The non-directional nature of the F-test means that the direction of
the contrast (e.g. having a high beta in condition 4, relative to
condition 1) shouldn't effect the results.
If you multiple the threshold by 2 (e.g. .001 to .002) then you should
get a closer map. The reason this is necessary is that the F-test is
equivalent (1 row) to a two-sided t-test AND spm shows results for a
one-sided t-test. If you repeat your t-test contrast for the canonical
HRF with double the p-value (.002 instead of .001) you will get
identical results to the t-test.
Also, it would be good to investigate t-test of the TD term as well.
If there is no difference in the TD term, it might be weakening the
significance.
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 Thu, Sep 9, 2010 at 10:35 AM, Watson, Christopher
<[log in to unmask]> wrote:
> That looks correct. Look at the beta values for those conditions. Maybe e.g. the parameter estimate for the TD for condition 4 is high, making the con value close to 0.
>
> ________________________________________
> From: SPM (Statistical Parametric Mapping) [[log in to unmask]] On Behalf Of megha sharda [[log in to unmask]]
> Sent: Thursday, September 09, 2010 2:30 AM
> To: [log in to unmask]
> Subject: [SPM] Specifying F contrast for hrf with time derivative
>
> Hi all,
>
> I'm trying to analyse data from a sparse sampled event related design. I have four conditions for which I'm modelling the canonical hrf with time derivative. When I simply model the hrf (using t-test in contrast) I get some activations. However, when i'm trying to use the hrf with time derivative, I get no activations at all. I specified the F-contrast using two vectors - [ 1 0 0 0 0 0 -1 0]; [ 0 1 0 0 0 0 0 -1] to look at the contrast between condition 1 and condition 4. (c1-c4). Any suggestions?
>
> Thanks!
> Megha
>
> --
> Megha Sharda
> Graduate Student
> National Brain Research Centre
> Manesar, Gurgaon -122050
> India
> Ph: +919810943202
>
|