Christian - to the degree that there is low-frequency noise in the
data (which there always is in fMRI data), this noise will impact
your signal estimates at the first level (as well as affecting the
variance estimates, as you mention), and thus introduce additional
noise variance into the second-level analysis if you don't remove it
using high-pass filtering
cheers
russ
On Mar 11, 2006, at 2:42 AM, Christian Keysers wrote:
> Dear SPM community,
>
> We use fMRI designs where the interval between two repetitions of a
> certain condition is often around half the length of a session.
> Till now, we have generally calculated the maximum interval, added
> approximately 10% to that, and used that setting as the high-pass
> filtering.
>
> Since we usually concentrate on RFX second level analysis, I'm
> actually wondering what the impact of high-pass filtering is at the
> second level? If all it does is reduce the unexplained variance at
> the first level, this should not affect the con images that flow
> into the second level analysis. Is then not using any high-pass
> filtering the cleanest approach?
>
>
> Christian
>
> --
> Christian Keysers, PhD
> Assistant Professor
>
> BCN Neuro-Imaging Center
> University of Groningen
> Antonius Deusinglaan 2 (room 120)
> 9713 AW Groningen
>
> Phone: +31 50 3638794
> Fax: +31 50 3638875
---
Russell A. Poldrack, Ph.d.
Associate Professor
UCLA Department of Psychology
Franz Hall, Box 951563
Los Angeles, CA 90095-1563
phone: 310-794-1224
fax: 310-206-5895
email: [log in to unmask]
web: www.poldracklab.org
|