(1) The main effect of group is inflated in your model. To test the
main effect of group properly, you need to average your conditions and
then compute a model using the averages. If you only have two groups,
I'd skip the F-test and just do the T-test.
(2) If you have more than 2 groups, you need to do post-hoc t-tests. I
would say that the had greater activation rather than more voxel as
more voxel seems to imply extent, which isn't being tested.
(3) For the interaction, make sure you have subject in your model and
in the design matrix. If you still get no significant results, you can
say that the interaction was not statistically significant or no
voxels passed the significance threshold.
(4) There are a number of ways to determine thresholds. Personally, I
like using AlphaSim in AFNI if possible. For exploratory analysis, you
can lower the voxel extent criteria a bit. In general, most studies
use p<0.001.
Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Postdoctoral Research Fellow, GRECC, Bedford VA
Research Fellow, Department of Neurology, Massachusetts General
Hospital and Harvard Medical School
Office: (773) 406-2464
=====================
This e-mail contains CONFIDENTIAL INFORMATION which may contain
PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED
and which is intended only for the use of the individual or entity
named above. If the reader of the e-mail is not the intended recipient
or the employee or agent responsible for delivering it to the intended
recipient, you are hereby notified that you are in possession of
confidential and privileged information. Any unauthorized use,
disclosure, copying or the taking of any action in reliance on the
contents of this information is strictly prohibited and may be
unlawful. If you have received this e-mail unintentionally, please
immediately notify the sender via telephone at (773) 406-2464 or
email.
On Sun, Feb 20, 2011 at 7:22 PM, Anna McCarrey
<[log in to unmask]> wrote:
> Hi all,
>
> I have a 2x2 flexible factorial design in fMRI and wish to report main effects and the interaction.
>
> When I run the F contrast for Main effect of Group and click whole brain analysis, many voxel clusters appear. So I must follow up with t-tests to understand in which the direction the groups differ. Do I write this as: Group1 had significantly more voxel activation in the (e.g.) amygdala region than Group2 (t(64) = etc. and then report the t and p values?
> If there are many voxel clusters, what's the best way to set a minimum threshold?
>
> Lastly, when I run the Group x Condition interaction, no suprathreshold clusters appear - with no F statistic, how do I formally report this?
>
> Many thanks in advance.
>
> Anna
>
|