 |
 |
Hello Mark, Thanks a lot for your helpful answer, I really appreciate it!!!! Just a follow up question: Let's suppose that the stroop task has a baseline condition ( for instance a resting block of 3 sec after each neutral and interference block, or a fixation condition at the beginning and at the end of the task). In that case the most appropriate set up of the "Custom (3 column format)" could be the second option that I explained in my previous email, Am I right? Once again thank you very much Lorena -- Lorena Jimenez-Castro, MD Postdoctoral Fellow Research Imaging Institute University of Texas Health Science Center 8403 Floyd Curl Drive San Antonio, TX 78229 ____________________________________ Subject: Re: fmri set up From: Mark Jenkinson <[log in to unmask]> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Date: Wed, 14 Nov 2012 00:48:03 +0000 Dear Lorena, I would basically do the first option, which treats the neutral as the baseline. This means having one EV, but I would specify it without the rows ending with zeros. That is, the specification of this EV is just like the second one (EV2) in the second case. Then you just have a +1 contrast for incongruent > neutral and a -1 contrast for neutral > incongruent. There are no other contrasts that you can estimate if you have no baseline condition. All the best, Mark On 13 Nov 2012, at 20:02, Lorena Jimenez-Castro <[log in to unmask]> wrote: > Hello, > > Thanks for the answer but I don't quite understand what you mean. I am not talking about groups in a higher level. My question is about the first level analysis, specifically the full model setup choosing of "Custom (3 column format)". Please any hint on this would be appreciated! > > Thanks > > Lorena > > > > > > Hi, > It depends of what you are looking for. 1. just one group with one condition without intercepts at the slopes; 2. two groups, with two conditions = modeling different events in different columns and looking for intercepts in the slopes,... > Cheers, > Rosalia. > El 13/11/2012 20:34, "Lorena Jimenez-Castro" <[log in to unmask]> escribió: > Hello, > > I am re-posting my question. Any help would be greatly appreciated. > > "Hello Fsl users and experts, > > > I will analyze a Counting stroop task starting with a first level in FEAT. In this experimental subjects completed 5 min of alternating 30-sec blocks of neutral and Interference (incongruent) trials. The task did NOT include baseline condition (resting/fixation) > So, I was wondering which of the following designs are more suitable for analyzing my data and Why? > > A) Modeling the stroop effect (incongruent > neutral) in just one EV: > Where 0s in the third column are neutral condition and 1s are incongruent condition > > 0 30 0 > 30 30 1 > 60 30 0 > 90 30 1 > 120 30 0 > 150 30 1 > 180 30 0 > 210 30 1 > 240 30 0 > 270 30 1 > > Contrast: > incongruent > neutral: 1 > > OR > > 2) Modeling each of condition with a separate EV, > > neutral (EV1): > 0 30 1 > 60 30 1 > 120 30 1 > 180 30 1 > 240 30 1 > > Incongruent (EV2): > > 30 30 1 > 90 30 1 > 150 30 1 > 210 30 1 > 270 30 1 > > Contrast: > > Neutral mean: 1 0 > Incongruent mean: 0 1 > Neutral > incongruent: 1 -1 > Incongruent > neutral: -1 1 > > Thanks" > > Lorena