PhD defence by Hyojin Kim

On Tuesday 17 May, Hyojin Kim will defend her PhD thesis "Physiological correlates of the audibility of masked signals at supra-threshold levels".

Time: 11:00
Place: Bldg. 344, room 105 & Zoom:
Please be aware that the PhD defence may be recorded - This will also be informed at the beginning of the PhD defence.

Supervisor: Associate Professor Bastian Epp
Co-Supervisor: Professor Torsten Dau

Assessment Committee:
Associate Professor Jens Hjortkjær, DTU Health Tech
Professor Laurel Carney, University of Rochester
Professor Jesko L. Verhey, University of Magdeburg

Senior Researcher Jens Bo Nielsen, DTU Health Tech

Hearing is essential for communicating with others. Therefore, hearing loss can significantly impact a person’s quality of life. Hearing healthcare can alleviate hearing impairment by providing clinical solutions, but current technologies only provide limited improvement. This is due to the complexity of auditory processing in the brain. Thus, understanding neural sound computation is essential to provide accurate diagnostic measures and efficient treatments for hearing healthcare.

This thesis investigated how the auditory system uses sound properties to perform sound analyses in noisy background situations. First, behavioral listening tests were used to examine what sound properties are beneficial for listening in a noisy environment. Second, the feasibility of electroencephalography (EEG) as an objective measure of audibility was explored. Third, computational auditory modeling with a spiking neural network (SNN) was demonstrated. Results show that modulation patterns, spatial information, and temporal contexts can enhance the audibility of masked signals. In addition, neural signals measured with EEG can be correlated with behavioral audibility measures. Lastly, the SNN shows its potential as a tool for investigating underlying neural mechanisms of sound perception.

Overall, this thesis provides an understanding of sound perception in various perspectives using psychoacoustics, EEG, and computational modeling approaches. The resulting data will help identify the shortcomings of current hearing solutions and provide new insights for improving hearing healthcare.


Tue 17 May 22
11:00 - 14:00


DTU Sundhedsteknologi


Bldg. 344, room 105 and Zoom