Dear Xin,
Will's slides, especially slide 19, might help understanding the whitening:
http://www.fil.ion.ucl.ac.uk/spm/course/slides09/02_General_Linear_Model.ppt
That said, for a paired t-test, I would set "Equal Variances" (that's
the SPM8 default).
Best regards,
Guillaume.
Xin Chen wrote:
> Thanks McLaren!
> I found a paper: Correcting for non-sphericity in imaging data using
> classical and Bayesian approaches
> <http://apps.isiknowledge.com/full_record.do?product=UA&search_mode=GeneralSearch&qid=1&SID=3D1mJp56p65no2aMEf3&page=1&doc=1&colname=WOS&cacheurlFromRightClick=no>
> Glaser, D E; Penny, W D; Henson, RN, et al. NEUROIMAGE V13(6)
> S127-S127 2001
> It mentioned this problem but I am not sure. I am new to SPM.
> Anyhow, if this correcting won't hurt the analysis I won't have problems
> with it.
> Thanks again.
>
> Xin
>
> On Tue, Apr 6, 2010 at 3:07 PM, MCLAREN, Donald
> <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
> Since standard statistics use a different formula for equal versus
> unequal variance and SPM only is programmed to use the equal
> variance formula, this is a way to account for the unequal variance.
> I'm not sure about the math behind it, but trust that is works.
>
> Best Regards, Donald McLaren
> =================
> D.G. McLaren
> University of Wisconsin - Madison
> Neuroscience Training Program
> Office: (608) 520-0586
> =====================
> 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 (608) 520-0586 or email.
>
>
> On Tue, Apr 6, 2010 at 1:51 PM, Xin Chen <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> Thanks a lot for your reply! Could you please educate me a
> little bit more on this? How should I deal with it? Thanks again!
>
> Xin
>
>
> On Tue, Apr 6, 2010 at 2:44 PM, MCLAREN, Donald
> <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
> This is an effect of setting the conditions to unequal variance.
>
> Best Regards, Donald McLaren
> =================
> D.G. McLaren
> University of Wisconsin - Madison
> Neuroscience Training Program
> Office: (608) 520-0586
> =====================
> 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 (608) 520-0586 or email.
>
>
> On Tue, Apr 6, 2010 at 1:33 PM, Xin Chen
> <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
> Dear Friends of SPM,
>
> I was doing a simple 2nd level analysis, paired T test
> to be specific. The design matrix looks fine before I
> ran the spm5 estimation, please see the attached
> beforeestimate.jpg. However, after the estimation the
> design matrix changed somehow, please see the
> afterestimate.jpg, it turned a little bit gray but not
> clear cut black/white. I am wondering is it just a
> graphic artifact of spm5, or I did something wrong? Or
> spm5 adjusted my original design matrix because of some
> weakness of my data?
>
> Thanks in advance.
>
> Xin
>
>
>
>
> Xin Chen Ph.D.
> Brain and Behavior Discovery Institute
> Medical College of Georgia
>
>
>
>
>
> --
> Xin Chen Ph.D.
> Brain and Behavior Discovery Institute
> Medical College of Georgia
>
>
>
>
>
> --
> Xin Chen Ph.D.
> Brain and Behavior Discovery Institute
> Medical College of Georgia
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|