Residual hearing and sound quality in cochlear implant patients

Cochlear implants have transformed the treatment of severe hearing loss, allowing adults who once struggled to follow even shouted conversation to recognize words and sentences with remarkable accuracy. Yet a curious paradox has long shadowed the technology: patients whose audiograms look nearly identical often describe their hearing in strikingly different terms. Some call the sound natural and full; others describe something robotic, muffled, or metallic. A new study from Thomas Jefferson University Hospital, published in Frontiers in Audiology and Otology, suggests that part of the answer lies in what happens to the low-frequency hearing patients bring with them into surgery—and in whether the clinical system actually helps them use it afterward.

The research team, led by otolaryngologist Anika S. Walia, surveyed 31 adult cochlear implant recipients and sorted them into three groups based on the trajectory of their residual acoustic hearing. One group had no measurable low-frequency hearing before implantation. A second group entered surgery with residual hearing in the low frequencies—thresholds better than 80 decibels at 250, 500, and 1,000 hertz—and still had it afterward. The third group had that hearing going in but lost it as a consequence of the procedure. Each participant completed detailed questionnaires about how their implant-mediated hearing actually sounded to them, alongside the standard objective tests of speech recognition that clinicians have relied on for decades.

The technical logic behind the question is worth unpacking. A cochlear implant bypasses damaged hair cells and delivers electrical pulses directly to the auditory nerve, but the sound it conveys differs fundamentally from acoustic hearing. Electrical stimulation offers limited spectral resolution, a compressed dynamic range, and poor transmission of temporal fine structure—the subtle timing cues that help acoustic listeners distinguish a violin from a flute or follow one voice among many. When surgeons preserve residual low-frequency hearing, they can pair the implant’s high-frequency electric signal with acoustic amplification of the low frequencies in the same ear, an approach called electroacoustic stimulation, or EAS. This hybrid mode reliably outperforms electric-only listening for speech in noise, music appreciation, and sound quality. But preservation is not guaranteed: even with modern soft-surgery techniques such as round-window insertion, many patients lose their residual hearing anyway.

The study’s most provocative finding concerns the patients who lost something they once had. Those who lost residual hearing after implantation rated the character of their implant sound—whether it felt artificial, harsh, thin, hollow, muffled, or reverberant—lower than patients who had never had residual hearing to begin with. The difference persisted as a trend even after the researchers statistically adjusted for age, duration of deafness, and time since surgery. The authors propose a perceptual-mismatch mechanism: patients who lose residual hearing must transition from a lifetime of acoustic processing to purely electric hearing, and their vivid auditory memory of natural sound may heighten their awareness of the implant’s inherent distortions. Patients with no residual hearing, by contrast, may compare their electric hearing primarily to their memory of profound deafness rather than to acoustic hearing—and rate it more favorably.

Just as striking is what the study did not find. Objective speech recognition improved by clinically meaningful margins in all three groups, with mean gains on word-recognition testing of roughly 31 to 46 percentage points, and the groups did not differ significantly on those gains. Yet the patients with preserved residual hearing reported the worst subjective speech perception of all, with a stepwise gradient on the Speech, Spatial and Qualities questionnaire running from preserved-hearing patients at the bottom through hearing-lost patients to no-residual-hearing patients at the top. The dissociation between measured performance and perceived benefit was stark: patients whose audiometric thresholds showed intact low-frequency hearing paradoxically reported more difficulty following speech in complex, noisy, reverberant environments.

The explanation may lie in a physiological phenomenon called electric-acoustic interference. Psychoacoustic research has shown that electrical stimulation can mask acoustic hearing in the same cochlea, elevating the detection thresholds of acoustic tones even in patients not programmed for EAS. For the two-thirds of preserved-hearing patients in this cohort who used electric-only maps, that interference may have degraded the utility of their preserved acoustic hearing in exactly the demanding listening situations the questionnaire probes—multi-talker scenes, reverberant rooms—without delivering the optimized integration that EAS programming provides. Their intact thresholds, in other words, did not translate into intact perceived benefit.

Which brings the study to its most actionable finding: underutilization of EAS. Only three of the nine patients with preserved residual hearing—27 percent—had an active EAS mapping component at the time of survey. Those three patients, though a tiny subgroup, showed medium-to-large advantages in sound quality descriptors and perceived clarity over their non-EAS counterparts, with effect sizes of roughly 0.75, and a moderate advantage in pitch perception consistent with EAS’s known benefit for musical sound. The authors argue that this low uptake reflects a broader pattern documented across the literature, with barriers including physical discomfort from acoustic amplification, lack of perceived benefit, and difficulty adapting when low-frequency hearing deteriorates. The 80-decibel threshold used to define residual hearing may also have included patients whose preserved hearing was too poor for acoustic amplification to help much, since EAS is generally effective at thresholds of 70 decibels or better. Whatever the cause, the gap between surgical hearing-preservation success and clinical programming implementation represents, in the authors’ framing, a modifiable deficiency in care.

The findings also carry a message about how cochlear implant outcomes should be measured. Recent evidence, including work by Berg and colleagues, suggests that sound quality—not speech recognition—is the primary driver of implant-related quality of life, and this study reinforces that conclusion. Objective speech recognition scores showed only weak associations with patient-reported outcomes across the cohort. The hearing-lost group, despite speech-recognition gains equivalent to everyone else’s, posted the lowest numerically on the Cochlear Implant Quality of Life-35 instrument. Quiet-room word tests, the study suggests, simply fail to capture how patients experience hearing in the real world—conversation in background noise, music, traffic—where the perceptual character of sound matters most.

The authors are careful to frame the work as exploratory. Thirty-one patients divided across three groups, a cross-sectional design that precludes causal inference, variable timing of survey completion relative to surgery, heterogeneous device manufacturers, and an unvalidated sound-quality instrument all limit statistical power and generalizability. The timing and mechanism of hearing loss in the hearing-lost group—immediate surgical trauma versus delayed progressive deterioration—remain unknown, and the two may carry different implications for adaptation. Larger prospective studies with standardized pre- and post-operative questionnaires, datalogging of device use, and investigation of the factors driving EAS uptake are needed before the findings change clinical practice.

Still, the study’s central insight is likely to resonate far beyond audiology clinics. What patients hear through a cochlear implant is not simply a function of how well they recognize words on a test; it is shaped by auditory memory, expectation, and the perceptual contrast between what they hear now and what they heard before. Losing residual hearing may mean losing something perceptually meaningful, even when the audiogram suggests nothing was lost at all. And preserving hearing, without the programming and follow-through that lets patients actually use it, may preserve little of subjective value. In an era when implant candidacy keeps expanding to patients with better and better preoperative hearing, the study suggests that the most important outcome may not be what the audiogram records—but what the patient reports.

Subject of Research: Residual hearing and sound quality in cochlear implant patients

Article Title: Residual hearing and sound quality in cochlear implant patients

Article References: Walia, A. S., Chiles, N., O'Brien, L., Uralov, D., Willcox, T., Chiffer, R., & Hunter, J. B. (2026). Residual hearing and sound quality in cochlear implant patients. Frontiers in Audiology and Otology, 4, Article 1913231. https://doi.org/10.3389/fauot.2026.1913231

Image Credits: AI Generated

DOI: 10.3389/fauot.2026.1913231

Keywords: Residual, hearing, sound, quality, cochlear, implant, patients, scientific research