I am not quite sure I understand your problem description. You write that "... after
the simulation the Nodes top reacting to the noise input and become stuck on the refresh
Voltage ...". But in the figure you show, the neurons have V_M ≈ -75 mV ≈ E_L for the
first 30 ms, which would correspond to neurons only receiving noise input during this time
and afterward seem to spike at roughly 20 ms intervals. It does seem curious that they
seem to have a more or less constant membrane potential of ≈ -65 mV between the phases
So one question is what happens around 30 ms? The other question is why the
"locking" to -65 mV.
Could you send the full simulation script leading to the figure you included? Or even
better, a minimal version of it, i.e., a complete, runnable script producing a figure with
the effect you showed, but with as few layers and as few neurons per layer as possible?
Prof. Dr. Hans Ekkehard Plesser
Head, Department of Data Science
Faculty of Science and Technology
Norwegian University of Life Sciences
PO Box 5003, 1432 Aas, Norway
Phone +47 6723 1560
On 26/07/2021, 14:28, "Filip Blastik"
To whom it may concern,
I am writing regarding the problem with giving my topology network noise after simulation
with the signal. The network is oscillating with noise generator which is created as
single layer which provides noise for other nodes (which consist of layers). I stimulate
my network with stronger signal(another different layer) and after the simulation the
Nodes stop reacting to the noise input and become stuck on refresh Voltage and its for
longer periods than refractory period.
The model is derivation of Hill-Tononi example network.
The Nest is running on Ubuntu 20.04 , Conda/Spyder The nest is version v2.20.1
To visualize the problem here are the images which illustrate the problem
Here is the snippet of the code with the connections and layer setups
Bc. Filip Blaštík