voxels activated

Dear SPMers,

Please forgive this very basic question; I am new to SPM.

I have looked through the SPM99 manual and through the e-mail archives, and
though I see much discussion on the issue, I still do not feel I know the answer:

When looking at the Volume Summary in the Statistics Table results in SPM99, is
there an easy way to determine the number of voxels surviving the statistical
threshold?  I often see number of voxels activated used as a measure, but I have
not seen how this number is derived.

If I’m reporting the corrected p level at the voxel-level, does the
cluster-level k tell me about number of voxels “activated”?

Many thanks in advance...

Rebecca

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Marsbar-users mailing list
Marsbar-users <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/marsbar-users

(hopefully) simple error message question

Dear Matthew and list,

I have a number of what I think must be relatively simple questions pertaining to MarsBaR 0.38.

First of all, the tutorial says, when setting the design from file, to choose the appropriate spm.mat file. 
However, the default choice seems to be the SPMcfg.mat file.  I would think the SPMcfg. Mat file would have
all the necessary information, but I’m not sure.  Please clarify.

Second, when I do a MarsBaR analysis on my group results (using the SPMcfg.mat file), it appears to work just
fine, but when I try to get ROI contrasts for any of my individual data, I get the following error:

Design reporting                        : ??? No appropriate methods for function ui_report.

Error in ==> C:\MATLAB6p5\marsbar-0.38.2\marsbar.m
On line 881  ==> ui_report(marsD, 'DesMtx');

Error in ==> C:\MATLAB6p5\marsbar-0.38.2\marsbar.m
On line 869  ==> marsbar('design_report');

Error in ==> C:\MATLAB6p5\spm99\spm.m
On line 1182  ==> evalin('base',CBs{v-1})

??? Error while evaluating uicontrol Callback.

Any thoughts???

Many thanks in advance.

Susan McLaughlin
Dept. of Speech and Hearing Sciences

MarsBar on Spm

Dear experts,

I want to do a ROI analysis using marsbar. My question is very simple: 
Can I run the 1-st and the 2-nd level analyses with spm2 and apply the
structural masks for my two regions of interest to my spm.mat (2-nd level
analysis) already estimated by SPM so I ask MarsBaR to extract the voxel
information in the space of the masks without re-analyze time-series data
with marsbar?   

If yes, Can I follow the steps in the "MarsBar tutorial pag. 30 Using a
structural ROI"?

Thank you.

--

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.7/409 - Release Date: 04/08/2006

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: MarsBar on Spm

Hi,

> I want to do a ROI analysis using marsbar. My question is very simple:
> Can I run the 1-st and the 2-nd level analyses with spm2 and apply the
> structural masks for my two regions of interest to my spm.mat (2-nd level
> analysis) already estimated by SPM so I ask MarsBaR to extract the voxel
> information in the space of the masks without re-analyze time-series data
> with marsbar?

Is this covered by the FAQ?

http://marsbar.sourceforge.net/faq.html#rfx

Best,

Matthew

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: (hopefully) simple error message question

Hi,

> First of all, the tutorial says, when setting the design from file, to choose the appropriate spm.mat
file.  However, the default choice seems to be the SPMcfg.mat file.  I would think the SPMcfg. Mat file would
have all the necessary information, but I'm not sure.  Please clarify.

The SPMcfg mat file, as far as I remember it, does not include the
images, so the SPM.mat file would probably be better - maybe try
that...

> Second, when I do a MarsBaR analysis on my group results (using the SPMcfg.mat file), it appears to work
just fine, but when I try to get ROI contrasts for any of my individual data, I get the following error:
>
> Design reporting                        : ??? No appropriate methods for function ui_report.
>

Hmm - are there no warnings previously?  Does this happen if you use
the SPM.mat file?

Best,

Matthew

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Import matlab matrix as data

Dear experts,

Thanks to Matthew I've red in the marsbar FAQ that running random effects
analysis I can load the spm random effects design into marsbar, import my
matlab matrix as the data and run the random effects analysis in marsbar. To
do this I follow the steps: 1)Design -> Set design from file, select spm.mat
(for example one sample t-test before defined in spm), 2) Data -> Import
data -> for Import what I set Summary time course(s) for region(s) -> for
Import from I set Matlab workspace (I've loaded before my contrast value
matrix)-> for MATLAB EXPRESSION what do I write?

Are these correct steps?
The spm random effects design that I define in spm and I'm using in marsbar
is only SPECIFIED in spm and it isn't either assigned or estimated because I
assign data and estimate it in marsbar, is this correct?

Thank you so much.

--

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.7/409 - Release Date: 04/08/2006

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Multiple ROIs

Dear experts, 

I think that if I've more than one roi it's better run separately one
analysis for each roi rather than one analysis for all rois to avoid pooling
covariance estimate and pooling autocorrelation estimate. Is this correct?

For example if I've 3 rois then I've 3 second level analyses, one for each
roi. Can I get corrected p-values for each roi dividing each uncorrected
p-value for the ROI number (3 in this case)?

Thank you.

--

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.7/410 - Release Date: 05/08/2006

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Re: Multiple ROIs

Hi,

> I think that if I've more than one roi it's better run separately one
> analysis for each roi rather than one analysis for all rois to avoid pooling
> covariance estimate and pooling autocorrelation estimate. Is this correct?

Yes, that would be more correct in most cases I think.

> For example if I've 3 rois then I've 3 second level analyses, one for each
> roi. Can I get corrected p-values for each roi dividing each uncorrected
> p-value for the ROI number (3 in this case)?

Yes, it's just a simple Bonferroni correction...

Best,

Matthew

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Anova or t-tests for single roi

Dear Matthew and dear All,

My design matrix has four regressors of interest (my experimental dedsign is
a 2x2 factorial within subjects) convolved with canonical hrf. At 1-level
analysis I think to use a t-contrast to test the two main effects ([1 1 -1
-1]; [(1 -1 1 -1)] and the interaction [1 -1 -1 1]). At 2-level I think to
use three one sample t-tests setting [1] and three one sample t-test setting
[-1].
If I want consider all the effects (main and interaction) in both directions
for a roi is this the correct manner or shall I use an ANOVA at 2-level?    
I know that in a roi analysis I apply a Bonoferroni correction for the
number of rois considered. But shall I applay also a correction for the
number of t-test (six in this case) I do on a single roi?

Thank you in advance. 

--

-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.7/410 - Release Date: 05/08/2006

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Re: questions about event time course

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

_______________________________________________
Marsbar-users mailing list
Marsbar-users <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/marsbar-users