Benita K. Nordenstrom
Magdalene H. Chalikia
Elizabeth M. Ebsen
Dept. of Psychol., Moorhead State Univ., Moorhead, MN 56563
When listeners hear a repeated sequence of steady-state vowels (of the same duration and pitch) phonemic transformations occur, and they report hearing words and phrases absent in the original stimulus. A previous study [M. H. Chalikia, R. Meyer, and R. Lindemann, 34th Psychon. Society Meet. (1993)] investigated the possible effects of variations in vowel duration, pitch, or both. Listeners successfully matched these vowel sequences to the verbal forms heard with vowels of equal duration and pitch (base-line stimuli). In this study a broader variation in pitch was employed, and the vowel sounds were computer-generated rather than naturally produced. Six base-line sequences of six 60-ms vowels (at 100 Hz), followed by a 300-ms silent gap, were used. Four variations of these were created by randomly changing the pitch of individual vowels. Most listeners were able to match the modified vowel sequences to the verbal forms heard with the base-line stimuli, thus confirming the robust, stable nature of the verbal organizations. It is suggested that these organizations are based upon objective acoustic characteristics, and the present stimulus manipulations were probably perceived as prosodic variations.