help
help > RE: Effect strength of factor
Apr 11, 2023 12:04 PM | Alfonso Nieto-Castanon - Boston University
RE: Effect strength of factor
Dear Peet
To look at whether the effect of a between-subjects factor ("performance") on functional connectivity-change differs between two groups you would typically define an interaction analyses:
1) in Setup.Covariates (2nd-level) select the two groups (e.g. variable "CONTROL" and "INTERVENTION") and the "PERFORMANCE" variable and click on 'Covariate tools. Create interaction of selected covariates'. That will create some new "CONTROL*PERFORMANCE" and "INTERVENTION*PERFORMANCE" interaction variables.
2) then back in the Results (2nd-level) tab, select (for example, assuming that you are running seed-based connectivity analyses) the options that read:
- "Analysis of seed-based connectivity change between "time1" and "time2"
(that will automatically select the two conditions and enter a [-1 1] between-conditions contrast)
- "Does the correlation between connectivity change and PERFORMANCE differ between CONTROL and INTERVENTION subjects".
(that will automatically select all four CONTROL, INTERVENTION, CONTROL*PERFORMANCE, INTERVENTION*PERFORMANCE variables and enter a [0 0 -1 1] between-subjects contrast)
Best
Alfonso
Originally posted by Peet Nowo:
To look at whether the effect of a between-subjects factor ("performance") on functional connectivity-change differs between two groups you would typically define an interaction analyses:
1) in Setup.Covariates (2nd-level) select the two groups (e.g. variable "CONTROL" and "INTERVENTION") and the "PERFORMANCE" variable and click on 'Covariate tools. Create interaction of selected covariates'. That will create some new "CONTROL*PERFORMANCE" and "INTERVENTION*PERFORMANCE" interaction variables.
2) then back in the Results (2nd-level) tab, select (for example, assuming that you are running seed-based connectivity analyses) the options that read:
- "Analysis of seed-based connectivity change between "time1" and "time2"
(that will automatically select the two conditions and enter a [-1 1] between-conditions contrast)
- "Does the correlation between connectivity change and PERFORMANCE differ between CONTROL and INTERVENTION subjects".
(that will automatically select all four CONTROL, INTERVENTION, CONTROL*PERFORMANCE, INTERVENTION*PERFORMANCE variables and enter a [0 0 -1 1] between-subjects contrast)
Best
Alfonso
Originally posted by Peet Nowo:
Dear fellow CONN-enthusiasts,
My problem:
I am comparing resting state fMRI functional connectiviy change of 2 groups (control & intervention) between 2 sessions (time 1 & time 2). Now I would like to know if the influence of a factor ("performance") on functional connectivity differs between the groups. Performance is a metrical numerical variable with a range from 0 to 3.
My goal is to know the influence of group and performance on functional connectivity.
Can somebody help me? How can I do that?
Cheers
My problem:
I am comparing resting state fMRI functional connectiviy change of 2 groups (control & intervention) between 2 sessions (time 1 & time 2). Now I would like to know if the influence of a factor ("performance") on functional connectivity differs between the groups. Performance is a metrical numerical variable with a range from 0 to 3.
My goal is to know the influence of group and performance on functional connectivity.
Can somebody help me? How can I do that?
Cheers
Threaded View
| Title | Author | Date |
|---|---|---|
| Peet Nowo | Apr 6, 2023 | |
| Alfonso Nieto-Castanon | Apr 11, 2023 | |
| Peet Nowo | Apr 17, 2023 | |