Abstract
Considering the recognition performance obtained by an implanted patient, the authors have developed models to explain the decrease in performance when the number of open channels on the prosthesis is increased. The French cochlear implant Chorimac was used in this experiment. Two models have been developed. The first is monodimensional and the second is multidimensional. They respectively represent an increase in information and its superposition. Results suggest that for the patient, the superposition factor prevails and is detrimental to recognition. Its elimination should be a major goal. A good selection of electrodes in a relatively small number seems to be the best policy. This is already done in some cochlear implants. Some other parameters in the signal that seem worth being analyzed are introduced.
Get full access to this article
View all access options for this article.