Low-frequency pitch sensitivity and speech perception performance in adult cochlear implant users fitted with fine structure strategies

TL;DR

This study shows that cochlear implant users using fine structure strategies can better detect low-frequency pitch changes and perform better in speech recognition, especially in noisy environments.

Contribution

The study demonstrates that fine structure strategies improve low-frequency pitch sensitivity and speech perception in cochlear implant users.

Findings

CI users with fine structure strategies showed significantly better low-frequency pitch discrimination via repetition rate.

RP performers had better speech recognition in noise compared to PP performers.

Bilateral CI users showed significant benefits in word recognition and matrix test performance.

Abstract

Cochlear implant (CI) users may perceive pitch changes via repetition rate (RP) and place-of-stimulation (PP) coding mechanisms. This study investigated whether CI users fitted with fine structure (FS) strategies can discriminate low-frequency (LF) pitch changes via RP and whether RP performers show better speech recognition than PP performers.

Thirty postlingually deafened adult CI users (15 unilateral, 15 bilateral) participated in this study. LF pitch discrimination linked to temporal fine structure (TFS) sensitivity was assessed with the Disharmonic Intonation test (A§E psychoacoustic test suite), while speech perception was evaluated with phonetically balanced words and everyday sentences (both in quiet and noise), and the Italian Matrix test (in adaptive mode).

Just noticeable differences (JNDs) in RP performers were significantly better compared to PP JNDs (p < 0.001, r = 0.80). No significant differences were found between unilateral and bilateral CI users for LF pitch perception (p > 0.05). Within-group comparisons (better ear versus bilateral listening) revealed that better ear JNDs were not significantly different from the bilateral performance (p > 0.05), whereas significant bilateral benefit was observed for word recognition score (WRS) + 10 (p = 0.016, r = 1.00), WRS + 5 (p = 0.021, r = 1.00), and Matrix performance (p = 0.033, r = 0.80). Speech recognition in noise was significantly better in RP performers compared to PP performers across multiple tests, including WRS + 10 (p = 0.002, r = 0.90), WRS + 5 (p = 0.001, r = 0.90), and the Matrix test (p = 0.03, r = 0.60).

These findings highlight the advantages of FS coding strategies in improving TFS sensitivity and speech perception in complex listening environments.

The online version contains supplementary material available at 10.1007/s00405-025-09449-y.

Figures