Effect of number of channels and speech coding strategy on speech recognition in mid-scala electrode recipients

Abstract

This study investigated the number of channels needed for maximum speech understanding and sound quality in 5 (anticipated n = 10) adult cochlear implant (CI) recipients with mid-scala electrode arrays completely within scala tympani (ST). CI programs with 4–16 active electrodes using a CIS-based strategy were created along with two n-of-m programs (8-of-16 and 12-of-16). Measures of speech understanding and sound quality were assessed. Our hypothesis was that individuals with precurved electrodes localized in ST would have greater channel independence than previous studies demonstrating gains in performance beyond 8 channels. Participants demonstrated performance gains up to 8–10 electrodes for speech recognition and sound quality ratings. Significantly poorer performance was observed using an n-of-m strategy as compared to CIS conditions with 8+ electrodes. These data are in contrast with recent studies for precurved arrays (e.g., Croghan et al., 2017; Berg et al., in review) reporting significant improvement up to 16 to 22 channels. However, the current results are consistent with previous literature (e.g., Fishman et al., 1997; Friesen et al., 2001; Shannon et al., 2011), demonstrating no more than 8–10 independent channels for CI recipients with straight arrays. Implications for device selection and the impact of electrode-to-modiolus distance will be discussed.