Hi Jen,
You're right - we've changed that recommendation (though the temporal
filtering program itself hasn't changed at all). This is because we looked
a little harder at the frequency reponse characteristics of the filter we
use to remove slow drifts in the data. The filter has a fairly slow
roll-off below the cutoff frequency (in order to avoid inducing negative
autocorrelations (or "ringing") in the data) and therefore we feel that it
is safe to move to this slightly more "aggressive" default for the cutoff
frequency. The more slow noise you can remove the more powerful your
stats.
Hope this makes sense, Steve.
On Fri, 1 Nov 2002, Jen Lewis wrote:
> Hello,
> I noticed that the "sensible" setting for the high pass freq cutoff has
> changed from 1.5(A+B)in the last version of FEAT to (A+B)in the current
> version. Would you please tell me how you define this.
> Thanks,
> Jen Lewis
> The Ohio State University
>
Stephen M. Smith
Head of Image Analysis, FMRIB
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|