Taking part in verbal conversation involves multiple cognitive processes, as listeners must process and retain speech while preparing a response. These processes rely on the efficient allocation of perceptual and cognitive resources. In comparison to healthy hearing, sounds transmitted via cochlear implants (CI) are limited, especially in terms of spectro-temporal cues. CI users are therefore likely to work harder in verbal conversation than typical-hearing (TH) listeners due to the degraded signal quality, which may require additional cognitive compensation.

This ongoing study investigates the extent to which CI users and TH listeners can memorize and retrieve speech information, and how this affects speech processing and cognitive load. To this end, participants completed an auditory n-back task in which matrix-sentences consisting of three words—a numeral, an adjective, and an object (e.g., “seven red flowers”)—were presented in quiet. Participants were asked to repeat each sentence and indicate whether a specified word was presented in the current (zero-back) or the previous (one-back) sentence.

We show preliminary results on outcome measures, including speech intelligibility, retrieval accuracy, and reaction times. Listening effort and memory load—indexed by pupil dilation—and subjective ratings of task load were also assessed. In order to examine the influence of individual factors, several cognitive functions (processing speed, cognitive flexibility, focused attention, and working memory) and participant age were considered.

We hypothesize that increasing task difficulty—from the zero-back to the one-back task—will elevate working memory load, resulting in lower retrieval accuracy, longer reaction times, and greater pupil dilation, and that these outcomes will be associated with individual factors. Increasing task difficulty is expected to have a more detrimental effect on CI users compared with their TH peers.