Discrimination of Complex Speech-Related Signals with a Multichannel Electronic Cochlear Implant as Measured by Adaptive Procedures

Major emphasis has been placed on identifying speech, with and without lipreading, after cochlear implantation. Although this is pragmatically important, identification measures provide limited information as to device and patient performance along acoustic dimensional continua known to underlie the phonetic features that differentiate one phoneme from another. We have undertaken a series of discrimination studies for a patient implanted with the Nucleus multichannel prosthesis, using synthetic, speechlike stimuli and other complex signals that incorporate acoustic changes important to speech perception. Measures of temporal resolution, transition discrimination, and second formant difference limens were made using adaptive procedures with feedback. All signals were presented free field to assess the complete prosthesis in relation to patient performance. Similar measures were also obtained for a group of normal-hearing subjects for comparative purposes.