Across-Frequency Temporal Processing and Speech Perception in Cochlear Implant Recipients and Normal Hearing Listeners

Abstract

Objective: Individual differences in temporal processing contributes strongly to the large variability in speech recognition performance observed among cochlear implant (CI) recipients. Temporal processing is traditionally measured using a behavioral gap detection task, and therefore, it can be challenging or infeasible to obtain reliable responses from young children and individuals with disabilities. Within-frequency gap detection (pre- and post-gap markers are identical in frequency) is more common, yet across-frequency gap detection (pre- and post-gap markers are spectrally distinct), is thought to be more important for speech perception because the phonemes that proceed and follow the rapid temporal cues are rarely identical in frequency. However, limited studies have examined across-frequency temporal processing in CI recipients. None of which have included across-frequency cortical auditory evoked potentials (CAEP), nor was the correlation between across-frequency gap detection and speech perception examined. The purpose of the study is to evaluate behavioral and electrophysiological measures of across-frequency temporal processing and speech recognition in normal hearing (NH) and CI recipients. Design: Eleven post-lingually deafened adult CI recipients (n = 15 ears, mean age = 50.4 yrs.) and eleven age- and gender-matched NH individuals participated (n = 15 ears; mean age = 49.0 yrs.). Speech perception was evaluated using the Minimum Speech Test Battery for Adult Cochlear Implant Users (CNC, AzBio, BKB-SIN). Across-frequency behavioral gap detection thresholds (GDT; 2 kHz to 1 kHz post-gap tone) were measured using an adaptive, two-alternative, forced-choice paradigm. Across-frequency CAEPs were measured using four gap duration conditions; supra-threshold (behavioral GDT x 3), threshold (behavioral GDT), sub-threshold (behavioral GDT/3), and reference (no gap) condition. Group differences in behavioral GDTs, and CAEP amplitude and latency were evaluated using multiple mixed effects models. Bivariate and multivariate canonical correlation analyses were used to evaluate the relationship between the CAEP amplitude and latency, behavioral GDTs, and speech perception. Results: A significant effect of participant group was not observed for across-frequency GDTs, instead older participants (> 50 yrs.) displayed larger GDTs than younger participants. CI recipients displayed increased P1 and N1 latency compared to NH participants and older participants displayed delayed N1 and P2 latency compared to younger adults. Bivariate correlation analysis between behavioral GDTs and speech perception measures were not significant (p > 0.01). Across-frequency canonical correlation analysis showed a significant relationship between CAEP reference condition and behavioral measures of speech perception and temporal processing. Conclusions: CI recipients show similar across-frequency temporal GDTs compared to NH participants, however older participants (> 50 yrs.) displayed poorer temporal processing (larger GDTs) compared to younger participants. CI recipients and older participants displayed less efficient neural processing of the acoustic stimulus and slower transmission to the auditory cortex. An effect of gap duration on CAEP amplitude or latency was not observed. Canonical correlation analysis suggests better cortical detection of frequency changes is correlated with better word and sentence understanding in quiet and noise.