This study examined age-related differences in the use of dynamic acoustic information (in the form of formant transitions) to identify vowel quality in CVCs. Two versions of 61 naturally produced, commonly occurring, monosyllabic English words were created: a control version (the unmodified whole word) and a silent-center version (in which approximately 62% of the medial vowel was replaced by silence). A group of normal-hearing young adults (19–25 years old) and older adults (61–75 years old) identified these tokens. The older subjects were found to be significantly worse than the younger subjects at identifying the medial vowel and the initial and final consonants in the silent-center condition. These results support the hypothesis of an age-related decrement in the ability to process dynamic perceptual cues in the perception of vowel quality.
Dynamic Information Networks: Geometry, Topology and Statistical Learning for the Articulation of Structure