Associations Between Cognition and Speech Understanding in Cochlear Implant Candidates

Individuals who are being considered for cochlear implantation are assessed with tests of speech understanding to estimate hearing aid adequacy and, thus, cochlear implant (CI) candidacy. Lower scores suggest a CI may provide benefit, and specific thresholds (e.g., <60% correct responses) are established as criteria for CI. The influence of cognition on speech understanding is not considered in most CI clinics or in candidacy criteria, despite a scientific literature describing its importance for speech understanding when the signal is ambiguous (e.g., due to hearing loss or background noise). It is unclear if there is an upper limit of hearing loss (e.g., as determined by pure-tone threshold average [PTA]) beyond which the afferent signal does not provide enough information to facilitate cognitive processing for speech understanding. The objectives of the study were to determine, among adults presenting for CI candidacy assessment, the degree to which measures of executive function and memory are associated with performance on common tests of speech understanding used in CI clinics, and whether associations are larger in patients with lower (better) PTA.

Adult participants presenting for CI candidacy assessment at a tertiary referral center were recruited. An existing dementia diagnosis was an exclusion criterion. Participants underwent routine CI candidacy assessments. Cognition was assessed using the Brief Visuospatial Memory Test-Revised, the Mental Alternation Test, the Animal Fluency Test, semantic fluency tests, and the Stroop test. Tests of speech understanding included the AzBio sentence test (in noise and quiet) and the consonant-nucleus-consonant word test (in quiet). Associations between cognition and speech understanding were determined using multivariable linear regression, adjusting for age, sex, deafness duration, and PTA. Multiplicative interaction terms were used to test for interactions between cognitive variables and PTA.

There were 64 participants in total (PTA range 70 to 120+ dB HL). Multivariable regression results were as follows: Mental Alternation Test was associated with AzBio sentence-in-noise performance for individuals with PTA <92 dB HL and with AzBio sentence-in-quiet performance for individuals with PTA <87 dB HL. Brief Visuospatial Memory Test-Revised was associated with consonant-nucleus-consonant for individuals with PTA <100 dB HL (all associations in the expected direction).

Measures of memory and executive function were positively associated with speech understanding among potential CI candidates with better hearing (i.e., PTA <87 to 100 dB HL, depending on the speech test). The results suggest that speech understanding scores are a function of both the adequacy of the afferent signal and cognitive processing among CI candidates with better PTA.