Exploring and Improving Hearing by Cochlear Implant listeners

Cochlear implants (CIs) restore hearing to severely and profoundly deaf people by electrically stimulating the auditory nerve. Many CI patients understand speech well in quiet surroundings, but all struggle to hear well in noisy situations. In addition, the perception of pitch is usually very poor and this greatly reduces the enjoyment of music. Because normal-hearing listeners use differences in pitch between sounds to tell them apart this also contributes the difficultiess CI listeners experience when many people are talking at the same time. Our research proposal investigates ways of alleviating these problems.

One strand of our approach aims to produce selective activation of the auditory nerve whilst still producing a sound sensation that is loud enough. This is important for two reasons. First, information from each frequency band of speech is sent to one electrode, with the aim of stimulating just a few neurons close to it. Unfortunately the electrical current spreads broadly to other electrodes, thereby smearing the neural response to the sound. Second, current sometimes spreads outside of the cochlea and stimulates the facial nerve, causing unpleasant twitching that can prevent the patient from using their CI. We have designed new ways of stimulating the electrodes that we hope will solve these problems, and will test them with CI patients. To better understand our results we compare them to the predictions of a computer model of the auditory system, and, in turn, use the experimental results to improve the model. We are particularly interested in how the health of the auditory nerve, which degenerates following deafness, influences the effectiveness of methods - including our own - that are designed to produce selective stimulation and improve speech perception. To do so we include measurements of two particular groups of patients. One of these are those who have become suddenly deaf in a way that is believed to leave the auditory nerve intact, and we compare them to long-term-deaf users. The other consists of children with a condition known to affect the auditory nerve, with recent evidence that it may particularly affect neurons that innervate the apex of the cochlea, which normally responds to low-frequency sounds.

A second strand focusses on the poor pitch perception by CI users. Some manufacturers have tried to improve pitch perception by presenting fine timing information to a subset of the electrodes, in the cochlear apex, as part of the speech-processing strategy (which converts sound to a pattern of electrical impulses). Unfortunately, very little is known about how CI listeners actually process this information, and this is the subject of the first part of this strand. These methods usually present different patterns to each electrode, and we suspect that pitch perception would be better with the same pattern applied to all of these apical electrodes. If our first experiments show that this is indeed the case, we will implement and test new speech-processing strategies which we hope will improve pitch perception while still clearly conveying all the other information that is needed for good speech perception. Finally, we use electrophysiological methods to help understand the neural basis for poor perception by CI listeners, especially that occurring when the pitch is quite high