Question about the parameter bounds for the Wilson-Cowan model

[NeuroLife]

Hi,

I was wondering what was the motivation behind the bounds for the different parameters of the Wilson-Cowan model. I noticed that some parameters have lower bounds because if they go below that value (e.g., tau_x = 0) the system explodes or vanishes, but some other parameters have an upper bound with a specific value much larger than what is used in the different papers mentioned (e.g., theta_x has 60.0 as upper bound). 

I am exploring single node model calibration and was using those bounds as hard priors, so I was wondering where they came from and why they were defined that way. I am hoping to stay within biologically realistic settings/behavior.

Thank you for your time!

Dominic Boutet

[WOODMAN Michael]

Hi,

Parameter bounds are best when there's a mathematical or physical reason for them e.g. probabilities are between 0 and 1.  In the absence of such constraints, TVB tries to provide helpful bounds beyond which the behavior saturates, since exploring parameter spaces in saturated regimes is not useful.  You are free to ignore those bounds of course.  

cheers,

Marmaduke

--
You received this message because you are subscribed to the Google Groups "TVB Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tvb-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tvb-users/1f1c0830-de49-44ad-b606-75cbffa87365n%40googlegroups.com.

[NeuroLife]

Hi,

I see, that makes sense. Thank you!

Have a nice day!

Dominic Boutet