were using modern techniques, could not see the nonlinearity and dismissed it as a species-specific peculiarity. Today it is clear that the non-linearity exists-but only if measurements are made with great care-and the key to our new clarity was the discovery that mechanical tuning and the degree of nonlinearity vary in a most radical manner with the condition of the experimental animal.
Thus, both tuning and nonlinearity are physiologically vulnerable: interference with the cochlea or deterioration of an animal subject's health leads to a decrease in the sharpness of tuning and in the nonlinear behavior. Early experiments had been performed on cadaver ears or drained cochleas or animals that tolerated the rigors of experimentation poorly. To obtain sharp mechanical tuning one must measure at low sound levels in exquisitely maintained animals.
Sharp mechanical tuning is thus seen to be intimately tied to a process that is nonlinear and physiologically vulnerable.
These characteristics and some others suggest that activity inside the cochlea is more than simple, passive, and mechanical. In other words, the traveling-wave mechanism, which produces the tuning, may do more than convert sound energy into mechanical vibratory energy. It is likely to utilize some additional, metabolic energy source.
Certain investigators contend, in fact, that only by assuming the presence of an active, energy-producing mechanism in the cochlear partition can the sharp mechanical tuning be explained. And aside from theorizing, some experimental observations almost compel one to accept the presence of active processes in the cochlea. For instance, after the cessation of brief sound-stimuli, echoes in the ear canal have been measured with sensitive microphones. These vibrations are demonstrably born in the cochlea as an aftereffect of the stimulation and are retransmitted by the bones of the middle ear and the eardrum to produce the sound. These acoustic emissions clearly contain greater energy than could be expected from a purely passive reflection.
In some subjects, a brief sound-burst can produce a lasting or even continuous emission of sound from the ear; and there are animals, and some human subjects, who produce spontaneous emissions of sounds. All of these originate in the cochlea and in the vibrations of the cochlear partition, and imply the presence of active vibratory processes in the inner ear. The mechanisms are nonlinear, sharply tuned, and physiologically vulnerable. It is thus not farfetched to relate them to the filtering process of the traveling wave, which is itself nonlinear, sharply tuned, and physiologically vulnerable. In fact, it is possible that the active processes disclosed by the echoes and emissions from the ear are the very mechanisms that enable the cochlear partition to produce its sharp tuning.
At present, it seems best to assume that incoming sound produces a traveling wave which, in the region of its maximum vibrations, initiates a process that draws on a local metabolic-energy pool and feeds vibratory energy back to the partition. Thus, around the maximum displacement of the traveling wave, the partition assumes greater amplitudes than it could if it were driven only by the incoming sound. Such a process can account for both the remarkable sensitivity and frequency selectivity of the ear.
Probably the most straightforward means of producing the required sensitivity is to generate a force upon the partition in a direction and with proper timing so as to boost the ongoing movement. Thus the presence of some motor elements within the cochlear partition seems to be required, and the most likely candidate for the role of motor element is-the outer hair cell! This is an audacious suggestion, and the evidence for it is largely circumstantial. What structural basis, if any, exists for the notion that the outer hair cell operates as a motor?
Aside from the striking difference in the number of nerve fibers that receive their information from inner and outer hair cells, and the different means whereby the central nervous system influences the hair cell systems, the actual structural properties of the two cell types are strikingly different. We have already noted the relation between the tectorial membrane and the cilia emerging from outer and inner hair cells. Among other differences, the most interesting is the presence, inside the envelope of the outer hair cell, of many subsurface layers whose presumed role is to store calcium. The outer hair cells are virtually unique in the prominence and spatial organization of these layers, and it is known that calcium must be present for either secretory or contractile activity in cells. It has been found through experiments that a moderate hearing loss results when outer hair cells alone are destroyed. And the region of the frequency of the loss is quite clearly related to the spatial extent of missing outer hair cells.
But the absence of outer hair cells also produces other quite dramatic effects. When studying the response properties of auditory-nerve fibers originating in an area having no outer hair cells, researchers have learned that the fibers' tuning characteristics are altered. The sharply tuned segment is either eliminated or reduced. The implication of these experiments is that the outer hair cells influence the inner ones. It has been suggested that this interaction takes the form of a boost of inner hair cell activity by some motor action of the outer hair cells-and moreover, that the boost is frequency-dependent.
One additional observation is most germane. Several nonlinear effects, very pronounced in the normal ear, disappear entirely when outer hair cells are destroyed. These nonlinearities are commonly associated with mechanical events within the cochlea. The overall implication is that the absence of outer hair cells profoundly affects the neural outflow from the cochlea, an outflow which comes mostly from inner hair cells. An additional implication is that the influence is due to an effect of outer hair cells on the mechanical properties of the cochlear partition.
The latter notion is reinforced by the following. Certain cochlear nonlinearities are manifested by peculiar sounds emitted from the ear-peculiar because they are sounds whose frequencies were not present in the input. They are produced by the nonlinear mechanical behavior of the cochlear partition and are transmitted backwards by the chain of bones in the middle ear and the eardrum to appear in the ear canal. These sounds can be influenced by electrically activating the descending neural tract that leads to the organ of Corti.
The significance of this fact can be appreciated by recalling that the termination of these fibers is either on the neurons originating from inner hair cells or on the bodies of outer hair cells.
Since it is hardly conceivable that the former could mediate a mechanical event in the cochlea, which is needed to interfere with the sound emissions from the ear, the action must be related to the latter. The almost inescapable conclusion is that the neural signals traveling down the fibers activate a mechanical change of the outer hair cells which modify the mechanical properties of the organ of Corti and hence of the cochlear partition. Thus the evidence for a motor behavior of outer hair cells is mounting. In some very preliminary experiments with cell cultures, an alternating electrical field placed across isolated outer hair cells resulted in alternating contracting and expansions of the cell bodies along their longitudinal axes. No such shape change could be observed for inner hair cells.
What might be the mechanism of the motor action of outer hair cells, and how might it influence inner hair cells? Recent research has shown that the region of the hair cell where cilia are anchored contains all the proteins that are normally associated with active nonmuscle contractile systems. If in fact contractions are possible, then the site of the active region is strategically situated. The best means for outer hair cells to influence cochlear mechanics would be via their cilia. Mechanical changes at the bases of the cilia can be transmitted to these structures. Since the cilia of outer hair cells are firmly connected to the tectorial membrane, a mechanical change should influence the stimulation of the inner hair cells. Why? Because the inner hair cells depend on the fluid flow between tectorial membrane and organ of Corti, and that flow is driven by the differential movement of the two surfaces.
Experimental evidence for the existence of such a process is strong, and the implication is that, at least in the cochlea of the mammal, a sensory-receptor cell has developed motor capabilities. Since outer hair cells do send some, albeit few, nerve fibers to the central nervous system, it is likely that those cells perform both sensory and motor functions. This is a most peculiar state of affairs and may be a unique occurrence in the nervous system of mammals.
Outer hair cells are clearly more influential in the hearing process than their minuscule nervous-system ties would suggest. Their importance is underscored by their fragility. Virtually any damage to the auditory system is first manifested by outer hair cell destruction. Drugs, noises, and aging all affect outer hair cells before influencing other structures. To take just one example, humans' loss of outer hair cells begins at a very early age and tends to progress inexorably. The average sixty-five-year-old male has lost most of his outer hair cells from the higher-frequency half of the cochlea.
This sort of damage may be more pernicious than the mere removal of the neural connections would suggest. It implies a profoundly altered mechanical state in the basal cochlea and the corresponding abnormal processing of sounds by the inner hair cells. Hair cell dysfunction may be the most significant contributor to the hearing disorders that afflict some eighteen million Americans. Our understanding of these marvelous information-transmitting and (likely) motor devices yields not only intellectual rewards but also significant implications for medicine and national health care.
This article originally appeared in the June 1986 issue of the World & I magazine and is reprinted by permission.
Ed. note: It is now estimated that more than 28 million Americans are hearing impaired. Dr. Dallos continues his research on cochlear development and function.
Author
Peter Dallos, PhD
John Evans Professor of Neuroscience; Hugh Knowles Professor of Audiology
Departments of Communication Sciences and Disorders, Neurobiology and Physiology, Otolaryngology and Biomedical Engineering
Northwestern University
|
