Background:
Cochlear implant (CI) users’ enjoyment of music and
understanding of spoken language depend dependsnot only on hearing
sounds but also on distinguishing between them . While sound
sensation is typically measured from the auditory brainstem response,
no common objective measure has been established for testing sound
discrimination ability. We here propose t the cortical mismatch
negativity response (MMN) [1] and spike densitity component analysis
(SCA) [2] as novel diagnostical tools.
Methods:
11 experienced CI users and 14 normal hearing controls
listened to the CI MuMufe MMN parad paradig m [1]. MMN responses to
deviant tones for 4 sound features each tested at 4 magnitude levels
were measured with electroencephalography (EEG). The individual
MMN was automatically detected with SCA statistics. Attentive sound
discrimination ability was measured with a behavioral test.
Results:
Automatically detected individual MMN responses predicted
CI users’ behavioral sound discrimination ability with high (~90%)
accuracy , even for complex music stimuli. SCA statistics, tailored for
analysis on individuals, showed significantly higher diagnostic
accuracy ( p<.001) and reliability (replication across individuals)
(p=.003) than previous methods inspired by group group-level EEG analysis.