| Learning in dynamical motor theory is viewed as the modification of pre-existing coordination dynamics or attractor states. The purpose of this study was to address the relationship among voice onset, voice quality, and fundamental frequency during voice motor skill acquisition in order to determine the intrinsic dynamics of the voice production system. Three levels of each independent variable, voice onset (glottal, simultaneous, breathy), voice quality (modal speech, mixed, falsetto), and fundamental frequency (low, mid, and high), were manipulated by vocally untrained, female subjects as they sustained the vowel /i/. The following optimal triads were predicted to emerge: (1) glottal onset/speech quality/low F 0, (2) simultaneous onset/mixed quality/mid F0, and (3) breathy onset/falsetto quality/high F0.; Three dependent measures, % correct of acquisition trials, self-report of effort, and jitter, were used to discriminate more stable phonations indicative of an intrinsic dynamic. Using intensity as a covariate, a 3-way ANCOVA with repeated measures on onset and fundamental frequency F0, was used to analyze data for each dependent variable. Results for % correct of acquisition trials revealed a statistically significant 3-way interaction for F0 x onset x quality [F(8, 131) = 3.10, p = .0031]. Results for self-report of effort revealed a significant main effect for onset [F(2,65) = 3.99, p = .0242] and significant 2-way interaction for F0 x quality [F(4,65) = 23.16, p < .0001]. Results for jitter revealed significant main effects for F0, [F(2,65) = 12.16, p = .0001] and quality [F(2,32) = 16.11, p < .0001].; Although complex, collaborative results of post hoc analyses from all dependent measures provide support for only two attractor states in untrained, female speakers: (1) glottal onset/speech quality/low F0, and (2) breathy onset/falsetto quality/high F0, The absence of evidence for the attractor state, simultaneous onset/mixed quality/mid F0, adds to speculation that practice further from these initial attractor states may lead to the development of a third intrinsic dynamic prevalent among trained singers. Results of this study suggest that certain parameters of voice motor control, onset, quality, and F0, exist as part of a dynamical system that can be identified, manipulated, and exploited in voice motor skill acquisition and learning. |
