Music perception in bimodal cochlear implant users

Abstract

While most cochlear implant users can perceive speech signals in quiet conditions very well, the perception of music is reported to be poor for most of them. In this thesis the perception of music in cochlear implant users was the main focus of the research. It has been found that most cochlear implant users cannot perceive pitches of melodies which are very important to music perception. Another limitation in their perception of music was in the recognition of musical instruments or more generally the timbre of sounds. It is widely believed that the current technology of cochlear implants is suitable only for coding of speech signals and not music signals. There is a need to investigate the reasons for such poor perception of music. The poor perception of pitch is believed to be the main reason for poor music perception. A literature review showed that although pitch perception is not satisfactory in cochlear implant users, there is also a possibility of interference between the perception of pitch and other aspects like intensity and quality of sounds. Therefore in the rest of the research, the effects of these two aspects on the perception of pitch were investigated. Three mechanisms were identified by which intensity could influence the perception of pitch (Current spreading, Electrode activation spreading, and Spectral spreading). Each mechanism was tested by inputting a specific stimulus type to the cochlear implant sound processor at two different intensity levels. Twelve bimodal cochlear implant users were the participants in this part of the research. The perceived pitch was quantified through a selected matched frequency value in the non-implanted ear (bimodal pitch matching). The results showed that the effect of intensity was observed when current spreading happened. Another finding of this research was that the perceived pitches due to stimulating different electrodes of the cochlear implants were lower than Greenwood’s prediction. In addition, when spectral spreading was present in complex musical notes, the perceived pitch for low frequency was not significantly different from that of high frequency. A wide range of frequency created a narrow range of different pitches in the cochlear implant. There were large individual differences among people. Some participants perceived lower pitch at higher intensity while others perceived higher pitch at higher intensity. To test the effect of sound quality on perceived pitch all of the above mechanisms were tested twice: once with a pure tone as the matching sound in the non-implanted ear and again with a complex tone in the non-implanted ear. The matched frequencies using complex tones were significantly lower than their counterparts with pure tones. This showed that the type of sound had a significant effect on the perception of pitch. In addition when pitch matching was done in one ear (monaural pitch matching) and the types of both sounds in matching were completely similar, the participants could match pitch with more precision. This was another indication that sound quality had an effect on the perception of pitch. Monaural pitch matching in the implanted ear showed that the effect of intensity in the implanted ear was greater in comparison with monaural pitch matching in the non-implanted ear. An earlier research study conducted by the author on a star performer indicated that current CIs can provide enough information for pitch perception of sounds from a single instrument when the listener is very familiar with that instrument. Bimodal cochlear implant users may experience different pitch percepts in implanted and non-implanted ears. The implication of this research is that a larger number of electrodes with narrower analysis filters in the sound processor could provide better fine structure information and improve pitch perception. The results suggest that more restricted current spreading could improve pitch perception as well.