Influence of Electric Frequency-to-Place Mismatches on the Early Speech Recognition Outcomes for Electric–Acoustic Stimulation Users

Abstract

Purpose: Cochlear implant (CI) recipients with hearing preservation experience significant improvements in speech recognition with electric–acoustic stimulation (EAS) as compared to with a CI alone, although outcomes across EAS users vary. The individual differences in performance may be due in part to default mapping procedures, which result in electric frequency-to-place mismatches for the majority of EAS users. This study assessed the influence of electric mismatches on the early speech recognition for EAS users. Method: Twenty-one participants were randomized at EAS activation to listen exclusively with a default or place-based map. For both groups, the unaided thresholds determined the acoustic cutoff frequency (i.e., > 65 dB HL). For default maps, the electric filter frequencies were assigned to avoid spectral gaps in frequency information but created varying magnitudes of mismatches. For place-based maps, the electric filter frequencies were assigned to avoid frequency-to-place mismatches. Recognition of consonant–nucleus–consonant words and vowels was assessed at activation and 1, 3, and 6 months postactivation. Results: For participants with default maps, electric mismatch at 1500 Hz ranged from 2 to −12.0 semitones ( Mdn = −5 semitones). Poorer performance was observed for those with larger magnitudes of electric mismatch. This effect was observed through 6 months of EAS listening experience. Conclusions: The present sample of EAS users experienced better initial performance when electric mismatches were small or eliminated. These data suggest the utility of methods that reduce electric mismatches, such as place-based mapping procedures. Investigation is ongoing to determine whether these differences persist with long-term EAS use. Supplemental Material: https://doi.org/10.23641/asha.22096523