Abstract
Vowel and consonant confusions were obtained from two patients, one using the Melbourne implant (M3) and one the Utah device (U1). The confusion matrices were analyzed using information transfer analysis and multidimensional scaling. The results from the two patients were generally consistent with the different processing strategies of their implants. For the vowels, M3 appeared to be processing second formant (F2) and duration, whereas U1 appeared to be attending primarily to first formant (F1) and duration. For the consonants, the responses of M3 reflected sensitivity to frication, voicing, and envelope, whereas the results from U1 suggested sensitivity to voicing, nasal, frication, and possibly “place” information.
Get full access to this article
View all access options for this article.