 |
 |
Dear Peter, Thank you so much. Actually, the ultimate goal is to know better the human mirror neuron system and using our computational model to predict the task performance and also to make predictions about a completely different set of data. I really have no idea about CMC or other models at the moment. But as we have done our measurements with fMRI and found the connections of mirror neuron system with DCM analysis, we are trying to change the parameter of our spiking network model to match the results from our experiments! Best, Sadjad On 6 July 2017 at 17:37, Zeidman, Peter <[log in to unmask]> wrote: > Dear Sadjad > That sounds like a very interesting analysis. To clarify - what's your > objective for this analysis? Is it to validate DCM? Or to ask something > specific about haemoynamics? I am wondering if there's a particular reason > you went for DCM for fMRI, rather than a biophysical model which explicitly > models firing rates i.e. the canonical microcircuit (CMC) model. > > Best > Peter > > -----Original Message----- > From: Sadjad Sadeghi [mailto:[log in to unmask]] > Sent: 06 July 2017 16:34 > To: [log in to unmask]; Zeidman, Peter <[log in to unmask]> > Subject: Re: VOI Time series adjustment > > Dear Peter, > > Thanks a lot for your reply. Yes I am working on DCM for fMRI and I have > found the best model for effective connectivity between 3 regions, then I > have simulated the same connectivity model for a spiking network model > which originally was designed for the prefrontal cortex. Each region > consists of 1000 neurons and it gives the spiking times of each neuron, > then I converted the spiking times of the neurons to have the firing rate > vs. the simulation time (in order to have the same output as predicted data > from DCM). For the input to this simulated network, I can only determine > the frequency of it, that is why I am also looking for the frequency to > enter the same strength of input as my experimental tasks! (as at the > moment the timeseries in DCM.U.u has only the value of 1 and 0 ). Then I > need to match this time series with time series from DCM analysis to > implement experimental tasks within spiking network model afterward. I hope > this was enough and looking forward to your more feedbacks about my > questions. > > Best wishes, > Sadjad Sadeghi > >