Functional magnetic resonance imaging (fMRI) of a real-time cochlear implant acoustic simulation and auditory modeling of the medial olivocochlear efferent system

Cochlear implants (CIs) can be utilized in profoundly deaf individuals to restore some sensation of hearing. Individuals who undergo surgery to receive a cochlear implant must subsequently adapt to an input neural signal that is distorted and acoustically underspecified. In general, performance is not restored to those levels achieved prior to the onset of deafness, especially cases of speech recognition in background noise. The goal of this thesis is to improve our knowledge of the auditory system, producing information that can be used to optimize cochlear implants for clinical benefit. A portable real-time speech processor that implements an acoustic simulation model of a CI was developed for the iPhone to examine the nature of perceptual learning by assessing how speech intelligibility improves after training. Participants\u27 underwent fMRI before and after exposure to the processor, measuring changes in neural activation following perceptual learning under real-time CI simulation. A second project modeled the time-varying, frequency, and level dependent feedback of the medial olivocochlear reflex (MOCR), which is known to be important for speech recognition in noise. This system was implemented within the framework of a well-established computational model for normal and impaired hearing auditory nerve responses based on physiological evidence. The developed auditory nerve model is the first of its kind to be binaural, and includes feedback for both the crossed and uncrossed ipsilateral and contralateral pathways. Applications of this model, which include the efferent feedback loop, may contribute to our understanding of sensorineural hearing loss in noisy situations for hearing-impaired listeners