*Study Limits and Future Improvements*

Subjects were asked to vote at least one time per day during the two-week take-home period. A total of 205 votes were recorded from the 11 patients, resulting in over one vote per subject per day, on average. All patients were able to use the remote control for data capture, successfully demonstrating the feasibility of EMA within a CI population. However, our self-initiated data capture method had some limitations. Six of the eleven patients provided at least one vote per day, while the remaining five subjects voted less often (averaged over the two-week period). Due to the low number of votes, four of those subjects had preference outcomes that were non-conclusive. In future such EMA studies, it would be beneficial to include a forced patient prompt, to request patients to conduct a comparison and vote. Additionally, an incentive, e.g., progress bars and gamification, for the study participant may help to minimize missing votes. Furthermore, the recipients should have the chance to withdraw an accidental vote. To support this, a possible review by the participants themselves of all votes might be worth considering for future studies. To summarize, most of the above-mentioned deficits cannot be addressed within a feasibility study.

The problem of obtaining an adequate number of responses is also described by Wu et al. [38]. The prompting frequency, take-home duration, number of acoustic scenes captured, and number of subjects need to be optimized to achieve sufficient data for subsequent statistical analysis. Nevertheless, a sufficient number of valid votes were collected in this study, showing the feasibility of EMA as well as this methodology in CI recipients.

The current fitting philosophy is to provide beneficial sound processing algorithms for each listening scene, which has been shown to provide benefits to a group, on average [31,39]. However, it may be that individual algorithm selection could provide further individual benefit [12]. These varying individual performance benefits or preferences can provide input for a machine learning approach [40] to select individualized sound processing algorithm options for specific listening scenes.

A central question for clinical application remains: is the individual scene-specific preference aligned with the benefits shown in speech-audiometry tests. The EMA methodology is expected to elucidate such real-world individualized patient algorithm preferences and may help to fit speech processing algorithms better to the individuals' needs.

## **5. Conclusions**

This study found that program preference varied significantly among subjects. Some demonstrated an overall program preference, others demonstrated scene-specific preferences, and others demonstrated no conclusive preference. The data collection tool was integrated into the patient's cochlear implant system and was suitable for real-world selfassessment. To improve data collection, future research should encourage participants to realize a higher response rate.

**Author Contributions:** Conceptualization, A.A.H., S.J.M., M.H.; methodology, S.J.M., M.H., B.B.; software, A.A.H., S.J.M.; validation, B.B., A.M., M.H.; formal analysis, A.A.H.; investigation, B.B., A.M., M.H.; resources, M.H.; data curation, B.B., M.H.; writing—original draft preparation, S.J.M., A.A.H., M.H., T.H.; writing—review and editing, all authors; visualization, A.A.H.; supervision, S.J.M., M.H.; project administration, M.H.; funding acquisition, M.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Cochlear Europe, grant number IIR-Ki15.

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki and approved by the CAU Ethics Committee (D467/16 1 June 2016).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We wish to cordially thank all the patients who kindly took time to participate in the investigations. Our special thanks go to the medical-technical assistants at the Kiel University ENT Clinic, who conducted measurements for this study.

**Conflicts of Interest:** A.A.H., S.J.M. and T.H. are/were employees of Cochlear. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
