Department of Psychology
2019-2020 COLLOQUIUM SERIES
and
The Data Science Institute
Present
Teague Henry, PhD
Postdoctoral Research Associate
Psychology and Neuroscience
University of North Carolina at Chapel Hill
“Localization and Control in Psychological and Brain Networks”
Network science approaches are increasingly used in psychology and neuroscience to analyze both psychological and neural systems as networks. By representing a given psychological or neural set of systems as networks, researchers can assess how the network topology or shape might matter. However, there are two considerations that traditional network analysis tools fail to address in psychological and brain network data. The first is that researchers are often interested in how specific elements of a network (e.g., specific brain regions or psychological symptoms) contribute to the overall network topology. The second is that many networks of interest are representations of complex, dynamic systems (i.e. brain activity, change in psychological symptoms over time). In this talk, I describe these two linked issues and discuss three approaches to account for them. I discuss the Network Statistic Jackknife, a method for localizing the contribution of specific network elements, and illustrate its use within a study on the neural correlates of Attention Deficit/Hyperactivity Disorder (ADHD). Next, I show how network controllability centrality measures can clarify the ways in which specific network nodes may modulate the underlying dynamic system; I illustrate this approach in a study of methylphenidate administration on children with ADHD. Finally, I show how optimal network control methods can be used to evaluate the potential impacts of interventions on psychological systems using a treatment study of patients suffering from complicated grief. Future directions for research in both localization and control methods will be discussed.
Monday, March 2, 2020
3:30 p.m. Gilmer 190
Coffee/cookies at 3:15pm. Reception will be held after the talk.