Okay got it,
For 2nd level, 1 subject, 3 sessions, I use ME stage 1 (to make sure the
between session variance is modeled; that's what I did).
For 2nd level, 1 subject, 1 session, I can run a fake 2nd level using
fixed-effects by manually changing the number of inputs in the fsf to 1.
Tx, Theo
on 10/22/05 1:43 AM, Stephen Smith at [log in to unmask] wrote:
> Hi - I presume you mean for combining, at second level, across difference
> sessions for a given subject - yes, ME stage 1 is the best thing to use
> for that (see the FEAT manual on the web for examples of how to do this if
> you only have a few sessions per subject).
>
> Cheers.
>
>
> On Fri, 21 Oct 2005, Theo van Erp wrote:
>
>> Would running "Mixed Effects Stage 1 only" lead to incorrect results when
>> used for combining the single subject runs (maybe only to be used when
>> combining runs from multiple subjects with run as a repeated and subject as
>> random factor)?
>>
>> Tx, Theo
>>
>> on 10/17/05 7:46 AM, Stephen Smith at [log in to unmask] wrote:
>>
>>> Hi Theo, yes, this all makes sense:
>>>
>>> You can either follow the mixed inputs instructions at:
>>> http://www.fmrib.ox.ac.uk/fslfaq/#feat_mixedup
>>>
>>> Or just run a "fake" 2nd level on these single cases using fixed-effects
>>> (though you'll need to save the design.fsf and edit it by hand to change
>>> the number of inputs to 1).
>>>
>>> Cheers. Steve.
>>>
>>>
>>>
>>> On Thu, 13 Oct 2005, Theo van Erp wrote:
>>>
>>>> Hi,
>>>>
>>>> I have another question related to empty EVs. I'm working on a single
>>>> trial
>>>> study with multiple runs and analyses based on subject responses.
>>>>
>>>> Sometimes there are missing EVs because a subject did not make a certain
>>>> type of response. This is not a problem, since as suggested below I then
>>>> leave out the empty Evs and leave the contrast order the same. I put zeros
>>>> for the non-existent contrasts (due to the missing EV) and correct the
>>>> contrasts to account for having fewer EVs. This appears to generate empty
>>>> (0) copes, and I then combine only the non-empty copes in a 2nd level
>>>> analysis to create 1 cope per subject to be included in a 3rd level
>>>> analysis.
>>>>
>>>> However, I'm running into a problem when a contrast only exists in 1 run
>>>> (you can't do a 2nd level, since the dof would be 0). Should I in this
>>>> case
>>>> just register the cope from one individual run to standard space in the
>>>> first level analysis such that I can include it into a 3rd level analysis
>>>> or
>>>> does this somehow mess up the dofs?
>>>>
>>>> I'll try a couple of things, but any feedback would be appreciated.
>>>>
>>>> Best, Theo
>>>>
>>>> on 6/5/04 12:17 PM, Mark Jenkinson at [log in to unmask] wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> This depends on what contrasts you are trying to use.
>>>>> If you have a contrast that only contains "missing" EVs then
>>>>> you can't have this if you drop these EVs. However, I'm
>>>>> assuming that the contrasts of interest for the group analysis
>>>>> are not going to include only missing EVs for any subject or
>>>>> session. Otherwise there is no information that this subject
>>>>> or session can contribute to the group analysis in this case
>>>>> and just shouldn't be included for the higher level analysis.
>>>>>
>>>>> If you do make contrasts to see certain effects in the lower
>>>>> level analysis that you don't pass up to the group analysis
>>>>> (e.g. a 0 0 1 0 0 ... 0 type contrast for each EV) then it doesn't
>>>>> matter what you put in this place for the lower level analysis
>>>>> as far as the group analysis is concerned.
>>>>>
>>>>> The best approach for you is probably to restrict yourself to
>>>>> first level analyses that only contain the contrast of interest
>>>>> for the group analysis. This should include at least one
>>>>> non-missing EV for each session/subject (if not, exclude
>>>>> that session/subject from the group analysis). Once you've
>>>>> done this the group analysis is straightforward.
>>>>>
>>>>> If you are interested in looking at other contrasts in the lower
>>>>> level, you can always run different contrasts again under
>>>>> post-stats to have a look (which is quick and easy).
>>>>>
>>>>> All the best,
>>>>> Mark
>>>>>
>>>>>
>>>>> On Thursday, June 3, 2004, at 10:42 pm, Sam Harris wrote:
>>>>>
>>>>>> Hi, Mark
>>>>>>
>>>>>> Thanks for your response. It seems to me, however, that I'm not out of
>>>>>> the woods yet--because once I drop the missing EVs for any given run,
>>>>>> that run will then not have the same number of contrasts.
>>>>>> Have I misunderstood you somewhere?
>>>>>>
>>>>>> Best,
>>>>>> Sam
>>>>>>
>>>>>> On Jun 3, 2004, at 11:15 AM, Mark Jenkinson wrote:
>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> What you really need to do is create consistent *contrasts* not EVs.
>>>>>>> It is contrasts that get fed up into the higher level analyses, not
>>>>>>> EVs.
>>>>>>> If an EV is missing, then do not have it in the model. However, when
>>>>>>> formulating your contrasts make sure that the numbering of the
>>>>>>> contrasts
>>>>>>> is the same in each subject/session so that these feed up
>>>>>>> consistently.
>>>>>>>
>>>>>>> You'll have to be careful to select the appropriate EVs each time, as
>>>>>>> their numbering will change whenever some are missing.
>>>>>>> For example, if you had a contrast like 1 0 -1 0 0 1 0 and in one case
>>>>>>> EV2 was missing then the design matrix would only have 6 columns,
>>>>>>> not 7, and the contrast would become 1 -1 0 0 1 0 in this
>>>>>>> instance (assuming the other EVs were present).
>>>>>>>
>>>>>>> Obviously these missing EVs cannot contribute to the contrast, but
>>>>>>> as long as each contrast contains at least one EV then you are fine.
>>>>>>>
>>>>>>> All the best,
>>>>>>> Mark
>>>>>>>
>>>>>>>
>>>>>>> Sam Harris wrote:
>>>>>>>
>>>>>>>> I'm attempting to analyze event-related data acquired on 14 subjects,
>>>>>>>> 3 runs each, in which
>>>>>>>> certain EVs, in any given run, were not represented. For example,
>>>>>>>> subject #6, in run #2, may not
>>>>>>>> have provided a single example of EV 8 (while all other subjects did,
>>>>>>>> as did subject #6 himself in
>>>>>>>> runs 1 and 3). I'm wondering if there is any way of creating a
>>>>>>>> "place-holder" model (in three-
>>>>>>>> column format), so that all subjects and all runs can contain the
>>>>>>>> same number of EVs, thereby
>>>>>>>> allowing analysis at the group level. Could I, in the above case,
>>>>>>>> create a spurious model for EV 8
>>>>>>>> (in subject #6, run #2), with a fictional onset time and duration,
>>>>>>>> and scale it at (or near) 0? Or
>>>>>>>> could I create an extra volume, tacked onto to the end of all scans,
>>>>>>>> that represented the average
>>>>>>>> value of each functional run, and then reference this timepoint in my
>>>>>>>> model as a dummy-EV?
>>>>>>>> Needless to say, I'm looking for a solution that will produce, in the
>>>>>>>> worst case, a type 2 error.
>>>>>>>>
>>>>>>>> Thanks for your help.
>>>>>>>> Sam
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>
>>>>
>>>> _____________________________________________________________
>>>> Theo van Erp
>>>> Lab Manager, PhD Candidate
>>>>
>>>> Cannon Lab
>>>> Department of Psychology [log in to unmask]
>>>> University of California Los Angeles voice (310) 794-9673
>>>> 1285 Franz Hall, room 5556 fax (310) 794-9740
>>>> Los Angeles, California, 90095-1563
>>>> http://www.bol.ucla.edu/~vanerp
>>>> _____________________________________________________________
>>>>
>>
>> Best, Theo
>>
>> _____________________________________________________________
>> Theo van Erp
>> Lab Manager, PhD Candidate
>>
>> Cannon Lab
>> Department of Psychology [log in to unmask]
>> University of California Los Angeles voice (310) 794-9673
>> 1285 Franz Hall, room 5556 fax (310) 794-9740
>> Los Angeles, California, 90095-1563
>> http://www.bol.ucla.edu/~vanerp
>> _____________________________________________________________
>>
Best, Theo
_____________________________________________________________
Theo van Erp
Lab Manager, PhD Candidate
Cannon Lab
Department of Psychology [log in to unmask]
University of California Los Angeles voice (310) 794-9673
1285 Franz Hall, room 5556 fax (310) 794-9740
Los Angeles, California, 90095-1563
http://www.bol.ucla.edu/~vanerp
_____________________________________________________________
|