On Mon, 28 Nov 1994, Andrew Baek wrote:
> In article <3ba25p$pkk at portal.gmu.edu> you wrote:
> : I'm modeling a network that has dendro-dendritic synapses, and I need a
> : pointer to a synaptic model that allows me to address graded potentials.
> : Everything I've seen assumes the synapse is triggered by an action
> : potential, and that isn't valid for me. The closest book I've seen to what
> : I need seems to be Shepherd's 3rd Edition of Neurobiology, and that
> : doesn't address the modeling issues.
>> The fact that the synapse is trigeered by an action potential is widely
> accepted. But the postsynaptic potential generated in response to an
> action potential can vary in its amplitude and time constants. The
> variabilities depend on the synaptic efficacy (synaptic weight) and the
> synaptic location where the action potential arrived.
Actually, it's triggered by membrane depolarization, which need not be
associated with an action potential, particularly in the dendritic tree.
(Since the number of vesicles released is a logistic function in the
depolarization, you tend to get an action potential on the post-synaptic
side if you get anything at all.) Levitan and Kaczmarek, the Neuron, and
Freeman, Mass Action in the Nervous System, have some useable materials,
and I guess I'll just work from those. One question does have to be
resolved--what is the nature of the refractory period on the presynaptic
side of the synapse? References gratefully appreciated.
>> The book to consult on computer simulation of synapto-dendritic trees is
> "Methods in Neuronal Modeling", by Koch and Segev.
Spent the weekend working through Koch and Segev. Didn't help that much
for the presynaptic side. Mascagni's chapter on numerical methods was
extremely interesting--it confirmed some suspicions I had about the
stiffness of compartmental models. Unfortunately, compartmental models
result in extremely large connectivity matrices--the one I'm using in a
massively simplified model of a piece of the olfactory bulb is
608x608--pretty much ruling out implicit methods. To avoid stability
problems with explicit schemes, I have to reduce both time stepsize and
spatial stepsize, driving me towards even larger connectivity matrices.
Now I see why Freeman uses Katchalsky Networks.
>> For the anatomy and neurophysiology, look at "From Neuron to Brain", by
> Nicholls, and at. el.
Internet: herwin at gmu.edu