The long-term objectives of this project are to understand better the relationship between the perception and production of speech and to improve services provided to individuals with phonological disorders. It has been postulated that one underlying cause for a phonological disorder may be an inability to correctly auditorily discriminate the error sound(s). Individuals who misarticulate a sound, however, often have little difficulty in discriminating that sound. In fact some individuals who display a phonological disorder produce and perceive acoustic differences in their own speech that are neither produced nor perceived by persons with normal speech, while also being able to consistently determine when a sound they incorrectly produce is accurately produced by other speakers. These individuals may have two different underlying representations (URs) for their error sound. each with different auditory characteristics. The auditory characteristics of one of these URs are derived from the input of other speakers. The auditory characteristics of the other UR are derived through the association of these speakers' own oral movements with the auditory consequences of the movements. This explains why these individuals can recognize correct articulation produced by other speakers, while at the same time perceiving subtle acoustic variations in their own speech. It also predicts that some clients might fail to attain correct production of their error sound because what sounds """"""""right"""""""" for their own speech is their error production. The possibility that certain people who misarticulate [r] have two different URs for their error sound. each with distinctive auditory characteristics, will be tested during year l of this project. The speech of subjects who misarticulate [r] will be tape recorded. Some of their productions will be """"""""corrected"""""""" electronically and recorded on audio tape along with their error productions. The tape will also contain incorrect and """"""""corrected"""""""" utterances spoken by another speaker matched for age and gender. Subjects will listen to the tape and determine whether each utterance is correct or incorrect and whether it was spoken by the subject him- or herself or by another speaker. If the subjects do have a UR for their own speech that differs from the UR they have for the speech of others, they may label their own error productions as correct and their """"""""corrected"""""""" utterances as incorrect while accurately identifying the errors of the other speaker. Research during year 2 will explore the use of visual feedback provided by a speech spectrograph to elicit correct articulation of [r] from individuals who have not responded to traditional therapy. It is hypothesized that individuals who failed to make progress in therapy have a UR for their own production of [r] which has auditory characteristics that differ from those of the UR that underlies their perception of others' speech. Subjects will be instructed to use visual feedback, but not auditory feedback, while they attempt to produce III. Preliminary data indicate that the speech spectrograph may prove useful in eliciting correct articulation.
| Shuster, L I (1998) The perception of correctly and incorrectly produced /r/. J Speech Lang Hear Res 41:941-50 |
| Shuster, L I (1996) Linear predictive coding parameter manipulation/synthesis of incorrectly produced /r/. J Speech Hear Res 39:827-32 |
