Beginning in the late 1950s, researchers began wondering whether it might be possible to replace the electric signals from the missing hair cells in people with sensorineural hearing loss, especially the majority of those people who had intact auditory nerves. The researchers' effort to create a cochlear implant faced a great deal of skepticism and daunting technical obstacles. But they were fortunate to be starting at a time when a good deal was known about the electric signals produced by the organ of Corti and sent down the auditory nerve. From the work of Eberhardt Zwicker, Smith Stanley Stevens, and Gordon Flottorp which culminated in 1957 at Harvard University, researchers knew that the auditory system was able to organize sounds into 24 channels. From animal experiments by Hallowell Davis and Robert Galambos, also at Harvard, it was known that the organ of Corti and the auditory nerve were at the base of this organization and that fibers in one part of the nerve carry information about low tones, the fibers in the next part carry information about slightly higher tones, and so on, in a predictable fashion.

The early experimental implants, however, did not exploit the cochlea's tonotopic organization. Several different research groups started implanting single-channel electrodes into the cochleas of deaf volunteers. The researchers and the volunteers knew that these crude devices would not provide enough information to encode speech. They thought, based on work that had been done by Glenn Wever and C. W. Bray at Princeton University in the 1930s, that the timing of electrical discharges from the electrodes would allow the volunteers to determine the pitch of a sound. Indeed, the volunteers were able to extract a huge amount of auditory information from the single channel. Although their speech perception was poor, they could tell, for example, whether a spoken word had one syllable or two, and they had some sensation of the pitch of a sound by the timing of neural spikes; this was enough to serve as a substantial aid to lipreading.

That surprising success emboldened researchers. By the early 1970s several groups were at work on more sophisticated devices with multiple electrodes. But how many electrodes would they need? The auditory nerve contains 30,000 fibers. Would the researchers have to provide 30,000 electrodes to stimulate all the nerve fibers individually in order to simulate intelligible speech sounds? If so, the project would clearly be impractical. But according to Zwicker and his co-workers, 24 channels were sufficient. In addition, Michael Merzenich of the University of California, San Francisco, simplified the system even further after he uncovered research results from an unexpected source.

Bell Laboratories, which was then the research arm of AT&T, was concerned with how much information needed to be sent over telephone lines to re-create intelligible speech sounds. Bell scientist James Flanagan, who is now at Rutgers University, determined that the frequencies of speech could be divided into as few as six or seven channels and still be understood. Michael Merzenich and others reasoned that, if only six or seven channels were needed to transmit speech over telephone lines, the same number of electrodes would likely suffice in a cochlear implant.

Would such an implant be safe? Many physicians and researchers thought that a cochlear implant with multiple electrodes would be like putting a telephone pole into the chambers of the inner ear and that it would probably destroy the delicate ganglion cells that transfer signals from hair cells to the brain through the auditory nerve. In a series of animal experiments, however, Merzenich and his colleagues proved that the implant did not harm the ganglion cells. In fact, the cells were reinvigorated by the stimulation.