I would run 4 models. Each model will have 7 subjects and 1 image per
subject. It will be a two-sample t-test.
Each image will be the difference of time1 and time2 for each of your
four conditions, hence the four models.
In setting up the model for each condition and using the difference
images, you can use FEAT and randomise.
If you were to setup a big model with all conditions, two groups, and
two timepoints, you'd have at least two problems:
(1) you'd have poor estimates of the effects since you have multiple
within-subject factors that increase the probability of violations of
sphericity, which are not corrected in FSL; and (2) you wouldn't be
able to use randomise. Thus, simplifying the model to a series of
paired t-tests since your not looking at the group*time*condition
interactions is probably your best option for the group models.
Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Postdoctoral Research Fellow, GRECC, Bedford VA
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
Office: (773) 406-2464
=====================
This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is
intended only for the use of the individual or entity named above. If the
reader of the e-mail is not the intended recipient or the employee or agent
responsible for delivering it to the intended recipient, you are hereby
notified that you are in possession of confidential and privileged
information. Any unauthorized use, disclosure, copying or the taking of any
action in reliance on the contents of this information is strictly
prohibited and may be unlawful. If you have received this e-mail
unintentionally, please immediately notify the sender via telephone at (773)
406-2464 or email.
On Fri, Feb 10, 2012 at 6:10 PM, Gautam, Prapti <[log in to unmask]> wrote:
>
> The 4 copes are each different conditions in a task and I want to look at
> them separately. Yes, there are two time points.
>
> Prapti
>
> From: Jeanette Mumford <[log in to unmask]>
> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
> Date: Fri, 10 Feb 2012 17:07:23 -0600
>
> To: <[log in to unmask]>
> Subject: Re: [FSL] help with design matrix-longitudinal study
>
> What are your 4 copes? How many time points are there, two? Are you trying
> to relate these 4 copes to each other or are you looking at each one
> separately?
>
> Jeanette
>
> On Fri, Feb 10, 2012 at 12:28 PM, Gautam, Prapti <[log in to unmask]>
> wrote:
>>
>> Hi Jeanette,
>>
>> Thank you for the link. However, I am really new at this, and I still have
>> a few more questions.
>> The model in the example seems to refer to mean group differences between
>> the two groups. While I am interested in that, I would also like to
>> investigate differences between different task levels with time for which I
>> am currently using four copes in the first level analyses. Do you know how I
>> would accommodate these copes in the design matrix and look for differences
>> between the two groups as well as looking at change with time ? Or should
>> these be two different types for analyses? One for group differences and one
>> for time point differences?
>>
>> Thanks,
>>
>> Prapti
>>
>>
>> From: Jeanette Mumford <[log in to unmask]>
>> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
>> Date: Thu, 9 Feb 2012 17:09:57 -0600
>> To: <[log in to unmask]>
>> Subject: Re: [FSL] help with design matrix-longitudinal study
>>
>> I think you'd follow the model described here:
>>
>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0809&L=fsl&P=R7901&1=fsl&9=A&J=on&K=2&d=No+Match%3BMatch%3BMatches&z=4
>>
>>
>> Hope that helps,
>> Jeanette
>>
>> On Thu, Feb 9, 2012 at 4:44 PM, Gautam, Prapti <[log in to unmask]>
>> wrote:
>>>
>>> Dear all,
>>>
>>> I am trying to design a longitudinal experiment looking at group
>>> differences over 2 time points but I am having trouble designing the
>>> contrasts and the design matrices.
>>>
>>> For example, if I had 3 patients and 4 controls in the study, I think
>>> that I can conduct the first and second level analyses separately for the
>>> two groups and then find group level differences in the third level
>>> analyses. For the analyses, I would have 4 copes for the first level, which
>>> could then be fed into the second level, the results of which would then be
>>> fed into the third level.
>>>
>>> However, I am not too sure whether this would be the best way to do the
>>> analyses. I have looked at the examples in the FSL pages for help with
>>> higher level analyses, but I couldn't figure out if the examples applied to
>>> my case.
>>>
>>> If I could get some advice on how to set up the design matrices and the
>>> contrasts, that would be very helpful,
>>>
>>> Thanks,
>>>
>>> Prapti
>>>
>>>
>
>
> ---------------------------------------------------------------------
> CONFIDENTIALITY NOTICE: This e-mail message, including any attachments,
> is for the sole use of the intended recipient(s) and may contain
> confidential
> or legally privileged information. Any unauthorized review, use, disclosure
> or distribution is prohibited. If you are not the intended recipient, please
> contact the sender by reply e-mail and destroy all copies of this original
> message.
> ---------------------------------------------------------------------
|