Melodic Contour Identification Training in Cochlear Implant Users with and without a Competing Instrument

Abstract

For cochlear implant (CI) patients, music perception can be especially difficult when multiple instruments are played. The relatively poor spectro-temporal resolution does not allow CI patients access to pitch and timbre cues that may be used to segregate and stream competing melodies and instruments. Auditory training improves CI users' melodic pitch perception. Given the more difficult and more common listening condition of multi-instrument music, it may be more beneficial to train with multiple instruments than with a single instrument. In this study, CI subjects were trained to identify melodic contours. One group (“no masker”) trained while listening only to the target contours. The other group (“masker”) trained while listening to target contours presented with a competing masker. Before training, baseline melodic contour identification was measured with and without a masker; the timing, pitch, and timbre of the masker was systematically varied. Subjects trained at home for a total of 10 hours during the 1-month training period. Results showed that baseline performance was poorer with the masker than without, and that performance improved with both training methods. However, the magnitude of improvement was greater for the masker group, suggesting that the more difficult training provided better outcomes.