Back to PLS Help

contrast for non-rotated behavioral PLS
kambiz
Posted on 01/14/14 11:14:26
Number of posts: 13
kambiz posts:

Dear PLS experts,

I have a question about how to set up a contrast for non-rotated behavioral PLS. I have two groups (Group A and B) and two conditions (Con C and D). I would like to seek a network which is functionally connected to a seed in greater extend during Con C than Con D.However, at the same time, we would like to see if functional connectivity of such a network that reflect ConC> Con D is greater in group A than group B.  This is exactly like having a functional connectivity map of a network which is more connected to the seed during Con C than Con D. Then, we would like to compare such a map across different groups (Group A and B). In short, something like GroupA (Con C>Con D) > Group B (Con C> Con D).

Using PLSgui, I  understood that I could easily set up a contrast for each group (Con C> Con D by setting 1 -1), but how can I set a contrast to investigate group difference for an specific contrast which seeks condition differences.

Do you have any suggestion on how to set up such a contrast?

 

Best

Kami

Replies:

Untitled Post
rmcintosh
Posted on 01/15/14 08:56:57
Number of posts: 394
rmcintosh replies:

quote:

Dear PLS experts,

I have a question about how to set up a contrast for non-rotated behavioral PLS. I have two groups (Group A and B) and two conditions (Con C and D). I would like to seek a network which is functionally connected to a seed in greater extend during Con C than Con D.However, at the same time, we would like to see if functional connectivity of such a network that reflect ConC> Con D is greater in group A than group B.  This is exactly like having a functional connectivity map of a network which is more connected to the seed during Con C than Con D. Then, we would like to compare such a map across different groups (Group A and B). In short, something like GroupA (Con C>Con D) > Group B (Con C> Con D).

Using PLSgui, I  understood that I could easily set up a contrast for each group (Con C> Con D by setting 1 -1), but how can I set a contrast to investigate group difference for an specific contrast which seeks condition differences.

Do you have any suggestion on how to set up such a contrast?

 

Best

Kami

hi Kami

 

the contrast you need is the same as you would do for coding a group X task interaction - if for group A you do 1, -1, then for group B you do -1, 1




Login to reply to this topic.

Contact Us:

3560 Bathurst Street
Toronto, Ontario
Canada M6A 2E1
Phone: (416) 785-2500

Baycrest is an academic health sciences centre fully affiliated with the University of Toronto