Croake Daniel J, Andreatta Richard D, Stemple Joseph C
University of Kentucky, Lexington.
J Speech Lang Hear Res. 2018 Mar 15;61(3):479-495. doi: 10.1044/2017_JSLHR-S-17-0227.
The purpose of this study is to quantify the interactions of the 3 vocalization subsystems of respiration, phonation, and resonance before, during, and after a perturbation to the larynx (temporarily induced unilateral vocal fold paralysis) in 10 vocally healthy participants. Using dynamic systems theory as a guide, we hypothesized that data groupings would emerge revealing context-dependent patterns in the relationships of variables representing the 3 vocalization subsystems. We also hypothesized that group data would mask important individual variability important to understanding the relationships among the vocalization subsystems.
A perturbation paradigm was used to obtain respiratory kinematic, aerodynamic, and acoustic formant measures from 10 healthy participants (8 women, 2 men) with normal voices. Group and individual data were analyzed to provide a multilevel analysis of the data. A 3-dimensional state space model was constructed to demonstrate the interactive relationships among the 3 subsystems before, during, and after perturbation.
During perturbation, group data revealed that lung volume initiations and terminations were lower, with longer respiratory excursions; airflow rates increased while subglottic pressures were maintained. Acoustic formant measures indicated that the spacing between the upper formants decreased (F3-F5), whereas the spacing between F1 and F2 increased. State space modeling revealed the changing directionality and interactions among the 3 subsystems.
Group data alone masked important variability necessary to understand the unique relationships among the 3 subsystems. Multilevel analysis permitted a richer understanding of the individual differences in phonatory regulation and permitted subgroup analysis. Dynamic systems theory may be a useful heuristic to model the interactive relationships among vocalization subsystems.
本研究旨在对10名嗓音健康的参与者在喉部受到扰动（临时诱发单侧声带麻痹）之前、期间和之后，呼吸、发声和共鸣这三个发声子系统之间的相互作用进行量化。以动态系统理论为指导，我们假设数据分组会出现，揭示代表这三个发声子系统的变量之间的依赖于上下文的模式。我们还假设分组数据会掩盖对于理解发声子系统之间关系至关重要的个体差异。
采用扰动范式，从10名嗓音正常的健康参与者（8名女性，2名男性）中获取呼吸运动学、空气动力学和声学共振峰测量数据。对分组数据和个体数据进行分析，以提供对数据的多层次分析。构建三维状态空间模型，以展示扰动之前、期间和之后这三个子系统之间的相互作用关系。
在扰动期间，分组数据显示肺容积起始和终止值较低，呼吸 excursion 更长；气流速率增加，同时声门下压力保持不变。声学共振峰测量表明，上部共振峰之间的间距减小（F3 - F5），而F1和F2之间的间距增加。状态空间建模揭示了这三个子系统之间变化的方向性和相互作用。
仅分组数据掩盖了理解这三个子系统之间独特关系所需的重要变异性。多层次分析有助于更深入地理解发声调节中的个体差异，并允许进行亚组分析。动态系统理论可能是一种有用的启发式方法，用于对发声子系统之间的相互作用关系进行建模。
