Mechanisms underlying speech masking release in hybrid cochlear implant users

Abstract

Hybrid cochlear implant (CI) users utilize low-frequency acoustic and high-frequency electric stimulation in the implanted ear, outperform traditional CI users in speech perception in noise (Gantz et al., 2017), and show “masking release,” an improvement in speech recognition in fluctuating noise relative to steady noise (Turner et al., 2004). Improved performance in noise has been attributed (but not proven) to the spectral resolution and temporal fine structure (TFS) provided by acoustic hearing. Spondee recognition in two-talker and ten-talker babble was assessed in Hybrid CI users, and masking release was calculated as the percent difference in performance in two-talker (easy condition) and ten-talker (hard condition) babble. Spectral ripple density discrimination thresholds and fundamental frequency discrimination limens was also assessed, as these psychophysical measures reflect underlying spectral resolution and TFS and correlate with speech perception in noise in traditional CI users (Won et al., 2007; Goldsworthy et al., 2013; Goldsworthy, 2015; Jeon et al., 2015). Preliminary data suggest masking release is possible in Hybrid CI users, and that psychophysical measures and masking release may be correlated, suggesting a role of spectral resolution and TFS in masking release. The extent of residual hearing also appears to correlate with psychophysical and masking release performance.