The tongue is one of the most complex motor systems in the human body. Due to the high number of precisely controlled muscles, different portions of the tongue are able to simultaneously pursue different speech tasks. This leads to very complicated patterns of deformation of the body of the tongue. And it is exactly due to this flexibility that the tongue is able to serve as the major tool for distinguishing and contrasting different consonants and vowels, and for being the most important influence on the dynamic acoustic signal emenating from the vocal tract. Unfortunately our understanding of tongue dynamics in normal and disfluent populations is still at its infancy, because it has been very difficult to observe and measure. What is needed is a device that is able to image as much of the tongue as possible, and at a high temporal rate, since significant movement occurs on the scale of tens of milliseconds. But available instruments for tongue measurement either track only a few points of the tongue at high speed (Electromagnetic Midsagittal Articulography and X-ray Microbeam) or image the shape of the tongue, but only statically (MRI) or at a slow pace (analog-output ultrasound and fast MRI). The instrument we seek to purchase, the Aloka SSD-5000 Digital Ultrasound Machine, allows us for the first time to image the shape of the tongue from blade to hyoid bone at 100-200 images per second, which is a high enough rate to capture the complexity of movement even in the fastest plosive consonants. The arrival of this new generation of digital high speed ultrasound machines will allow us to collect and analyze large amounts of data from a variety of normal and disfluent populations, which will provide an empirical basis for deepening our understanding of the kinematics of the tongue and the processes of motor control that drive the kinematics. Furthermore, this instrument will generate speech production data that will further our research on the speech perception process, since the tongue is the primary influence on the dynamic acoustic signal, and the speech perception system seems to acquire most information from the dynamic portions of the speech signal. We plan for the Aloka SSD-5000 machine to be part of a laboratory setup including an Optotrak machine, which will allow us to measure the position of the tongue in Euclidean space. A collection of Matlab applications will allow us to quantify the images collected by the instrument.
