(PP803) Influence of Maxima Peaks on Perception of Musical Sound Quality in Cochlear Implant Users

Lead Presenter(s)

Michael Z. Burchesky, BA

Audiology Extern
Mayo Clinic Jacksonville
Canonsburg, Pennsylvania, United States

Presenter(s)

GS

Greta Stamper, AuD, PhD

Audiology Chair
Mayo Clinic Florida
Jacksonville, Florida, United States
DZ

David Zapala, PhD

Associate Professor
Mayo Clinic Florida
Jacksonville, Florida, United States

Contributor(s)

TP

Terri Pratt
MG

Millicent Garry
CL

Courtney C. Luffler, AuD

Clinical Audiologist
Mayo Clinic
Jacksonville, Florida, United States

Abstract: Musical sound quality (MSQ) perception is a listening priority for many patients utilizing cochlear implants (CIs). CI audiologists can adjust parameters to affect sound processing algorithms, including the number of maxima selected for stimulation. There are limited data available in the literature examining the influence of maxima number on MSQ. In this study, the CI Multiple Stimulus with Hidden Reference and Anchor (CI-MUSHRA) test battery was utilized to objectively measure CI patients’ MSQ perception of musical stimuli under low-, middle-, and high-maxima conditions. Patterns of MSQ judgments and influences of the number of maxima on MSQ will be discussed.

Summary: Rationale. There are a number of programming parameters available to cochlear implant (CI) audiologists that can be fine-tuned to adjust CI sound processing to maximize patient performance. In some programming softwares, and for processors utilizing n-of-m strategies, the number of maxima can be altered to control how many channels are selected for stimulation during each sweep of a sampled acoustic input signal. The channels are selected based on the highest amplitude peaks of each fast fourier transform (FFT) calculated for a given sample. Previous research has established that a maxima of at least 9 is required for optimal speech recognition in noise (Dorman et al., 2002). According to “Programming Cochlear Implants” by Wolfe & Schafer (2015), a maxima of 10 to 12 is often used in clinical practice. While speech understanding is the first priority for CI processing schemes and is the central focus of the bulk of current CI literature, sound quality perception and improved music listening experience are equally valid priorities for many CI patients. However, there are limited data in the literature evaluating the effect of maxima settings on music listening perception. The purpose of this study is to strengthen quantitative research on the relationship between the number of maxima and patients’ musical sound quality (MSQ) perception. Importance. The present study answers a call from Caldwell, Jiam & Limb (2017) for further research into assessing and improving sound quality perception in CI users. The authors stress the utilization of quantitative tools such as the CI Multiple Stimulus with Hidden Reference and Anchor (CI-MUSHRA) developed by Roy, et al. (2012) to reach reliable and repeatable conclusions as a field. With MSQ being an inherently subjective construct, systematic and objective research approaches are essential for establishing evidence-based solutions for clinically relevant patient needs. Methods. The CI-MUSHRA test battery was implemented via MATLAB to present 7 versions of musical soundbites with varying amounts of low-frequency (LF) degradation. The availability of LF content was altered because this parameter is a known contributor to MSQ perception in normal hearing (NH) listeners (Roy, 2012). Ten (N = 10) Cochlear Americas CI users (CP1000 or CP1150, unilaterally) were asked to make sound quality ratings (0 = “very poor”; 100 = “excellent”) for each soundbite. Participants were prompted to base their MSQ ratings in comparison to a reference soundbite of the highest quality (no LF degradation). Such a study design allows for an objective measurement of CI user ability to perceive and discriminate between high- and low-quality soundbites. This test block was completed under three maxima conditions: low (5-of-m), middle (10-of-m), and high (15-of-m). Results & Conclusions. Patterns of MSQ ratings will be analyzed alongside ratings from a cohort of NH listeners. Relationships will be discussed pending the results of a between groups 2-way analysis of variance (ANOVA) and linear regression analyses. Implications. Our findings may help to guide CI audiologists’ programming decisions in a clinical setting. The data will provide evidence-based guidance regarding the maxima setting for patients for whom music enjoyment is a listening priority.

Learning Objectives:
describe the influence of adjusting the cochlear implant (CI) maxima setting on CI patients' musical sound quality perception.