Effects of Hearing Experience, Music Training and Spectral-Temporal Processing on Music and Speech Perception in Adult Cochlear Implant Users

Abstract

Cochlear implants (CIs) are a very successful hearing device for improving speech perception in quiet environments for individuals with severe-to-profound hearing loss. However, many CI users continue to face challenges in appreciating and enjoying music (particularly due to poor melody perception) and they also struggle to understand speech in noisy settings. This thesis addresses some of the gaps in our understanding of the factors influencing music perception and enjoyment among CI users, exploring the mechanisms behind chord discrimination abilities and their relation to speech-in-noise perception. The first study, experiment 1 (chapter 2) used questionnaires to collect details of an individuals’ background characteristics for hearing experiences and music-listening experiences and behaviours and explored how they contributed to CI users’ abilities to discriminate between musical chords using simulated piano tone stimuli. Exploratory factor analysis was used to reduce the data fields to a series of meaningful constructs; five latent variables and a direct measurement were derived (through the combination of associated variables). These were: ‘duration of hearing difficulties’, ‘age characteristics’, ‘years of CI experience’, ‘music quality of the implant’, ‘music engagement’ and ‘music listening habits’. A significant interaction between some of these variables revealed that older participants with a greater number of years of CI experience performed better in chord discrimination tasks, potentially due to enhanced auditory cortex function and neuroplasticity associated with extensive listening experience. This experiment was conducted online. A key finding from experiment 1 was that participants performed better when the differences between the chords involved lowering the second note (reduced chords), where notes are closely spaced, compared to raising the third note (expanded chords). The closely spaced notes in the reduced chords were likely to provide a higher number of temporal interactions , such as beats, which CI users may be able to detect due to their reliance on temporal envelope cues. Further investigation of these ideas were carried out in experiment 2 (chapter 3), where a hierarchy of frequency discrimination tasks, measuring frequency difference limens, were used (pure tones, pair of tones, amplitude-modulated tones) and findings were associated with chord (pure-tone) discrimination accuracy. For the chord task, it was found that as the difference between the chords increased in semitones that discrimination was easier (also a finding in experiment 1). In experiment 2, two different base frequencies were used for the chords. There was a lower frequency stimulus, with the root being middle C (C4), notes in this stimulus set covered the range 262-466 Hz. The higher frequency stimulus set was based around a root note of C6 and the stimuli covered a frequency range of 1046-1864 Hz. It was speculated in experiment 2 that the hierachy of tasks would result in increasingly higher thresholds due to the increase in complexity of the stimuli, unless for the pair of tone stimuli that the listeners benefitted from the temporal interactions, in which case pair of tone stimuli may result in better (lower) thresholds. This was indeed what was found, and it was more apparent for the lower frequency tones. This suggests that closer frequency spacing may enhance temporal interaction cues in the low-frequency range, accessible to CI users. These results align with auditory scene analysis theory and reflect auditory neuroplasticity, where limited access to temporal fine structure and spectral fine structure cues may lead to CI users depending more on temporal envelope cues. Better utilization of these temporal cues was associated with improved chord discrimination skills and enhanced speech-in-noise perception. The role of residual hearing was also highlighted, where individuals using Cochlear devices with more extensive low-frequency residual hearing exhibited better chord discrimination than those who did not have residual hearing. This difference was only apparent for the Cochlear users who had a wider range of performance for those without residual hearing than for the other devices. This analysis of the interaction between residual hearing and device type was based on a low sample size by the time the participants were separated into smaller subgroups, however, it does highlight that residual hearing can provide useful additional sound processing that can be used for optimizing and enhancing music perception. Suggesting the importance of preserving the hearing and clinically considering the hearing during mapping and rehabilitation sessions. Residual hearing and individual CI settings were further explored in Chapter 4, comparing participants who completed both the online and in- person chord discrimination tasks. Although there were only eleven participants, significant differences emerged between online and in-person performances, with better results in the in-person (pure tones) setting. Additionally, an in-depth exploration of three participants’ CI mapping settings showed that those with an additional electrode for the top note of the expanded chords performed better on those chords. A qualitative case study of a deafblind musician using CIs illustrated adaptive strategies such as tactile feedback and personalized rehabilitation approaches to overcome sensory limitations. This real-world application emphasizes the significance of individualized strategies in enhancing music enjoyment and quality of life for CI users. Overall, these experiments advance our understanding of music perception and enjoyment in CI users by exploring the mechanisms underlying chord discrimination abilities. The findings highlight the critical roles of temporal cues, residual hearing, and personalized approaches, offering valuable insights for improving CI technology, programming, and rehabilitation programs to better support music perception and enjoyment among CI users.