Summary
During the last 20 years, considerable advances have been made in the development of cochlear implants for the profoundly deaf. It has been shown that multiple-channel devices are superior to single-channel systems. Strategies in which several electrodes (six to eight) correspond to fixed-filter outputs, or the extraction of six to eight spectral maxima for 20 to 22 electrodes offer better speech perception than stimulation with second and first formants at individual sites in the cochlea, provided that nonsimultaneous or interleaved presentation is employed to minimize current leakage between the electrodes. Further refinements such as spectral maxima at rates of approximately 800 to 1600 pulsess and the extraction of speech transients also give improvements for a number of patients.
Successful speech recognition by many prelinguistically deafened children as well as by postlinguistically deaf children has been achieved. If children are implanted before 2 years of age and have good language training, they can achieve speech perception, speech production, and expressive and receptive language at levels that are normal for their chronological age. The main restriction on the amount of information that can be presented to the auditory nervous system is the electroneural “bottleneck” caused by the relatively small number of electrodes (presently 22) that can be inserted into the cochlea and the limited dynamic range of effective stimulation. Strategies to overcome this restriction continue to be developed.