nettron2000 at aol.com wrote:
> Joe Legris <jalegris at xympatico.ca> wrote in message news:<3FDFBB23.6060108 at xympatico.ca>...
>>>Matthew Kirkcaldie wrote:
>>>>>In article <Xns94539459E651FBilZ0rhotmailcom at 202.20.93.13>,
>>> BilZ0r <BilZ0r at TAKETHISOUThotmail.com> wrote:
>>>>>>>>>>>>>Yes, its very common. Just often another way a cell mediates negative
>>>>feedback.
>>>>>>>>>Why would an autapse from an excitatory neuron constitute "negative
>>>feedback"? It would be a positive feedback loop if anything.
>>>>>> Matthew.
>>>>Positive feedback might make a neuron "bursty" - so that any time it
>>fires, there's a good chance it will fire many times in rapid
>>succession. Maybe the length of the feedback loop sets the bursting speed.
>>> Bursting speed ? I was under the assumption that a typical neuron
> operated at its max "frequency" when bursting. If this isnt
> necessarily the case, then what could be the alleged purpose of
> bursting in neural computation ?
I don't know if there are various bursting speeds, I was only guessing.
Now that I think about it a little more, positive feedback would have
the effect of making the neuron progressively more active, so that its
firing rate would increase at a rate proportional to its present firing
rate. An exponential function?
--
Joe Legris