Dear Pegah,
I don't quite follow the relation between subcortical nuclei and
hyperparameters. In the case if DCM ERP hyperparameter is just a way
of telling DCM how much of your data can be assumed to be noise. So
the effect of it will be that if you force the scheme to think that
everything is noise it won't try to fit the model to the data. If you
look in spm_nlsi_N around line 217 you'll see that there is an option
to set the hyperpiors but it is not presently used by DCM ERP and the
code goes to the default setting which is something quite
uninformative.
Best,
Vladimir
On Wed, Mar 30, 2011 at 6:01 PM, Tayaranian Hosseini P.
<[log in to unmask]> wrote:
>
> Dear Vladimir,
>
> Actually, I have studied many papers relating to DCM including application and theory based papers. I have some questions on my mind so I am trying to manipulate every possible option in SPM to see their effects on the results and at the same time get more familiar with both DCM and SPM8. One of my questions is why these priors? I have not yet found out about it so I thought maybe by changing them I can at least have an idea about it. On the other hand, I am going to add some subcortical auditory nuclei to DCM to model the auditory path so changing hyperparameters will be of much importance for me in the near future.
>
> Best,
> Pegah
>
> ________________________________________
> From: Vladimir Litvak [[log in to unmask]]
> Sent: 30 March 2011 16:03
> To: Tayaranian Hosseini P.
> Cc: [log in to unmask]
> Subject: Re: [SPM] How to define priors in EEG SPM?
>
> Dear Pegah,
>
> Just out of curiosity why do you want to change the hyperparameters?
> This doesn't sound like the first thing a beginner would do.
>
> Vladimir
>
> On Wed, Mar 30, 2011 at 3:40 PM, Tayaranian Hosseini P.
> <[log in to unmask]> wrote:
>> Dear SPMers,
>>
>> I am very new to SPM so I have a basic question about it. How can I change hyperparameters in DCM section of EEG SPM?
>> There is a review priors button but I am not sure how to use it to change the hyperparameters.
>>
>> Best,
>> Pegah
>>
>
|