My Views about the Field of Linguistics
The field of linguistics is primarily concerned with characterizing the uniquely human ability to communicate using language and the nature of word and sentence structure in the world's languages. An important scholarly goal of linguists is to provide a theoretical account of the human cognitive process of language production and comprehension. Fundamentally, such an account will require an understanding of how speakers cognitively represent language in terms of atomic phonological, morphological, and syntactic structures and how these representations relate to articulation and meaning. It will also address how these cognitive structures develop, how they are utilized both individually and interactively in the process of speaking and understanding, and how linguistic interaction between individuals affects the system. Computational modeling provides important tools for testing the viability and cognitive adequacy of linguistic theory.
Phonetics and Interdisciplinarity
Phonetics investigates the articulatory, acoustic, and perceptual properties of spoken language. Spoken language relies on an elegant and complex acoustic structure to support communication between speakers and listeners. Consideration of the orchestration of articulatory activity that produces this acoustic signal is a necessary element in understanding the process of human communication. The complex messages and emotions of spoken language must be communicated by precise choreography of the jaw, tongue, lips, larynx, and respiratory system. This choreography is shaped by the structure of the language being spoken but is constrained by the shared biology of the human organism producing the linguistic message.
Phonetics is an interdisciplinary area of research, which in addition to requiring an understanding of linguistics, incorporates perspectives of cognitive science, motor control, acoustics, physiology, and psychology. I consider my own work in speech production to be part of the growing worldwide interdisciplinary research effort in cognitive science. In my view, the study of language is one of the cornerstones of this effort. It is a special human ability that makes possible the richness of human intellect, relationships, culture, and technology. Language is both public and private. It is a public phenomenon in that the acoustic signal (or optical signal, in the case of signed languages) is transmitted in the environment for the benefit of a perceiver. And, it is a private phenomenon in that it is generated by the thoughts of an individual and realized by movements, often internal, of their body. Because spoken language stands at this crucial interface between cognition and action, it is a fertile field of research for understanding the human cognitive system.
My Research Program
Speech Timing: Empirical Results, Computational Modeling, & Theoretical Paradigm
The long-term objective of my research program is to understand how linguistic structure conditions the temporal realization of speech. As research in speech production becomes more integrated with linguistic theory, it has become increasingly clear that segmental articulation can not be understood independently of prosodic structure; such structure includes, but is not limited to, the domains of syllables, words, and phrases. These high-level prosodic aspects of linguistic structure pervade low-level articulatory behavior. Despite the pervasiveness of these effects, few prosodic "signatures" have been identified at the level of articulatory patterning. Understanding the effects of syllable structure on articulatory coordination and of phrasal structure on temporal patterns in speaking are two areas in which my work has made important progress.
Prosodic Structure: Pursuing the view that dynamically-defined articulatory gestures are the atomic unit of the spoken language representation and of the execution of words in speech (see the work of Browman & Goldstein), the relative timing and the timing stability of these gestures as a function of their syllabic and phrasal position and prominence is investigated in a series of experiments. These data are providing a profile of the way in which articulatory coordination among multiple gestures is shaped by supra-segmental context, ultimately with the hope of defining general signatures of prosodic structure on articulatory organization. This work has led to a theoretical and computational model of phrase boundaries (& potentially phrasal accent) as "prosodic gestures," which, when active, shape the activation and/or parameter dynamics of the constriction gestures that they overlap with in time. (e.g., Byrd & Saltzman, JPhon, 2003)
Phonological Structure: A second aspect of the research program investigates the phonetic and phonological evidence for phonological units at many levels of granularity, ranging from gesture to segment to syllable onset to morphemes and words. This work examines the temporal cohesion and organization of multigesture units ("gestural molecules"), the gradient nature of speech errors, and the participation of a variety of gestural molecules in phonological processes.
Experimental Approach: My research depends on the collection of experimental data on speech production using human subjects. Articulatory data is collected using real-time MRI and the magnetometer to image and track the movements of the tongue, lips, and jaw. Magnetometer data can be collected for two-subjects synchronously to examine the interaction of interlocutors. I also place a priority on data-driven modeling, using in particular the Task Dynamics model developed by Elliot Saltzman.
Additional Interests and Collaborative Projects
Downloadable articles on these topics by me and colleagues can be found at my CV and ResearchGate.
This work has been supported in part by the National Institutes of Health and other funding sources.