Neuron charging

Joseph Strout jstrout at ucsd.edu
Thu Aug 29 10:51:13 EST 1996


On Tue, 27 Aug 1996, Gary Frank wrote:

> I am designing a computational neural network to do object recognition. 
>  I am basing it as closely as possible on the brain. 

If you really want to be biologically accurate, you should consider
compartmental modeling, or at least an integrate-and-fire model.  I have a
class library which is useful for both of these (though especially the
former); it runs on any platform, and is available at:

		http://www-acs.ucsd.edu/~jstrout/conical/

It includes some introductory material which you may also find useful.

>  I wonder if the post synaptic neurons that receive these impulses 
> are "charged up" as a result.

Yes, a cell membrane acts like a capacitor.  When a current is injected,
the membrane potential (voltage) approaches a new steady state
exponentially.  This is illustrated by the first CONICAL demo, c.f

     http://www-ncmir/~jstrout/conical/doc/demos/cylind-inject/

The time constant of this charging (e.g., time to change about 2/3 of the
way to its new steady state) is on the order of 5-10 milliseconds.

,------------------------------------------------------------------.
|    Joseph J. Strout           Department of Neuroscience, UCSD   |
|    jstrout at ucsd.edu           http://www-acs.ucsd.edu/~jstrout/  |
`------------------------------------------------------------------'




More information about the Neur-sci mailing list