How many conditions did you model at the first level?
If you did not model 9 conditions (3x3), then you can't create any of
these second level models. It would be better to have one big model,
but this means you need to have your 9 conditions modeled at the first
level.
Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
Postdoctoral Research Fellow, GRECC, Bedford VA
Website: http://www.martinos.org/~mclaren
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 Mon, Oct 8, 2012 at 5:51 AM, Juste S. <[log in to unmask]> wrote:
> Dear SPM’ers,
> I need some advice on the second level analysis model for my current experimental design. I am reposting this message since I haven’t received any reply and I would really use some help/suggestions.
> I conducted an event-related cued task/state switching paradigm with 3 types of cues indicating different types of trials, namely, “rest” (waiting) periods (cue1), task1 (cue2) (parity judgment on number stimuli) and task2 (cue3) (magnitude judgment on number stimuli) with 2 groups of participants (clinical and control). In this paradigm subjects have to alternate between the 2 task trials and also “rest” (waiting) periods. I am interested in first establishing the very cue-related effects, and also the switch or repeat related effects (with reference to n-1 trial).
> The first idea was to use the flexible (2x3x3) factorial design with a group factor with 2 levels (clinical and control), cue type factor (cue1, cue2, cue3) and switch/repeat factor with three levels as well (cue(1,2,3) to cue(1,2,3)). However, I run into a problem here since the switch/repeat factor involves the qualitatively different repeats and switches, namely, the switches from “rest” to task (vice versa) and between tasks, same with the repeats (state switches and task switches/state repeats and task repeats), and the results become uninterpretable.
> So, my question is, is it valid instead of having a model encompassing everything, to split it into a number of flex factorial models, involving only a few conditions, e.g. cue1 repeat (rest repeat) trials and cue2/3 switch to cue1 (rest trials preceded by task trials), for all the relevant switches and repeats. Or, maybe a more proper way to go is to do the t contrasts of interest (e.g. cue1cue1>cue2/3cue1) in the first level analysis, bring them to second level and do the 2 sample t test on them.
> Thanks in advance for your suggestions.
> Juste
|