Physiological markers of listening effort and their dependence on age

Speech understanding is a key function of human audition with important roles in learning, professional, and social functions. Ageing is associated with increased difficulties with speech recognition, particularly in noisy backgrounds, although traditional clinical tests frequently fail to account for subjective speech perception difficulties. A growing body of work has identified increased compensatory use of cognitive resources, collectively known as listening effort (LE), as a key change within the ageing and hearing-impaired populations. Consequently, accurate measures of LE may serve as critical objective diagnostic tools for quantifying subjective speech perception deficits. While multiple methods for quantifying LE have been proposed, little is known about whether these measures reflect usage of the same underlying cognitive processes, particularly in the context of continuous speech in the presence of other meaningful speech maskers. Moreover, little is known about whether changes in LE affect processing of lower- and higher-level features (e.g., acoustic envelope vs. lexical surprisal) throughout the auditory cortical hierarchy. This proposal will address these important gaps in our knowledge. In Aim 1, we will manipulate LE demands via changes in signal-to-noise ratio or spatial cues in a speech recognition task utilizing isolated sentences masked by competing speech, while measuring three commonly used measures of LE (self-report, pupil size, and neural measure of alpha band power). We will then correlate these measures with one another to determine whether they are similarly modulated by intelligibility manipulations, indicating common underlying cognitive processes. In Aim 2, we will address two key questions. First, we will test whether the LE measures used in Aim 1 exhibit similar relationships with one another in the context of more naturalistic, continuous speech, more closely mimicking scenarios in which speech perception difficulties are commonly encountered. Second, we will use advanced model-based analyses to decompose speech-evoked electroencephalographic activity into responses to lower- and higher-level features of both attended and ignored speech and assess how LE manipulations modulate these representations. Doing so, we will gain novel insights into the interaction between higher-order cognitive processes and the language processing pathways in the cortex, potentially revealing new neural signatures of LE. Finally, to gain deeper understanding of aging effects on LE and speech processing, we will assess how each Aim’s key measures change as a function of age. This work will provide foundation for future studies aimed at a more detailed characterization of the effects of different cognitive processes involved in LE on cortical processing of attended and ignored speech, with potential applications in diagnostics of speech processing deficits, particularly in scenarios where standard clinical tests fail to detect an impairment, and in testing speech-perception benefits from new algorithms used in hearing- assisting devices.