Hi Todd,
The results between SPM96 (either Matlab or MEDx) and SPM99 can definitely
be different as there have been changes made to the statistics modules
between the two versions. See attached a mail from Karl Friston that
discusses some of the differences one sees.
Also in one of the earlier versions there had been a bug in the SPM96
module which had been subsequently fixed. You could be running into this too
if you haven't upgraded the version of MEDx you are running.
Please go to Help -> About MEDx and send me the version of MEDx that you are
presently running and I could confirm on this.
Regards....Raj
P.S: Also this is to let you know that the latest version of MEDx (vers.
3.4) has spm99 integrated into it.
*********************************************
Raj Jagannathan
Medical Products Division
Sensor Systems, Inc.
http://medx.sensor.com
*********************************************
----- Original Message -----
From: "Todd Lewis" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Wednesday, June 13, 2001 12:42 PM
Subject: MEDx compared to spm99
> Hi,
> Our lab recently obatained a computer to use with spm99. We are currently
> running MEDx with spm96. We ran some data on spm 99 and the results did
not
> match the results attained in MEDx. I used this command line to pull up
the
> old dilogue box for spm96
>
> spm_spm_ui('cfg',spm_spm_ui('DesDefs_PET96'))
>
> design type was Multi-Subject different conditions.
>
> The data was in Picker Prism format and converted to Analyze using MEDx,
it
> was then normalize and smoothed to 10 mm.
>
> Any help would be greatly appreciated.
>
