Abstract: Complex behaviors that humans exhibit - especially sensorimotor behaviors, such as walking or speaking - are intricately choreographed in space and time, requiring the precise engagement of various neural processes. My research is focused on behavioral analytics for neurocognitive assessment, developing novel statistical and computational models that enable machines to find meaningful patterns in human behavior, and thereby gain awareness of human neurocognitive states. Such analytics have applications to noninvasive health assessment technologies, and can also shed light onto the fundamental principles and mechanisms underlying human behavior. Progress toward algorithms that can monitor and understand complex human behaviors is associated with many challenges, most of which center on effectively modeling the myriad variety and sources of behavioral variability exhibited across individuals and contexts. Fresh insights into the development of algorithms that can uncover the mechanisms and sources of variability in human behavior have been recently enabled by an interdisciplinary confluence of empirical, computational and theoretical advances. My research, situated at this confluence, incorporates each of these elements. This talk will present three related efforts in this domain, aimed at (a) cognitive status estimation from vocal biomarkers, based on advanced analytics and machine learning, (b) physiological phenotyping of sensorimotor deficits in traumatic brain injury using immersive virtual environments and combined statistical/mechanistic modeling, and (c) analysis and modeling of speech production variability with magnetic resonance imaging and neurocomputational models [work supported by NIH, NSF and DoD].
Biography: Dr. Adam Lammert is a member of the technical staff in the Bioengineering Systems & Technologies group at MIT Lincoln Laboratory (MIT LL). His research centers on science and technology related to human-centered problems in cognition, motor control, speech and hearing. His research efforts has Principal Investigator have been funded to DoD, NIH and NSF. He has published over 45 peer-reviewed papers in key journal and conference venues, several of which have been awarded special status. Prior to his arrival at MIT, he was Visiting Assistant Professor at Swarthmore College in the Computer Science Department (Swat CS). Before Swarthmore, he completed a Ph.D. at the University of Southern California in the Department of Computer Science, where he was a member of the Signal Analysis and Interpretation Laboratory (SAIL), under the advisement of Dr. Shri Narayanan in the Viterbi School of Engineering and Dr. Louis Goldstein in the Department of Linguistics. He was also an NIH pre-doctoral fellow in the Hearing and Communication Neuroscience program during this time. At USC, he worked on different aspects of speech, language, and human behavior, and his thesis work focused on interdisciplinary problems in speech motor control and auditory perception, which he performed in close collaboration with other members of the Speech Production and Knowledge Group (SPAN). His research efforts at USC were recognized with the Raymond H. Stetson Scholarship in Phonetics and Speech Science, which is given once annually by the Acoustical Society of America, as well as the USC Computer Science Department Best Dissertation Award in 2014 for his dissertation entitled Structure and Function in Speech Production. Immediately before coming to USC, he worked with Dr. Pierre Divenyi as Lab Manager for Speech and Hearing Research at the Veterans Affairs Northern California Health Care System (SHR). His work there was focused on speech production and perception. Prior to that, he studied in the Department of Computer Science at North Carolina State University under the mentorship of Dr. Edward W. Davis toward my Master's thesis. Thesis was focused on automating the design of efficient digital logical circuits using intelligent search techniques (Master's thesis). The journey to his current coordinates began at Vassar College, where he studied Cognitive Science and Computer Science. While there, he worked with Dr. John H. Long, Jr., as a member of his biorobotics and biomechanics laboratory (JHL Lab), toward understanding biologically inspired control of perception-action systems (undergraduate thesis).