One of the main reasons people can carry on conversations in noisy environments is that the perceptual system can rely on multiple high-level cognitive mechanisms to restore the degraded auditory input into meaningful speech. The brain is able to reconstruct parts of speech not audible due to noise, forming an intelligible speech stream. This phonemic restoration effect can be shown with increased intelligibility of periodically interrupted speech after filling the silent intervals with noise bursts. Having a conversation in background noise is very difficult for hearing-impaired people; the incoming signal is so poor that the compensation mechanisms may fail to help. This study investigated underlying mechanisms of top-down speech restoration, a previously poorly understood phenomenon, in the presence of sound degradations that can happen in cochlear implants (CI, prosthetic devices for deaf individuals). The results show that the understanding of interrupted speech can be increased through training, perhaps teaching individuals to make better use of their top-down restoration mechanisms. We have shown that top-down cognitive restoration mechanisms are less effective if the bottom-up auditory signal is of insufficient quality (as occurs in CI speech processing). Both the cognitive and auditory processes play an important role in the restoration of interrupted speech. Especially high-level linguistic mechanisms have a large influence on the restoration of interrupted speech; vocabulary and verbal intelligence are significant predictors of successful restoration of interrupted sentences. Since linguistic skills play an important role in the restoration of spectrally degraded interrupted speech, CI users can possibly be trained to improve these skills.
|Qualification||Doctor of Philosophy|
|Place of Publication||[Groningen]|
|Publication status||Published - 2015|