Forwarding message to list. I await Francesca's reply on whether she
estimated the model.
Karl/John note that there may be an occasional problem with filling in
xVol-- see below.
Darren
>Reply-To: "Dr. Garry Honey" <[log in to unmask]>
>X-Phforward: V2.4@drjimmy (NORTHWESTERN.EDU)
>From: "Dr. Garry Honey" <[log in to unmask]>
>To: "Darren R. Gitelman" <[log in to unmask]>
>Subject: Re: error
>Date: Wed, 8 Jan 2003 15:58:08 -0000
>Organization: University of Cambridge
>X-Mailer: Microsoft Outlook Express 6.00.2800.1106
>X-MIME-Autoconverted: from 8bit to quoted-printable by
>drjimmy.it.northwestern.edu id h08Fw1qQ016786
>X-MIME-Autoconverted: from quoted-printable to 8bit by
>merle.it.northwestern.edu id h08Fw50Q025028
>
>Hi Darren,
>
>Thanks for your input on this. I had actually estimated the model, and I
>tried to estimate it again just to confirm this following Karl's suggestion,
>and accordingly got the warning message: "Current directory contains
>estimation files. Existing results will be overwritten." However, when I
>re-estimated the model (over-writing previous estimation files), the xVol
>and other missing fields were produced, and I was able to go on and specify
>contrasts. I therefore continued to re-estimate models for other subjects,
>and initially presumed that the error related to an earlier post, regarding
>the portability of the results across computers/platforms, due to the
>specification of the complete directory structure in spm.mat, given recent
>changes in our local computing infrastructure. However, I have now done this
>for several subjects, and have found that sometimes Xvol is produced, but
>sometimes not! This is despite the correct estimation of the model, and no
>errors reported in matlab. I had presumed that this may still relate to
>residual issues following the change in our local computing facilities,
>however, perhaps there may be something else wrong, if another user is
>experiencing the same problem in a different institution...
>
>Francesa - I'm presuming you did, but could you just confirm that you had
>estimated the model?
>
>Thanks,
>
>Garry
>
>PS - Darren, please could you forward this to the mailing list, as I'm
>unable to post from my current address - thanks very much.
>
>
>
>
>
>----- Original Message -----
>From: "Darren R. Gitelman" <[log in to unmask]>
>To: <[log in to unmask]>
>Sent: Wednesday, January 08, 2003 3:38 PM
>Subject: Re: error
>
>
>Francesca:
>
>Karl responded to a similar question the other day and suggested the user
>may not have explicitly estimated the model.
>http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0301&L=spm&P=R1476&I=-3
>
>Did you estimate this model?
>
>Darren
>
>
>At 04:01 PM 1/8/2003 +0100, Francesca Sabattoli wrote:
> >Dear Darren,
> >thank you for your answer.
> >
> > > Did you also do the VBM analysis in SPM2?
> >YES, I DID.
> >
> > > Otherwise, if you did analyze it in SPM2 when the analysis finished was
> >there a message such as there are no in-mask
> > > voxels?
> >NO, THERE WAS NOT.
> >
> >When I try to display the results, I see this error massage:
> >??? Reference to non-existent field 'xVol'.
> >Error in ==> C:\MATLABR11\spm2b\spm_getSPM.m
> >On line 163 ==> XYZ = SPM.xVol.XYZ; %-XYZ coordinates
> >Error in ==> C:\MATLABR11\spm2b\spm_results_ui.m
> >On line 260 ==> [SPM,xSPM] = spm_getSPM;
> >??? Error while evaluating uicontrol Callback.
> >
> >In the statistical VBM analysis the file that is saved is only SPM.MAT.
> >
> >Francesca
|