|Title:||Phonetic Effects on the Timing of Gestural Coordination in Modern Greek Consonant Clusters||Add Dissertation|
|Author:||Jonathan Yip||Update Dissertation|
|Email:||click here to access email|
|Degree Awarded:||University of Michigan , Department of Linguistics|
Theoretical approaches to the principles governing the coordination of
speech gestures differ in their assessment of the contributions of
biomechanical and perceptual pressures on this coordination.
Perceptually-oriented accounts postulate that, for consonant-consonant
(C1-C2) sequences, gestural timing patterns arise from speakers'
sensitivity to the listener's need to perceptually recover the input,
whereas biomechanically-oriented accounts focus on physical factors that
might constrain the relevant articulators. This dissertation contributes
to current understanding of gestural coordination by examining the
influences of order of place of articulation (front-to-back,
back-to-front), manner of C1 (plosive, fricative), and manner of C2
(plosive, fricative, lateral) on the timing of constrictions formed by the
tongue tip, tongue dorsum, and lips. If speakers produce CC sequences in
order to accommodate listeners' needs, temporal separation between C1 and
C2 is expected in contexts in which acoustic masking due to intergestural
overlap is especially likely. If speakers' productions are instead
directed by physical limitations of the vocal tract, overlap should be
reduced when the gestures for C1 and C2 are not independent.
Specific instantiations of these broad hypotheses were tested in a
production experiment in which eight Greek speakers' productions of initial
CC sequences [pt ps pl ft kt ks kl xt] were imaged using ultrasound and
video camera technologies. Degree of gestural overlap was measured in
terms of temporal lag between the release of C1 constriction and the
achievement of C2 constriction. Although perceptual-recoverability and
biomechanical accounts made similar predictions for the effect of place
order, they differed in their predictions for effects of C1 and C2 manner
in the two place orders.
Results showed that, consistent with biomechanics, dorsal-coronal [kt ks kl
xt] were produced with greater intergestural lag than labial-coronal [pt ps
pl ft]. Consistent with perceptual recoverability, plosive-plosive [pt kt]
were produced with longer lag than fricative-plosive [ft xt]. An outcome
not clearly predicted by either hypothesis was that lag was longer in [pt
kt] than [ps ks]. Patterns, especially for plosive-lateral [pl kl], varied
across speakers. These findings revealed an interplay between physical and
perceptual—and potentially language-specific—demands on the timing of
gestural coordination in speech production.