I have a question stemming from this discussion about creating design
matrices. In my study each subject completes a matching paradigm three
times--they complete another paradigm in between each matching paradigm so
there are ~8 mins between matching paradigm sessions and the subject remains
in the scanner the entire time. For one subject I have analyzed each
session individually and I have also created a design matrix with three
identical sessions representing each time the subjects completes the
matching paradigm. If I examine a contrast from one single session within
my design matrix that contains all three sessions the results are different
than if I were to model that one session of interest in a design matrix by
itself. I presume this is because the degrees of freedom are different. Is
that a correct assumption? Are there other reasons the results are slightly
different? In general should I only use the design matrix containing all
three sessions if I am looking at effects across the three sessions or is it
legitimate to look at effects within a single session in a design matrix
containing multiple sessions?
Thanks
Patrick
On Tue, 5 Dec 2006 13:45:19 +0000, Christina Moutsiana
<[log in to unmask]> wrote:
>(-:
>
>Thanks for your reply.
>
>On 12/5/06, PESENTI Mauro <[log in to unmask]> wrote:
>>
>> LLN, le 5/12/06
>>
>> Dear Christina,
>>
>> Cyril is right: concatenation is not recommended. If the scanner
>> stopped and restarted between runs, yet you modelled your conditions
>> as a single session, then your data no longer represent a continuous
>> timeseries, and several aspects of SPM will be disrupted, eg.:
>>
>> - highpass filtering,
>> - temporal autocorrelation estimation,
>> - grand-mean scaling,
>> - session-meaning.
>>
>> (I quote Rik Henson, from this list).
>>
>> Note that I once compared "concatenated runs + constant" to
>> "separated sessions" and the results were quite similar, though not
>> identical.
>>
>>
>> >I will need to reanalyse 15 peoples data . . .).
>>
>> Hmmmm, what are 15 individual analyses in the life of a brain-mapper...?
>>
>> Hope this helps,
>>
>> Mauro
>>
>>
>> >Dear Cyril,
>> >
>> >Thanks for your quick reply.
>> >Each session starts and ends with a 'baseline' condition to prevent
>> >problems due to the scanner starting, and also the first two volumes
>> >are discarded. Do you think that there is still a problem with that?
>> >(well, conditions order was different between sessions, but if there
>> >is a problem with that there is no question I will need to reanalyse
>> >15 peoples data . . .).
>> >
>> >Thanks so much,
>> >Christina
>> >
>> >
>> >On 12/5/06, cyril pernet
>> ><<mailto:[log in to unmask]>[log in to unmask]> wrote:
>> >
>> >Hi Christina:
>> >
>> >>Dear SPMers,
>> >>
>> >>I am new to SPM and I would appreciate any help on the following
>> question.
>> >>I have a four session design. Each session contains all the conditions I
>> >>want to compare later on through contrasts. When presented, the sessions
>> >>follow each other and there is a very short break between sessions
>> >>presentation (30sec approx.).Participants stay still and they only get
>> out
>> >>of the scanner after they have completed all the sessions. I have
>> analysed
>> >>the data as one single session (instead of 4) as this made contrast
>> >>definition easie .... I am now unsure about how correct this is. Do I
>> >>violate any assumptions? Is it a common procedure to regard multiple
>> >>sessions as one? Is this ‘eligible’?
>> >>
>> >>
>> >I think that's a bad idea - each time you stop start the scanner you
>> >have variations. Therefore you should use 4 sessions in your design,
>> >each one having its' own cste. Contrasts are not more complicated, if
>> >you enter the conditions the same way you should be able to use exactly
>> >the same contrasts.
>> >
>> >Best
>> >Cyril
>>
>>
>> --
>> _____________________________________
>>
>>
>> Help fighting hunger: http://www.hungersite.com
>>
>> Just click your mouse and sponsors of The Hunger Site donate a
>> serving of food to a person in need - at no cost to you.
>>
>> ______________________________________
>>
>> Mauro PESENTI
>> Research Associate, National Fund for Scientific Research (Belgium)
>> Unite de Neurosciences Cognitives
>> Departement de Psychologie
>> Universite Catholique de Louvain
>> Place Cardinal Mercier, 10
>> B-1348 Louvain-la-Neuve
>>
>> tel.: +32 (0)10 47 88 22
>> fax: +32 (0)10 47 37 74
>> E-mail: [log in to unmask]
>> http://www.nesc.ucl.ac.be
>> http://www.nesc.ucl.ac.be/mp/pesentiHomepage.htm
>>
>
|