Computing LFP with use_fast_imem and extra_scatter_gather

Anything that doesn't fit elsewhere.
Post Reply
pascal
Posts: 106
Joined: Thu Apr 26, 2012 11:51 am

Computing LFP with use_fast_imem and extra_scatter_gather

Post by pascal »

I have successfully implemented computation of the LFP in a previous network model using cvode.use_fast_imem and cvode.extra_scatter_gather (see viewtopic.php?f=8&t=3656&p=15713&hilit= ... m+1#p15713)

I am trying to use this same approach in a new network model, but I am having some trouble. I am trying to record the LFP contribution from a subset of neurons in a network of 800 total neurons. The problem is that the LFP registers action potentials, but not post-synaptic potentials. For instance, I ran a simulation in which I recorded both the intracellular voltage and the LFP from one single neuron in the network. (I recorded the LFP from a single neuron by defining a subset of sections belonging to just that neuron, then passing that subset of sections to a callback using extra_scatter_gather.) The intracellular voltage shows EPSP’s:
Image

But the LFP is a perfect flatline (and yes, I checked to make sure there are not small fluctuations when you zoom in):
Image

But if I increase the strength of excitatory synapses onto this cell, then spikes are elicited:
Image

And this does show up in the LFP:
Image

I am at a loss to explain why AP’s register in the LFP, but PSP’s do not. I made a toy simulation with just one neuron, but unfortunately I was unable to replicate the problem (i.e., PSP's did show up in the LFP in the toy simulation).

Any ideas what the problem might be, or how I might troubleshoot this?
pascal
Posts: 106
Joined: Thu Apr 26, 2012 11:51 am

Re: Computing LFP with use_fast_imem and extra_scatter_gather

Post by pascal »

Never mind, I figured out the problem...I only printed out the LFP trace to three decimal places, and apparently the signal was so small that it required greater precision to see the fluctuations. Everything is working as it should.
Post Reply