Hi, I am new to the CONN toolbox and have a question about setting the contrasts for 2nd lvl analysis. I have two groups (patients and controls), and results from a psychological test for each subject and I am performing connectivity analysis. I want to find some relationship between connectivity, group and score on the test.
Suppose I define the following 2nd lvl covariates
AllSubjects: [1 1 1 1 1 1 1 1 1 1]
Patients: [1 1 1 1 1 1 0 0 0 0]
Controls: [0 0 0 0 0 0 1 1 1 1]
Score: [12 14 18 9 7 15 20 14 10 8]
ScorePatient: [12 14 18 9 7 0 0 0 0 0]
ScoreControl: [0 0 0 0 0 15 20 14 10 8]
Which model is better?
1) AllSubjects + ScorePatient + ScoreControl, contrast [0 1 -1]
2) Patients + Controls + ScorePatient + ScoreControl [0 0 1 -1]
What are these models? Two-sample t-test and ANCOVA? Which one is better and what do they exactly test?
Suppose I define the following 2nd lvl covariates
AllSubjects: [1 1 1 1 1 1 1 1 1 1]
Patients: [1 1 1 1 1 1 0 0 0 0]
Controls: [0 0 0 0 0 0 1 1 1 1]
Score: [12 14 18 9 7 15 20 14 10 8]
ScorePatient: [12 14 18 9 7 0 0 0 0 0]
ScoreControl: [0 0 0 0 0 15 20 14 10 8]
Which model is better?
1) AllSubjects + ScorePatient + ScoreControl, contrast [0 1 -1]
2) Patients + Controls + ScorePatient + ScoreControl [0 0 1 -1]
What are these models? Two-sample t-test and ANCOVA? Which one is better and what do they exactly test?