The effort of mentally repairing speech in individuals with hearing loss

Abstract

Over 460 million people worldwide have hearing loss (HL) that negatively impacts their ability to communicate (Davis & Hoffman, 2019). In the clinic, performance is measured by the percentage of words a listener repeats correctly. However, these scores reflect not only the health of the auditory system but also the listener’s ability to mentally repair misperceptions by using knowledge of the language and context (“cognitive repair”). Standard measures of speech perception cannot detect if a person used cognitive repair or if they accurately heard speech (with no need for repair). Detecting a person’s reliance on cognitive repair is important because while reliance on an extra moment to use context is helpful in the testing booth, it may break down in the real world as the next sentence would be heard before the previous sentence was fully processed. We hypothesize that continual need for cognitive repair is at the heart of what makes listening effortful, and what ultimately leads to increased fatigue (Edwards, 2017), anxiety (Morata et al., 2005), and social withdrawal (Hughes et al., 2018) for people with HL. The goals of this dissertation are to (1) identify listener reliance on cognitive repair, (2) measure the timeline of cognitive repair and its interference with ongoing processing, and (3) measure the relief from effort resulting from priming. The first study demonstrates the use of a clinically feasible test using behavioral measures that identifies when a listener relies on the moment immediately following the sentence to use context. Importantly, this test will better identify patients with HL who use cognitive repair during clinical testing which can lead to improved individualized patient centered care. The second study uses a dual-task paradigm to better identify the amount of time needed for cognitive repair after a sentence. During this time the listener would be susceptible to interference from an upcoming sentence in real-world conversation. The third study uses pupillometry to measure how the effort of repairing speech is affected by listeners having a preview of the missing word.