-----Original Message-----
From: Stephen J. Fromm [mailto:[log in to unmask]]
Sent: Friday, March 18, 2005 9:01 AM
Subject: Re: Global normalisation

On Fri, 18 Mar 2005 11:44:03 +0800, Wang, Bo <[log in to unmask]
wrote:

dear all,

        how to understand the scale/none option when i link data to
SPM.mat
(fMRI button - data)? in SPM99, it recommended to chose scale, but in
SPM2
i heard of another choice.
thanks!

If you mean whether you should do global scaling in fMRI analysis, the
current consensus is that you shouldn't (though the verdict on this isn't
unanimous).

Wang, Bo

>>>
To echo an email a few weeks back - the below manuscripts discuss the issue
of within session scaling (or global scaling).

While I am not aware of any dissenting opinions on this issue, I think what
is important is that you report in your methods exactly what you did in your
spm analyses.  If you do decide to use 'scaling' then you should report it
in the manuscript, and it is recommended (see below manuscripts) that you
report the correlation between your global signal and your regressors.  This
allows one to evaluate the potential influences of scaling on your data.

Kalina Christoff has some code on her old website to do this (as well as a
nice description of the problem).
http://www-psych.stanford.edu/~kalina/SPM99/Tools/rd_global.html

Best regards,
Kent


Macey,PM, et al (2004) A method for removal of global effects from fMRI time
series. NeuroImage 22. 360-366.

 Aguirre, G. K., Zarahn, E., & D'Esposito, M. (1998). The inferential
 impact of global signal covariates in functional neuroimaging
 analyses.  Neuroimage,  8(3), 302-306.

 Andersson, J. L. (1997). How to estimate global activity independent
 of changes in local activity. Neuroimage, 6(4), 237-244.

 Andersson, J. L., Ashburner, J., & Friston, K. (2001). A global
 estimator unbiased by local changes. Neuroimage, 13(6 Pt 1),
 1193-1206.

 Desjardins, A. E., Kiehl, K. A., & Liddle, P. F. (2001). Removal of
 confounding effects of global signal in functional magnetic resonance
 imaging analyses. Neuroimage, 13, 751-758.


 ------------------------------------------------------
 Kent A. Kiehl, Ph.D., Director
 Clinical Cognitive Neuroscience Laboratory Olin Neuropsychiatry
 Research Center, Institute of Living

 Assistant Clinical Professor
 Department of Psychiatry
 Yale University School of Medicine

 Mailing address and contact information:

 Institute of Living
 200 Retreat Avenue
 Olin Neuropsychiatry Research Center
 Hartford, CT USA 06106
 Tel: 860-545-7385; Fax: 860-545-7797
 Email [log in to unmask]
 http://www.nrc-iol.org/ccnlab