By PHILIP HADLEY
Larry Price, a researcher in the college of education at Texas State University, has developed a new graphical modeling and statistical analysis method. This breakthrough will allow neuroscientists to unlock the hidden meanings buried in the complex data collected during imaging studies of the human brain.
“This work is extremely practical in education, psychology, and the biomedical and health sciences,” Price said. “If we can better understand how the brain functions both regionally and system wide we can diagnose and treat people more accurately.
“A more accurate diagnosis can lead to understanding and treating conditions that impede a child’s ability to learn or to better understand how the human brain functions in children with conditions such autism,” he said. “This analytic method can also help individuals with traumatic brain injuries by leading to more effective rehabilitation therapies.”
A gap exists in the methods available to brain scientists to discern the obscured meanings in the large and often complex data sets obtained from brain imaging. Neuroscientists at the University of Texas Health Science Center-San Antonio’s Research Imaging Center turned to Price for assistance in solving this problem.
“The rapid advances in imaging technologies have provided neuroscientists with new tools to acquire data that can then be used in statistical models to explore the role of various regions of the brain and determine how these regions work together when an individual conducts a specific task or processes particular information,” said Price, a professor and consultant in biometrics, psychometrics and statistics.
Over a four-year collaboration, Price designed and tested structural equation models and developed several new methods, one of which is titled “Neuroimaging Network Analysis using Occam’s Window.”
“This work advances the theory in an area where little or no previous theory existed,” Price said. “It also provides a practical method for applying the theory by creating a graphical model of the dynamic connectivity among the different regions of the human brain.”
Price presented the new method at the annual meeting of the American Psychological Association in San Francisco last August. He presented a second paper titled “Modeling Dynamic Functioning Neuroimaging Data Using Structural Equation Modeling: A Monte Carlo Study,” at the American Educational Research Association’s special interest group on Structural Equation Modeling last April in Chicago.
Price and his colleagues, Peter Fox and Angela Laird, have two papers under review at the top neuroimaging journal Neuroimage and at Structural Equation Modeling: A Multidisciplinary Journal in an effort circulate these novel approaches to the neuroscience community.