Patterns of phoneme confusions with sense and nonsense CVC words for subjects with normal hearing and with presbycusis

Abstract

Differences in listening strategy between normal-hearing and hearing-impaired listeners will become apparent in different patterns of phoneme confusions. The confusions were studied with meaningful and meaningless word material of the consonant-vowel-consonant type (CVC words) presented in quiet at fixed presentation levels to a group of young normal-hearing subjects, a group of elderly subjects with normal hearing regarding their age, and subjects with presbycusis. The confusions of the vowels and initial and final consonants were subjected to INDSCAL analysis [J. D. Carroll and J. J. Chang, Psychometrika 35, 283–319 (1970)]. Parameters in the analysis were; word type (meaningful versus meaningless), group of subjects, and phoneme identification score. This latter parameter is perceptually more relevant than presentation level when comparing subjects with different hearing losses. For the normal-hearing subjects, vowel perception was dominated by the first and second formant, whereas for presbycusis subjects the contribution of the second formant was reduced and some influence of vowel duration was present. For the perception of both initial and final consonants voicing was all important. However, the feature of voicing was better perceived by presbycusis subjects than by the normal-hearing subjects.