Hi,
I have two regressors in each subject’s 1st-level model and I want to calculate the difference of % signal change (%SC) in a ROI between regressor 1 and regressor 2 (i.e. regressor 1 > regressor 2). As I’ve asked before, I know MarsBar can extract this %SC subject by subject but it’s time consuming because it’s iterative (I have 20 subjects).
I thus hope to calculate %SC from the output of a group-level model. Before this, in a 2nd-level model, I conducted a 1-sample t-test to examine group maps. I then used the methods as below:
(1) (matlab directory is in the 2nd-level analysis folder) “design” – “set design from file” – select the contrast image of 2nd-level SPM.mat file.
(2) “data” – “extract ROI data (default” – select this ROI
(3) “Result” – “estimates results”
(4) “Data” – “Export data” – “Export summary timecourse for region” – “Export to text file”
I then get 20 numbers.
My questions are:
(1).
I think the 20 numbers represent the difference in parameter estimates (i.e. beta1 – beta2). Am I correct?
(2).
What is the relationship between beta value and %SC? It seems that
%SC = beta * (regressor_max_height / session_baseline)
Am I correct? So if I want to transform the 20 numbers into %SC, how can I calculate [also I have no idea how to calculate (regressor_max_height / session_baseline)?
Thanks a lot.
|