EVALUATION COMMITTEE

The Faculty Council has appointed the following adjudication committee to evaluate the thesis and the associated lecture: 

• Dr. Jens Hjortkjær, Technical University of Denmark, Denmark
• Prof. Nathan Weisz, Paris Lodron University Salzburg, Department of Psychology, Austria
• Dr. Romulus Lontis, HST, Aalborg University, Denmark (Chairman).

Moderator:
Dr. Carsten Dahl Mørch, Health Science and Technology, Aalborg University

ABSTRACT

Listening effort is a major complaint of people with hearing loss. In recent years there has been growing interest in an objective quantification of listening effort. In this thesis, we further investigated neural measures alongside self-reported measures of listening effort with normal-hearing participants performing speech-in-noise tasks. Behavioral and electroencephalography (EEG) data were collected while participants listened to regular sentences and random word lists (with no syntactic and sentences-level semantic information) in background noise at different signal-to-noise ratios (SNRs) (-9, -6, -3, 0 dB).

The thesis consists of three studies. In the first study, we investigated the effect of SNR and linguistic violations (operationalized by sentences and word lists) and their interaction on listening effort. Behavioral and EEG data showed a significant interaction of SNR and linguistic violation on the self-reported listening effort rating and frontal theta band (4 - 7 Hz) power during the memory retention interval. That is, increased theta power was found for word lists than sentences at only SNR 0 dB, representing increased memory load for word slits. Decreasing SNR showed an increased trend for frontal theta power in sentences but not in word lists. Frontal theta power was found to be positively correlated with self-reported listening effort in sentences.

In study two, we assessed the reliability of self-reported and frontal midline theta power as measures of listening effort. Generally, self-reported measures showed an acceptable between-session variability that could be interpreted as a reliable measure of listening effort. However, frontal midline theta power showed greater between-session variability, leading us to be cautious about interpreting it as a reliable measure of listening effort, despite the acceptable effect sizes shown in study one.

 In study three, we examined the effect of SNR and linguistic violations and their interaction on cortical speech tracking, quantified by the phase-locking value (PLV). We then related PLV values to the self-reported listening effort. Results showed an interaction of SNR and linguistic violation on PLV values in the delta frequency band (1-4 Hz). We observed an increased PLV for sentences compared to word lists at SNR 0 dB in sentences, indicating that the delta band speech tracking reflects linguistic information consistent with the literature.  PLV only in sentences was also modulated by SNR that showed high PLV at SNR 0 dB compared to -9 dB. A marginally significant negative relationship between speech tracking and self-reported effort was observed in sentences; this trend corroborates the literature showing that decreased speech tracking is associated with increased listening effort.

The findings of this work emphasize the role of linguistic information in speech perception. This is evident from studies one and three, which showed lower frontal theta power during the memory retention interval, likely representing memory load,  and higher delta-band PLV at SNR 0 dB in sentences compared to word lists. The results also indicated the effect of background noise that increased memory load and decreased delta band PLV in sentences.