I Can Hear You Whisper Read online

  In his efforts to develop his own theory, Volta experimented with metals alone. He stacked pairs of silver and zinc disks separated by brine-soaked pads. When he touched the top and bottom of the pile with a wire, an electric current flowed through the pile and along the wire. The contraption became known as the voltaic pile; Volta had invented the battery.

  To explore the idea of using electricity medically, Volta applied his voltaic pile to the body. First, he made muscles contract. Then, connecting his battery to the optic nerve, he generated a flash of light when he touched any part of his face. Next he turned to hearing. Into his own ears he inserted two metal rods connected to a circuit of thirty or forty cells with about fifty volts of power. Electricity crackled through him, and he later described the boom he experienced:

  “I received a shock in the head and some moments after I began to hear a sound, or rather noise in the ears, which I cannot well define; it was kind of a crackling with shocks, as if some paste or tenacious matter had been boiling.”

  The scientific world was intrigued yet cautious. Few were willing to repeat the experiment on themselves. The connection between electricity and hearing had been made—literally—but for the next century and a half, there was little progress.

  Then in Paris, in the 1950s, electrical hearing became reality when “the impossible” was tried. André Djourno was a neurophysiologist who studied medical applications of electricity at the Institut Prophylactique (today called l’Institut Arthur Vernes). Working with rabbits and guinea pigs, he was stimulating nerves by implanting induction coils. Charles Eyriès was the chief of the hospital’s head and neck surgery department and an expert in facial nerve repair. In February 1957, a fifty-seven-year-old patient, Monsieur G., came to Eyriès in fairly bad shape. Surgery to remove two large cholesteatomas, a skin growth that pushes from the middle ear into the inner ear, had left the man deaf in both ears and with extensive facial nerve paralysis. Eyriès wanted to reanimate the facial nerve with a graft. A colleague proposed that perhaps the man’s deafness could be addressed as well if Eyriès implanted one of Djourno’s induction coils during the surgery. The patient, Eyriès wrote later, had “expressed the desire that the impossible be tried in order to put an end, however imperfect, to his total deafness.” Since he would be undergoing surgery anyway, it was thought he had nothing to lose. For his part, Djourno was fascinated by the opportunity.

  The facial graft repair (using fetal tissue) was a success, but Eyriès and Djourno found the cochlear nerve “significantly shredded.” They put the active electrode into the remaining stump of the nerve and placed the induction coil in the temporalis muscle. During the operation, they tested the device with a variety of stimuli: bursts of low-frequency current at a rate of fifteen to twenty pulses a minute, then low-frequency alternating current, and also words spoken into a microphone. From the start, Monsieur G. heard sounds. He could discriminate the loudness or softness of sounds but not their pitch. Speech was unintelligible. During extensive rehabilitation in the following months, increasingly complex signals were tried. Eventually, the man was able to tell the difference between low frequencies, which sounded to him like “burlap tearing,” and high frequencies (“silk ripping”). He could hear some environmental noises and a handful of words, but he never understood speech. Within a few months, the implant stopped working. Eyriès and Djourno found that the electrode in the muscle had broken; a second implant also failed. Eyriès washed his hands of the project. Djourno tried again with a different surgeon and a new patient, but the young woman was less enthusiastic and Djourno’s funding ran out.

  Eyriès and Djourno reported on their work in several French medical journals. It received little attention in other countries, although at least one researcher suggested to me that they deserve the credit for inventing the cochlear implant. Their successful surgery was, however, mentioned in a short article in an English-language publication that was seen by a patient in California. He clipped out the article and brought it to his otologist, who happened to be Bill House.

  “The light went on,” House told me as we sat in his small living room in Oregon. A few years earlier, he had moved up from California to live next door to his son, David. “See, we’d known that putting electricity near the ear you get a sound, put it across the eye and you get a flash of light. So the nervous system is very highly attuned to telling you what’s happening. [The French report marked] the first time I realized a patient had a total loss of the cochlea and could still hear with electrical stimulation.”

  How could it be possible to hear with a nonfunctioning cochlea? The cochlea is the hub, the O’Hare Airport, of normal hearing, where sound arrives, changes form, and travels out again. When acoustic energy is naturally translated into electrical signals, it produces patterns of activity in the thirty thousand fibers of the auditory nerve that the brain ultimately interprets as sound. The more complex the sound, the more complex the pattern of activity. Hearing aids depend on the cochlea. They amplify sound and carry it through the ear to the brain, but only if enough functioning hair cells in the cochlea can transmit the sound to the auditory nerve. Most people with profound deafness have lost that capability. The big idea behind a cochlear implant is to fly direct, to bypass a damaged cochlea and deliver sound—in the form of an electrical signal—to the auditory nerve itself. “The inner ear is a pretty beautiful natural platform for stimulation in the sense that from very early childhood, it’s in a stable adult size and form,” says auditory neuroscientist Michael Merzenich, who was instrumental in a later stage of cochlear implant development in the 1970s. “Several surgeons got at the idea that conceivably you could excite it and recover enough hearing to be useful.”

  To do that would be like bolting a makeshift cochlea to the head and somehow extending its reach deep inside. A device that could replicate the work done by the inner ear and create electrical hearing instead of acoustic hearing would require three basic elements: a microphone to collect sound; a package of electronics that could process that sound into electrical signals (a “processor”); and an array of electrodes to conduct the signal to the auditory nerve. How best to build those pieces was anyone’s guess. Some of it at least seemed achievable with time. Electronics could be engineered, for instance, and tolerable levels of stimulation for the tissues involved could be determined through animal studies. More difficult was the question of how to excite discrete groups of nerve fibers. Even if those technical problems were solved, and electrodes successfully and safely implanted, a basic science problem remained, to which no one in the 1960s had an answer: what signal to send.

  The processor had to encode the sound it received into an electrical message the brain could understand; it had to send instructions, and no one knew what those instructions should say. They could, frankly, have been in Morse code—an idea some researchers considered, since dots and dashes would be straightforward to program and constituted a language people had proven they could learn. By comparison, capturing the nuance and complexity of spoken language in an artificial set of instructions was like leaping straight from the telegraph to the Internet era. It was such a daunting task that realistically, most scientists thought the best they could hope for was to make speechreading easier. “The more a researcher knew about auditory neurophysiology or speech acoustics, the more confident he was that implants could not provide a high (or even useful) level of speech understanding,” wrote Michael Dorman and Blake Wilson in an account of some of the early research. The few who “imagined that you could just replace the signals in the ear in some magical way,” says Merzenich, didn’t really know much about “how the complexities of sounds that would be meaningful, like the sound of oral speech, had to be represented across the nerve to the brain.”

  No one was sure what, exactly, the brain needed to hear to distinguish between a dog barking and a baby crying, or to know to get out of the way when a car horn blows. They doubted it would ever be possible to make an implant that allowed a
child to hear his mother say “I love you.” Before sound could fly direct to the auditory nerve, someone would have to reinvent the airplane.

  Bill House aimed to try. In his first year of private practice in Los Angeles, House saw two families with two-year-old children they suspected were deaf. At the time, there was no test to uncover hearing loss at earlier ages. House found it painful to tell parents their children were deaf. “I felt I was presenting a very bleak outlook to these parents,” he said. “What I had to offer seemed very inadequate.” When he learned about the work of Djourno and Eyriès, he immediately saw the potential to do more and resolved to pick up where the French had left off. “I felt if there was anything we could do, we should.”

  It was an attitude he learned from his father. House grew up on a five-acre ranch in Whittier, California, not far from Los Angeles. Although they kept a few cows and grew avocados, oranges, and lemons, all of which were the responsibility of Bill and his brothers, the family business was dentistry. His father, Milus, set up a private practice in an old barn. Milus House wasn’t the kind of father who played ball with his boys or took them fishing, but he made a big impression when he talked about the satisfaction he gained from fixing serious dental problems that affected patients’ emotional and physical well-being. “I could feel the joy he had as he talked about ‘fixing mankind,’” wrote Bill House years later. “I knew then that I too wanted to be a ‘healer.’” After two years as a dentist in the Navy, House went to medical school to specialize in ear, nose, and throat surgery, then narrowed that down to ear surgery, or otology. In 1956, he joined his half-brother Howard, who was ten years older, in practice at what became known as the House Ear Institute, a leading West Coast center for otolaryngology then and now. (Today it’s called the House Research Institute and is run by Howard’s son and Bill’s nephew, John House.)

  Bill House told me he had been a mediocre student. Later in life, he realized he probably had dyslexia; writing and reading were always a challenge. But he excelled at working with his hands—a skill his dental training helped hone—and became a top-notch surgeon. He was also driven to solve problems. He spent hours in the morgue, practicing surgical techniques and new approaches on unclaimed bodies. His wife, June, was a registered nurse. If she could get a babysitter, she joined him in the morgue to hand over his instruments.

  One of House’s first innovations was to introduce and improve upon the use of a surgical microscope for ear surgery. He developed new surgical techniques for acoustic neuromas that he says helped reduce mortality from 40 percent to less than 1 percent and preserved the facial nerve. For patients with Ménière’s disease, a disease of the inner ear whose symptoms include vertigo, vomiting, tinnitus, and hearing loss, he created a small device, a shunt, to correct the condition. His most famous patient was astronaut Alan Shepard, who developed Ménière’s after his first space flight. With one of House’s shunts in place, Shepard went back into space as commander on Apollo 14. A grateful Shepard invited House and his wife to Cape Canaveral for the liftoff.

  When House got interested in cochlear implants, he enlisted the help of electrical engineer Jim Doyle. As a first step, Doyle built a battery-operated amplifier and electrode that could be applied to the round window, the membrane that leads to the inner ear. House tried it out during middle ear surgery with three volunteer patients who had lost their hearing after developing speech. Under local anesthesia, he lifted the eardrum and delivered small alternating currents to the round window. All of the patients heard sounds; furthermore, the sounds seemed related to the frequency and intensity of the electrical stimulus. Excited, House had Doyle set to work creating an implantable version. The new device had a silicone-covered coil to generate current, amplifiers, and an electrode, but nothing as fancy as a speech processing program. The coil would be placed in the mastoid bone and the electrode in the cochlea. Wires running from the coil ended in a plug in the skin behind the ear. In the lab, patients would be connected—plugged in—to an electronic stimulator that could send signals through the system to the auditory nerve. When not in use, the plug behind the ear would be covered with a bandage.

  Two adult patients volunteered, and they were implanted early in 1961. For two weeks, they underwent testing with a variety of sounds. “They could hear the sounds, and it was obvious that even though the sounds were not clear, the devices would be of great help for environmental warning sounds and lipreading,” says House. Soon, however, redness and swelling appeared around the external wires. “I got kind of scared at that, and I had to take it out,” says House. “They were disappointed and so was I.” He had run up against the problem of biocompatibility.

  Even in the 1960s, when research protocols did not preclude putting an untested device straight into human subjects, many researchers considered House’s approach at best unscientific and at worst dangerous. Certainly, it was emblematic of the strengths and weaknesses of House’s hands-on style. “He started building things and putting them in patients,” says Eisen. “He got his information from talking to people, not reading books. You’d get in big trouble if you did what he did today, but … if he’d read more books, he might have believed all the people who told him it couldn’t be done.”

  Biocompatibility was not the only problem. Doyle, the engineer, had made grand claims of eliminating deafness, which caught the imagination of technology buffs and science-fiction fans: “Electronic Firm Restores Hearing with Transistorized System in Ear,” read a typical headline in Space Age News. Doyle also saw commercial potential and set about creating a company around the device, a move that was at odds with House’s more altruistic vision. The ensuing publicity was too much too soon. It offended other doctors and scientists and led to a deluge of calls from people seeking help whom House had to send away. According to House, he told Jim Doyle he was going to put the project on hold until they had biocompatible materials and asked for a full report on the materials and electronics. Doyle refused. “He said I was a damn fool, and he was going to get a Nobel Prize,” House says, still visibly upset by it fifty years later. “It was my first disappointment of many that followed.”

  • • •

  News of what House had done reached an auditory researcher named Blair Simmons, who was a new assistant professor of otolaryngology at Stanford University School of Medicine. Simmons was interested in the physiology of sound reception. His studies of the auditory systems of cats had been published in the prestigious journal Science. A scientist’s scientist, Simmons was angry at what he regarded as “irresponsible claims” emanating from Los Angeles, but he was captivated by what he considered a true research problem.

  A year later, in 1962, Simmons saw an opportunity. An eighteen-year-old cancer patient with increasing hearing loss was going to have exploratory brain surgery under local anesthesia. As with House’s patients, the young man’s inner ear would be exposed during the operation. In his work with cats, Simmons had successfully implanted electrodes into the inner ear without destroying the auditory nerve. Now he wanted to try his technique in a human subject who would be able to describe what he heard. “We were amazingly lucky on our first try,” Simmons later wrote. With an electrode stimulating his auditory nerve, the teenager heard a wide variety of sounds. Most surprising was the boy’s ability to hear sounds at either end of the spectrum beyond the range required for speech.

  Two years later, Simmons went a little further. Sixty-year-old Anthony Vierra of San Jose suffered from retinitis pigmentosa, a condition that causes an increasing loss of peripheral vision. Eventually, the tunnel vision narrows completely and the patient is left blind. By the time he met Simmons, Vierra was also profoundly deaf in his right ear and was losing what hearing remained in his left. He gamely agreed to be permanently implanted with a six-electrode cochlear implant that Simmons had devised. Like the House implant, this one had wires threading through the skull, just behind the ear, that had to be connected to a computer or electronic stimulator in the laborator
y for Vierra to hear anything. The surgery was performed at Stanford, but even a respected researcher like Simmons had trouble finding basic-science colleagues willing to work with him on testing such a controversial project. So he turned to one of the few places in the country with an established interest in the alchemy of electricity and human speech: Four scientists at Bell Labs, which had moved from Manhattan to New Jersey, agreed to perform the audiological testing of Vierra. “They were outsiders,” Simmons commented later. “I don’t think they’d read the publicity.” Vierra had never been on a plane before, and since he was nearly blind, Simmons and his wife escorted him across the country.

  The combination of Vierra’s poor vision and hearing meant the researchers had to communicate with big block-letter signs saying things like: TELL US WHEN YOU THINK YOU HEAR SOMETHING. Vierra needed some lessons in how to accurately describe and compare the sounds he was hearing, but he ultimately was able to identify them in terms anyone could understand. At a slow rate of one pulse per second, he heard a ping or a ding. Three to four pulses per second resulted in clicks. As the rate increased, he heard a buzz, then a sound like a telephone ringing, and finally a car horn above thirty pulses per second. Vierra was able to recognize familiar tunes like “Jingle Bells” and “Mary Had a Little Lamb.” In the eighteen months he wore the device, however, he never could understand speech. In a Science article on the work coauthored by Simmons, his Stanford colleague John Epley (who collaborated on the surgery), and the Bell Labs researchers, the conclusions were cautious. They had succeeded in expanding technical knowledge about pitch perception, although in a qualified way. As to the larger goal, they wrote: “Much remains uncertain… . It is unlikely that stimulation with any speech-derived signal would permit this subject to discriminate an appreciable number of words, unless considerable learning were possible.”