Hi Doug,
On 28. Aug 2006, at 23:55 Uhr, freesurfer-request@nmr.mgh.harvard.edu wrote:
Date: Mon, 28 Aug 2006 17:53:30 -0400 From: Doug Greve greve@nmr.mgh.harvard.edu Subject: [Freesurfer] Re: newbie question event-related To: Sebastian Moeller sebastian.moeller@lur.rwth-aachen.de Cc: Freesurfer Mailing List freesurfer@nmr.mgh.harvard.edu Message-ID: 44F365DA.1040805@nmr.mgh.harvard.edu Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi Sebastian,
are you using an FIR or gammafit?
Gammafit (-gammafit 0 1.25 -gammaexp 2 in union with -taumax 20 and -dt event-related). To illustrate this, I used the same parameters with "-dt blocked", but my new paradigm is so sparse (200ms timulation every 16 seconds), that event related analysis seems like a good idea. Since we are at it ;), am I right that I should set the TER to 0.2000 and my stimulation epochs to one, while the only fixation periods are set to zero?
Ahoi & Thanks Sebastian
doug
Sebastian Moeller wrote:
Hi List, hi Doug,
for the first time I have now data, that realloy wants event-related analysis (200ms stimulation every 10 seconds). Now, using a TER of 0.2 in the para-file and using "-dt event-related" in mkanalysis-sess.new (3.0.3 pub x86_64 centos4) was the easy part, but what do I have to specify in mkcontast-sess? Especially thi "-ircorr" complex (with -deltarang and -taurange) is daunting. Is it necessary to specify anything here to get a real event-related analysis, or does fs-fast already know enough without these parameters (as I specified delta, tau and even taummax in mkanalysis)? So far any nicorr value I used produced errors on the way... I would be delighted for any pointer of how to properly convinve fs-fast to analyse my data ;).
Ahoi & thanks Sebastian
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
Yes, that's fine. But with gammafit, you cannot use the ircor option (it's only meant for FIR).
Sebastian Moeller wrote:
Hi Doug,
On 28. Aug 2006, at 23:55 Uhr, freesurfer-request@nmr.mgh.harvard.edu wrote:
Date: Mon, 28 Aug 2006 17:53:30 -0400 From: Doug Greve greve@nmr.mgh.harvard.edu Subject: [Freesurfer] Re: newbie question event-related To: Sebastian Moeller sebastian.moeller@lur.rwth-aachen.de Cc: Freesurfer Mailing List freesurfer@nmr.mgh.harvard.edu Message-ID: 44F365DA.1040805@nmr.mgh.harvard.edu Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi Sebastian,
are you using an FIR or gammafit?
Gammafit (-gammafit 0 1.25 -gammaexp 2 in union with -taumax 20and -dt event-related). To illustrate this, I used the same parameters with "-dt blocked", but my new paradigm is so sparse (200ms timulation every 16 seconds), that event related analysis seems like a good idea. Since we are at it ;), am I right that I should set the TER to 0.2000 and my stimulation epochs to one, while the only fixation periods are set to zero?
Ahoi & Thanks Sebastian
doug
Sebastian Moeller wrote:
Hi List, hi Doug,
for the first time I have now data, that realloy wants event-related analysis (200ms stimulation every 10 seconds). Now, using a TER of 0.2 in the para-file and using "-dt event-related" in mkanalysis-sess.new (3.0.3 pub x86_64 centos4) was the easy part, but what do I have to specify in mkcontast-sess? Especially thi "-ircorr" complex (with -deltarang and -taurange) is daunting. Is it necessary to specify anything here to get a real event-related analysis, or does fs-fast already know enough without these parameters (as I specified delta, tau and even taummax in mkanalysis)? So far any nicorr value I used produced errors on the way... I would be delighted for any pointer of how to properly convinve fs-fast to analyse my data ;).
Ahoi & thanks Sebastian
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
Hi Doug,
On 29. Aug 2006, at 00:27 Uhr, Doug Greve wrote:
Yes, that's fine. But with gammafit, you cannot use the ircor option (it's only meant for FIR).
Aah, so, that is simple, no -ircorr at all ;). So the specification for gammafit is done exclusively in mkanalysis-sess, nothing more? One more question, when is the FIR design the better choice than the gammafit design (if that question has an easy answer at all)?
Ahoi & Thanks again Sebastian
Sebastian Moeller wrote:
Hi Doug,
On 28. Aug 2006, at 23:55 Uhr, freesurfer-request@nmr.mgh.harvard.edu wrote:
Date: Mon, 28 Aug 2006 17:53:30 -0400 From: Doug Greve greve@nmr.mgh.harvard.edu Subject: [Freesurfer] Re: newbie question event-related To: Sebastian Moeller sebastian.moeller@lur.rwth-aachen.de Cc: Freesurfer Mailing List freesurfer@nmr.mgh.harvard.edu Message-ID: 44F365DA.1040805@nmr.mgh.harvard.edu Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi Sebastian,
are you using an FIR or gammafit?
Gammafit (-gammafit 0 1.25 -gammaexp 2 in union with -taumax 20and -dt event-related). To illustrate this, I used the same parameters with "-dt blocked", but my new paradigm is so sparse (200ms timulation every 16 seconds), that event related analysis seems like a good idea. Since we are at it ;), am I right that I should set the TER to 0.2000 and my stimulation epochs to one, while the only fixation periods are set to zero?
Ahoi & Thanks Sebastian
doug
Sebastian Moeller wrote:
Hi List, hi Doug,
for the first time I have now data, that realloy wants event-related analysis (200ms stimulation every 10 seconds). Now, using a TER of 0.2 in the para-file and using "-dt event-related" in mkanalysis-sess.new (3.0.3 pub x86_64 centos4) was the easy part, but what do I have to specify in mkcontast-sess? Especially thi "-ircorr" complex (with -deltarang and -taurange) is daunting. Is it necessary to specify anything here to get a real event-related analysis, or does fs-fast already know enough without these parameters (as I specified delta, tau and even taummax in mkanalysis)? So far any nicorr value I used produced errors on the way... I would be delighted for any pointer of how to properly convinve fs-fast to analyse my data ;).
Ahoi & thanks Sebastian
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
freesurfer@nmr.mgh.harvard.edu
-
Doug Greve -
Sebastian Moeller