New temporal coding scheme for auditory nerve stimulation

Abstract

Electrical stimulation of the auditory nerve provides deaf people with a perception of sound inferior to that of normal hearing people. Since any acoustic perception is derived entirely from the output of the auditory nerve, effective cochlear implant stimulation strategies should evoke an output of the electrically stimulated auditory nerve that is in some sense a good approximation to that of an acoustically stimulated auditory nerve. A typical cochlear implant may have around 20 independent electrodes, each of which transmits data in the form of discrete electrical pulses at a practical limit of a few thousand pulses per second. In contrast, the normal hearing ear has approximately 30,000 nerve fibers, each of which can achieve action potential independently of the others at any time during the stimulus. Clearly, it is beyond the scope of current technology to achieve an electrically evoked auditory nerve output identical to that evoked via the normal hearing situation (NHS), so cochlear stimulation strategies must be limited to approximating certain features of neural firing patterns. How should such features be chosen?