Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Statistical Analysis of Functional MRI DataAdditional Statistical Issues

The Statistical Analysis of Functional MRI Data: Additional Statistical Issues In this penultimate chapter, we examine a number of critical statistical issues that can only be fully appreciated from within a broader understanding and knowledge of the analysis methods currently in use for fMRI data. As we have seen in the preceeding chapters, this rich data source, and the fascination of the human brain, have created an attitude among researchers of “I have a hammer, fMRI data look like nails.”. Almost any statistical procedure that one can consider has been brought to bear on some aspect of the analysis stream, whether it be preprocessing, modeling, thresholding, or all three. This is not necessarily a bad thing, as it means that some very smart people have thought about a very hard problem. Still, it is worthwhile to take a step back from the minutiae of implementation, as outlined in the chapters devoted to particular techniques or approaches, and consider some general issues that can be gleaned from the more detailed discussions. The goal of this chapter is to provide that overview. 11.1 Whitening Versus Smoothing One of the main obstacles to easily fitting models to fMRI data is the compli- cated correlation structure, particularly in the spatial dimension. As we http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

The Statistical Analysis of Functional MRI DataAdditional Statistical Issues

Download PDF
19 pages

Loading next page...
 
/lp/springer-journals/the-statistical-analysis-of-functional-mri-data-additional-statistical-YsTNDLixjS
Publisher
Springer New York
Copyright
© Springer-Verlag New York 2008
ISBN
978-0-387-78190-7
Pages
1 –19
DOI
10.1007/978-0-387-78191-4_11
Publisher site
See Chapter on Publisher Site

Abstract

In this penultimate chapter, we examine a number of critical statistical issues that can only be fully appreciated from within a broader understanding and knowledge of the analysis methods currently in use for fMRI data. As we have seen in the preceeding chapters, this rich data source, and the fascination of the human brain, have created an attitude among researchers of “I have a hammer, fMRI data look like nails.”. Almost any statistical procedure that one can consider has been brought to bear on some aspect of the analysis stream, whether it be preprocessing, modeling, thresholding, or all three. This is not necessarily a bad thing, as it means that some very smart people have thought about a very hard problem. Still, it is worthwhile to take a step back from the minutiae of implementation, as outlined in the chapters devoted to particular techniques or approaches, and consider some general issues that can be gleaned from the more detailed discussions. The goal of this chapter is to provide that overview. 11.1 Whitening Versus Smoothing One of the main obstacles to easily fitting models to fMRI data is the compli- cated correlation structure, particularly in the spatial dimension. As we

Published: Jun 7, 2008

Keywords: Receiver Operating Characteristic Curve; Functional Connectivity; fMRI Data; Correlation Approach; Task Data

There are no references for this article.