Can a model of overlapping gestures account for scanning speech patterns?

Area of Science:

• Speech-language pathology
• Acoustic phonetics
• Neurolinguistics

Background:

• Scanning speech is a dysarthria characterized by irregular speech rate and rhythm.
• Understanding the acoustic underpinnings of scanning speech is crucial for developing effective interventions.
• Coproduction, or gesture overlap, is a key aspect of fluent speech production.

Purpose of the Study:

• To investigate whether coproduction is reduced in neurologic speakers with scanning speech patterns.
• To examine the influence of speaking rate and utterance position on acoustic measures of coproduction.
• To determine if temporal variability in speech gestures can be explained by variations in coproduction.

Main Methods:

• Utilized a simple acoustic model of overlapping, sliding gestures.
• Measured F2 onset frequency as an acoustic correlate of coproduction and gesture overlap.
• Analyzed the effects of habitual versus fast speaking rates and initial versus medial utterance positions.
• Employed regression analyses to assess the relationship between temporal variability and coproduction.

Main Results:

• Disordered speakers showed lower F2 onset frequencies, suggesting reduced gesture overlap.
• Speaking rate did not significantly affect F2 onset frequencies in either healthy or disordered speakers.
• Utterance-initial positions exhibited reduced coproduction compared to medial positions, though potentially confounded by articulatory scaling.
• Gesture sliding partially explained temporal variability in F2 trajectories across repetitions.

Conclusions:

• Findings support the hypothesis of reduced coproduction in neurologic speakers with scanning speech.
• Speech production theories for healthy speakers can offer insights into disordered speech mechanisms.
• Acoustic analysis of F2 onset frequency provides valuable data for understanding speech motor control deficits.