Kinematic linkage of the tongue, jaw, and hyoid during eating and speech

Abstract

Objective: Tongue movement is temporo-spatially coordinated with jaw and hyoid movements during eating and speech. As such, we evaluated: (1) the correlation between the tongue with jaw and hyoid movements during eating and speech and (2) the relative influence of the jaw and hyoid on determining tongue movement.

Design: Lateral projection videofluorography was recorded while 16 healthy subjects ate solid foods or read a standard passage. The position of anterior and posterior tongue markers (ATM and PTM, respectively), the jaw, and the hyoid relative to the upper occlusal plane was quantified with the upper canine as the origin (0,0) point for Cartesian coordinates. For vertical and horizontal dimensions, separate multiple linear regression analyses were performed with ATM or PTM position as a function of jaw and hyoid positions.

Results: Vertically, both ATM and PTM positions were highly correlated with the jaw and hyoid during eating (median r=0.87). The relative influence was higher for the jaw than the hyoid for ATM position (P<0.001), but lower for PTM position (P=0.04). Horizontally, tongue marker positions had moderate correlation with the jaw and hyoid during eating (r=0.47), due more to hyoid position than to jaw position. Overall, correlations were lower during speech than eating.

Conclusion: This study demonstrated distinct kinematic linkages between the movements of the jaw, the hyoid and the anterior and posterior tongue markers, as well as differing impact of the jaw and the hyoid in determining tongue movement during eating and speech.

Figure 1

Locations of the jaw, hyoid, and anterior and posterior tongue markers and their positions in Cartesian coordinates. (A) Small lead discs were glued to the dried dorsal tongue surface approximately 10mm posterior from the tongue tip along the midline (anterior tongue marker; ATM) and as posteriorly as possible along the midline (posterior tongue marker; PTM). Radiopaque lead markers (5mm diameter) were also cemented to the buccal surfaces of the left upper and lower canines and upper first molar. (B) The positions of the ATM, PTM, jaw and hyoid (the antero-superior corner of the hyoid bone) were expressed as X (horizontal) and Y (vertical) coordinates relative to the upper occlusal plane defined by a line passing through the upper canine and first molar markers. The vertical axis was defined as the line perpendicular to the upper occlusal plane at the upper canine.

Figure 2

Representative measurements of the movements of the anterior tongue marker (ATM), jaw, and hyoid in the horizontal and vertical dimensions over time. The complete sequence of eating banana is shown on the left, and the reading of a part of the “Grandfather Passage” by the same subject is on the right. Movement at the top of each figure is either forward or upward. During eating of banana, the ATM moves cyclically in horizontal and vertical dimensions. ATM movement was moderately correlated with jaw and hyoid movement in the horizontal dimension (r = 0.45) and highly correlated in the vertical dimension (r = 0.85). In contrast, the amplitude of structure movements was relatively small and movement patterns were more irregular during speech, which are reflected by lower correlation coefficients (r = 0.35 for the horizontal dimension and r = 0.64 for the vertical dimension).

Figure 3

Box and whisker plots showing the average multiple correlation coefficients (r) by behavior (eating each food, speaking) and anterior and posterior tongue markers (ATM and PTM, respectively) in the vertical and horizontal dimensions. In the vertical dimension, ATM and PTM positions were highly correlated with jaw and hyoid positions during eating, with the average r higher for ATM than for PTM. Vertical correlations were higher during eating than speaking. In the horizontal dimension, the ATM and PTM positions were moderately correlated with jaw and hyoid positions in the multiple regression model.

Figure 4

Box and whisker plots showing the average standardized partial correlation coefficients (β) by factor (jaw and hyoid), behavior (eating each food, speaking), and anterior and posterior tongue markers (ATM and PTM, respectively) in the vertical and horizontal dimensions. In the vertical dimension, jaw β was higher for ATM position during eating, but hyoid β was higher for PTM position. In the horizontal dimension during eating, β was higher for the hyoid than the jaw for both ATM and PTM positions. During speech, β was generally low except for vertical jaw position.