Dear DCM experts,
I am currently attempting to perform DCM analysis on my data in which I
have a 2x2 factorial design consisting of two eye movement types (Smooth
pursuit (SP) and saccades (SAC)); and two stimulus presentation types
(predictable (PRD) or randomized (RND)). This results in 4 conditions in
total: PRD SP, RND SP, PRD SAC and RND SAC. See attached figure for design
matrix.
I am interested to know how the modulation between areas is changed
between the RND and PRD conditions. Based on a block analysis the areas
that appear significant in all conditions include: V5, frontal eye fields,
prefrontal cortex (DLPFC), and the cerebellum. I was thinking of including
all these areas in my design but would appreciate any advice on other more
appropriate areas i should look into. My input area would be V5.
I was thinking that the feedforward modulations would be higher during the
RND condition, and the feedback modulations higher during the PRD
conditions but could really do with some advice on how to devise an
appropriate design matrix to test this. At present I have used each
condition as a separate regressor but for my input I will need all
conditions? (see current design attached).
Any advice or help on this would be greatly appreciated,
Mel
Dr Melanie Rose Burke,
Sensorimotor Neuroscience Lab.,
Faculty of Life Sciences,
University of Manchester,
P.O. Box 88,
Manchester,
M60 1QD
U.K.
Tel: +44 (0)161 3063779
E-mail: [log in to unmask]
|