May 2018

With nearly a year of training, testing, data collection, and analysis behind us, the Multimodal Neuroimaging Initiative is beginning to wrap up our first wave of research projects. In recent months, our teams have submitted abstracts, presented at conferences, and defended theses. Read on for more details about three projects that are on their way out to the scientific community this year!

This issue of the newsletter will be our last for the academic year, and we wish you all a safe and productive summer! As always, we hope you'll stay in touch and let us know about your own news and publications via Twitter ( @UT_MultiNeuro ) or the website . We look forward to reporting on lots of summer research and activity in September.


Tunc Akbas, Rachel Reetzke, and Ben Zinszer
Your Multimodal Neuroimaging communications team
In this newsletter...
Academic Year 2017-18: Project Updates
MultiNeuro Talk at Health Science Meeting in Mexico
Announcements | Resources | Publications
Academic Year 2017-18: Project Updates
Adaptive Neural Controller for Motor Rehabilitation Tasks
Electrical and Computer Engineering graduate student Allison Berman presented her Master's thesis project at the CARE (Clinically Applied Rehabilitation Research and Engineering) research day on April 13th with co-authors Ethan Oblak, Prof. Jarrod Lewis-Peacock (Psychology) and Prof. James Sulzer (Mechanical Engineering). Their study uses fNIRS to guide motor task instructions that modulate sensorimotor brain activity to a desired level, ultimately leading towards a system to restore function following stroke.
The adaptive neural controller monitors brain activity in real-time and updates the task instructions after every trail.
The participant performs an isometric precision grip task with the force sensor (left) while their brain activity is recorded with fNIRS (right).
The goal of their study was to control the dosage of activity during neurofeedback with an adaptive controller. The adaptive controller (depicted on the left) is tested using an precision grip task (depicted on the right). The controller instructs participants to pinch at varying forces at every trial to help participants achieve a desired level of brain activity, compared to pinching at a constant, predetermined level of force. Berman and her colleagues are currently validating this model across multiple participants.
Blood Oxygenation During and After Physical Activity
In April, graduate students Yeonhak Jung and Brett Baker, along with co-authors Dakota Skinner, Preeti Chopra, and Prof. Darla Castelli (Kinesiology and Health Education) submitted two abstracts to the 2nd International Neuroergonomics Conference. These projects, Comparison of Behavioral and Cerebral Hemodynamic Responses to Standing and Walking Conditions, and Executive Function and Cerebral Hemodynamic Responses Following an Acute Bout of Physical Activity, examined behavioral performance and changes in cortical blood oxygen concentration during different types of physical activity and task demands.
In the first project, Jung and colleagues used fNIRS to measure blood oxygenation in the prefrontal cortex during four types of physical activity that varied in complexity from standing to walking. In the second project, Baker and colleagues used fNIRS to measure blood oxygenation in the prefrontal cortex during three different cognitive tests. The participants completed these tests before and after physical activity conditions that varied in complexity to examine the effect of exercise on executive function. These projects highlight the utility of fNIRS to examine neural correlates of physical activity and related cognitive processes.
Pictured right: A participant walks on the treadmill while performing a cognitive task in the virtual reality environment.
Speech Sound Processing with fNIRS+EEG
The SoundBrain Lab and Applied Research Laboratories submitted two conference abstracts on a multimodal imaging experiment with fNIRS and EEG to study processing of Mandarin lexical tones. Dr. Benjamin Zinszer (Multimodal Neuroimaging Initiative) and colleagues Dr. Todd Hay (ARL), Dr. Alex Athey (ARL), and Prof. Bharath Chandrasekaran (Communication Sciences and Disorders) used machine learning models to identify patterns in brain activity for each of three lexical tones applied to the same syllable.

In their abstract submitted to the Society for the Neurobiology of Language, they applied a multilayer perceptron model to identify which fNIRS channels were most important for predicting which tone a participant was hearing, supporting the role of left anterior brain regions (i.e., frontal lobe) in discriminating between similar speech sounds. In a second abstract to the 2nd International Neuroergonomics Conference they compared the quality of the frequency-following response (FFR, a primarily subcortical signal measured with EEG) with fNIRS tone classification accuracy, suggesting a connection between participants' early auditory processes (FFR) and later anterior cortical activity (fNIRS).

Their preliminary findings support the proposal that EEG and fNIRS measures can mutually inform one another in discriminating between similar speech sounds and sets the stage for further language research with MRI-incompatible populations, such as children and cochlear implant users.
MultiNeuro Talk at Health Science Meeting in Mexico
The Multimodal Neuroimaging Initiative's technical director, Dr. Benjamin Zinszer, will present an invited talk on fNIRS research methods at the IX Congreso Nacional de Tecnología Aplicada a Ciencias de la Salud (9th National Congress on Technology Applied to Health Science) on June 15th. This conference, held annually in Puebla, Mexico, brings together scientists, engineers, clinicians, and private sector researchers to foster innovation in health sciences and promote educational opportunities for undergraduate students in these fields. Dr. Zinszer's talk, Multivariate pattern analysis and near-infrared spectroscopy: Decoding the developing brain, will describe the potential for applying the latest methods in functional brain imaging to low-cost, portable, child-friendly neuroimaging tools like fNIRS.
The conference with be held at Benemérita Universidad Autónoma de Puebla in Puebla, Mexico (left). MultiNeuro Technical Director, Benjamin Zinszer (right), will give an invited lecture on multivariate pattern analysis methods for fNIRS research.


  • Neurophotonics, the official journal of Society for fNIRS, will go fully Open Access in January 2019. More information here: 
  • Visit our Resources webpage to view tutorial videos, download training materials, and connect with other users for support.


The Multimodal Neuroimaging Intiative would like to thank:
Office of the Vice President of Research
Moody College of Communication
Cockrell School of Engineering
College of Liberal Arts
College of Education
Applied Research Laboratories
for their generous support!