Dear all,
I have two resting state sessions acquired one after the other for each subject and I want to use stochastic/spectral DCM to analyse the effective connectivity.
Since each session consists of only 100 volumes, I would like to concatenate the two sessions in order to obtain a more reliable estimation of the DCM parameters.
However, I am not sure if this is theoretically correct, in particular as resting state fluctuations are characterized by low frequency dynamics. Other problems may be the correlation in the neuronal noise for stochastic DCM and/or the hemodynamic response, which depends on previous states.
I would like to hear comments on this issue.
Thank you
Matteo Tonietto
PhD student at University of Padova
Department of Information Engineering
in the neuronal noise
|