"How To Be a Skeptical Neuroimager: Functional Connectivity & Causal Modeling"
Workshop at the 2011 Organization for Human Brain Mapping meeeting, Monday 27 June. Organized by Thomas Nichols.
Despite the surge in graphical modeling for functional and structural brain imaging data, often researchers only consider one method and just a pair of competing network models. The purpose of this workshop is to educate neuroimaging users about the "Network of Network Methods" that exists, and show the relationships between the different methods used in imaging and in other disciplines. We also will discuss how to choose the appropriate method for the data and the question at hand, and how to search large space of models that can be fit to your data.
- "The Network of Network Methods, and a Case Study of SEM"
Dr. Martin Lindquist, Columbia University - "The Dangers of Granger: What They Didn't Tell You About Granger Causality"
Dr. Victor Solo, University of New South Wales - "Don’t Fall In Love with your Method: Practical Issues for modeling networks in resting BOLD timeseries"
Dr. Steve Smith, Oxford University - "Don’t Fall In Love with your Model: Model Selection for Graphical Models"
Dr. Joseph Ramsey, Carnegie Mellon University
See also our OHBM 2010 'Skeptical' session.