hi - think this is a pretty basic question
so apologies for that
We are interested in modelling a learning effect
in a simple desigm
Each participant has 2 runs, 5 blocks each
For each block we modelled EVs for predictive and unpredictive cues, leading to 10 EVs
For the first-level analyses, we computed 5 copes for the difference
between predictive > unpredictive cues
We are interested in the learning effect, namely how the
difference between predictive and unpredictive cues changes across
the 10 blocks
We then submit, for each subject, the 10 copes from the 2 first level-analyses
into a Fixed effects analyses
specifying the design as
1 0 0 0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0 0
0 0 0 0 0 1 0 0 0 0 0
0 0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 0 0 1
and (for example) a linear contrast
-4.5 -3.5 -2.5 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5
I guess this model look fine?
or would it be more appropriate to do a Fixed effects with one EV for the relevant cope
and another covariate to model the learning effect, such as
1 -4.5
1 -3.5
1 -2.5
1 -1.5
1 -0.5
1 0.5
1 1.5
1 2.5
1 3.5
1 4.5
and then assess the covariate effect for each subject and bring that to the 3rd level
which model would be more appropiate in this case? guess the 2 models are pretty much
the same..what is the difference if any
or is there a better way to model for this experiment?
suggestions welcome and appreciated!
thanks so much
David
|