This study investigates how signal-to-noise-interference ratio (SNIR) affects auditory performance and neural responses associated with listening effort (LE), and whether other acoustic factors beyond SNIR contribute to this relationship. We collected a new dataset from individuals with mild-to-moderate hearing loss, all fitted with hearing aids (HAs). Participants listened to two competing audiobooks from front-facing loudspeakers, while 16-talker babble noise was presented from four background speakers to simulate realistic acoustic scenes. Six SNIR levels (5.47, –3.55, –2.13, –1.19, –0.64, and –0.27 dB) defined six conditions, each consisting of 12 one-minute trials. In each trial, participants attended to one audiobook while ignoring the competing audiobook and background noise, then completed comprehension questions and reported subjective LE.

Behavioral results showed a significant linear effect of SNIR on subjective LE ratings and a quadratic effect on comprehension questionnaire accuracy: perceived LE decreased steadily with improving SNIR, whereas comprehension performance plateaued at higher SNIR levels.

EEG analyses revealed a significant linear relationship between SNIR and local connectivity in the parietal alpha band. Parietal local connectivity correlated with subjective LE ratings, while spectral power in the same region and frequency band did not, indicating that local connectivity is a more sensitive neural marker of LE. While alpha power and connectivity may share partially overlapping neural origins, they appear to reflect distinct aspects of the neural processes underlying effortful listening.

Beyond SNIR effects, systematic differences emerged between the two talkers under identical acoustic conditions. One talker consistently elicited higher subjective and neural indices of LE, suggesting that talker-specific features—such as voice characteristics or linguistic content—can modulate cognitive demand independently of acoustic degradation.

Together, these findings highlight that LE is shaped not only by the external acoustic environment but also by intrinsic speech properties. Parietal alpha connectivity provides a robust neural correlate of LE, reflecting both graded SNIR effects as well as talker-specific effects, making it a promising tool to assess cognitive resources during complex, realistic listening scenarios in HA users.