Temporal fine structure processing, pitch and speech perception in adult cochlear implant recipients

Objectives: The aim of the study was to investigate the link between Temporal Fine Structure (TFS) processing, pitch and speech perception performance in adult cochlear implant (CI) recipients, including bimodal listeners who may benefit better low frequency (LF) temporal coding in the contralateral ear. Design: The study participants were 43 adult CI recipients (23 unilateral, 6 bilateral, 14 bimodal listeners). Two new LF pitch perception tests - Harmonic Intonation (HI) and Disharmonic Intonation (DI) - were used to evaluate TFS sensitivity. HI and DI were designed to estimate a difference limen for discrimination of tone changes based on harmonic or inharmonic pitch glides. Speech perception was assessed using the newly developed Italian STARR test where sentences relevant to everyday contexts were presented at low, medium and high levels in a fluctuating background noise in order to estimate a Speech Reception Threshold (SRT). Results: Although TFS and STARR performances in the majority of CI recipients were much poorer than those of hearing people reported in the literature, a considerable inter-subject variability was observed. For CI listeners, median Just Noticeable Differences were 27.0 Hz and 147.0 Hz for HI and DI respectively. HI outcomes were significantly better than those for DI. Median STARR score was 14.8 dB. Better performers with SRTs less than 20 dB had a median score of 8.6 dB. A significant effect of age was observed for both HI/DI tests, suggesting that TFS sensitivity tended to worsen with increasing age. CI pure tone thresholds and duration of profound deafness were significantly correlated with STARR performance. Bimodal users showed significantly better TFS and STARR performance for bimodal listening than for their CI only condition. Median bimodal gains were 33.0 Hz for the HI test and 95.0 Hz for the DI test. DI outcomes in bimodal users revealed a significant correlation with unaided hearing thresholds for octave frequencies lower than 1000 Hz. Median STARR scores were 17.3 versus 8.1 dB for CI only and bimodal listening respectively. STARR performance was significantly correlated with HI findings for CI listeners and with those of DI for bimodal listeners. Conclusions: LF pitch perception was found to be abnormal in the majority of adult CI recipients, confirming poor TFS processing of CIs. Similarly, the STARR findings reflected a common performance deterioration with the HI/DI tests, suggesting the cause probably being a lack of access to TFS information. Contralateral hearing aid users obtained a remarkable bimodal benefit for all tests. Such results highlighted the importance of TFS cues for challenging speech perception and the relevance to everyday listening conditions. HI/DI and STARR tests show promise for gaining insights into how TFS and speech perception are being limited and may guide the customization of CI program parameters and support the fine tuning of bimodal listening.