A trend in cochlear implants is the use of faster per-channel pulse carrier rates, promoted on the basis of increased information capacity. However, it is not clear that higher rates achieve this goal in most or ill individuals. Clinical studies present mixed results and research promoting high (e.g., 5000 pps) rales raises important questions. For example, while beneficial, desynchronizing, effects have been reported in many auditory nerve fibers, comparable numbers become adapted to the point of being unresponsive. Furthermore, physiologic dala have only been obtained from intact fibers, not from degenerated neurons typical of chronically deaf ears. Finally, while much is known about neural adaptation to acoustic stimuli, relatively little is known about electrical adaptation, even though the latter typically produces much larger functional changes. We hypothesize that at least part of the variability in performance with higher-rate carriers is due to across-user differences in the auditory nerve's response to high-rate stimuli. This research plan seeks to fill these gaps in our knowledge of how the auditory nerve encodes information presented as modulated pulse trains.
Three Aims are proposed.
Aim 1 will assess signal encoding in fibers excited by modulated carriers at rates relevant to modern and proposed speech processors. Data will be collected from intact and degenerated nerves for greater applicability to clinical cases. Fiber tracing techniques will help link physiology with anatomical status.
Aim 2 will use that data to help develop a computational model of the nerve that accounts for many fiber properties (e.g., integration, refractoriness, and adaptation). This model wsll be used to predict the electrically evoked compound action potential (EC'AP) so that we can tesi: the capacity of ECAP measures to assess fiber functionality, a clinically relevant issue. Finally, Aim 3 will explore the feasibility of applying specific transforms to modulated trains to compensate for adaptation and refractory effects that limit information carried by modulated pulse trains. We expect the results of this work will guide future designs of cochlear-implant speech processors so that modulated stimuli can be better tailored to the encoding capacity of the user's auditory nerve.
