Encoding Temporal Fine Structure for Cochlear Implants

The study goal is to improve music and speech perception for cochlear implant users. The relevant health outcome is their quality of life. This proposal focuses on how well cochlear implant users can learn to use temporal fine structure if provided as a clear and consistent cue for music or voice pitch. Historically, cochlear implants have discarded temporal fine structure and have only transmitted timing information of relatively slow envelope fluctuations. Attempts have been made to restore temporal fine structure into cochlear implant stimulation, but it is unclear whether previous attempts were limited by implementation, lack of experience, or inherently by physiology. The proposed approach is unique in that it examines the perceptual and physiological plasticity that occurs when temporal fine structure is restored. Proposed research is organized into two aims, which examine the relative salience of stimulation place and rate for providing a sense of pitch (Aim 1) and the salience of dynamic-rate stimulation compared to conventional methods (Aim 2). Both aims combine perceptual learning, computer-controlled electrode psychophysics, electrophysiology, and computational neural modeling to characterize the plasticity of pitch perception in cochlear implant users.

Aim 1 examines the perceptual and physiological plasticity associated with place and rate of cochlear implant stimulation. Cochlear implant users hear an increasing pitch associated with increasing stimulation rate, but this effect is difficult to measure above 300 Hz. Most studies of psychophysical sensitivity to cochlear implant stimulation rate have not considered perceptual learning. Preliminary results show that the sense of pitch provided by stimulation rate improves with training. The proposed research examines perceptual sensitivity and physiological encoding throughout a crossover training study with training provided for pitch based on place and rate of stimulation. The primary hypothesis tested is that cochlear implant users have a latent ability to hear pitch associated with stimulation rate, but they require training to learn how to use this new information.

Aim 2 is to determine whether dynamic-rate stimulation provides better sensitivity and better physiological encoding of fundamental frequency compared to conventional stimulation methods based on amplitude modulation of constant-rate stimulation. In normal physiology, auditory-nerve activity phase locks to the temporal fine structure of sound. Since cochlear implants typically discard this information, it is unknown how well cochlear implant users can learn to use it if provided. Aim 2 focuses on the comparison between dynamic-rate stimulation in which stimulation rate is dynamically adjusted to convey temporal fine structure compared to conventional methods based on amplitude modulation of constant-rate stimulation. The primary hypothesis is that dynamic-rate stimulation provides better pitch sensitivity and better physiological encoding compared to amplitude modulation of constant-rate stimulation.