When listening situations become increasingly challenging due to background noise, individuals tend to disengage from listening. Based on previous research on physical and mental effort, we hypothesized that the decision to disengage is influenced by multiple factors including the speech-to-noise ratio, the potential reward of listening, current fatigue levels and recent errors in recall.

This study used an effort discounting paradigm in which participants were presented with offers indicating the task difficulty and potential reward on the upcoming trial. On each trial, participants chose whether they wanted to work (perform a speech-in-noise recall task to potentially earn credits) or if they wanted to rest. After each trial, participants provided a subjective fatigue rating and received feedback on their performance.

Different computational models were tested for their ability to explain participants’ choices and fatigue ratings on a trial-by-trial basis. Additionally, pupillometry and fNIRS were used to measure pupil dilation and cortical blood oxygenation changes in response to the stimuli.

The data shows that participants chose to work less frequently in high-difficulty, low-reward conditions. These offers were also rejected more frequently in the second half of the experiment. Fatigue ratings increased slightly more after trials with recall errors compared to trials where the sentence was repeated correctly.

Preliminary results favor a computational model in which the subjective value of effort is characterized by the reward on offer, discounted by the associated task difficulty. The effect of task difficulty is scaled by an effort discounting factor that incorporates fatigue levels and performance feedback.