Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Communication Sciences and Disorders


The purpose of this research was to examine the use of prosody in a group of mentally retarded (MR) subjects who were impressionistically dysprosodic. Seven adult, institutionalized, severe mentally retarded subjects matched with two groups of normal subjects (matched for chronological age and for language age) were compared in two production experiments. The first experiment was a study of word level stress in which the subjects named pictures of two syllable, morphologically simple, non-derived words. The second experiment was a study of sentence level stress in which the subjects described changing toy locations with three word (subject-preposition-object) utterances, intended to require the use of prosody to distinguish contextually unchanged information from changed information. All subject productions were analyzed perceptually for stress accuracy by three sophisticated judges. Acoustic measurements were made using a Visi-Pitch 6095 interfaced with an Apple IIe computer and an Epson FX 100 printer. This instrument combination enabled extraction of fundamental frequency (F0) peaks, relative intensity (I0) peaks, mean F0, mean I0, and duration data for each subject's productions in each experiment. Comparisons were made of the subject groups' use of the acoustic parameters which might have cued stress. The lexical stress experiment revealed no differences in perceptual judgments of the groups' ability to mark stress on the appropriate syllable. The sentence accent experiment revealed that the MR group performed significantly poorer than two groups of normal subjects. The MR group was perceived to use utterance final stress in a majority of its productions. They did not appear to use prosody to distinguish contextually unchanged information from contextually changed information. No differences in stress cues were found for the first and third stress positions. However, variations in the use of these parameters occurred in the second stress position. The acoustic analyses did not reveal patterns which characterized the dysprosodic nature of the MR subjects' productions. Possible explanations for acoustic variations in the second stress position were discussed along with future research considerations.