In article <36F23B6B.814145E7 at axcess.co.uk> Karla Parussel,
kparussel at axcess.co.uk writes:
>Would this be relevant to inhibitory interneurons that are said to be
>found quite a lot in competitive networks? I have looked all over the
>place, searched the net and read through my books but I just can't seem
>to find the answers. My maths isn't really that strong which probably
>limits me a lot when it comes to understanding large mathematical
>formuli. If anyone could point me in the right direction I would be
>really appreciative.
>
There is a little ambiguity in the terminology surrounding the word
"interneurons". Taken literally, it means "between the neurons", i.e.,
these are neurons that are in between the "major" neurons in a particular
circuit. The "major" neurons are usually first identified anatomically,
so they are often the biggest ones around or they are arranged in some
really obvious laminar pattern, that makes them stand out as a population
under a microscope. Interneurons are often fewer in number, scattered
around sparsely without the same obvious laminar distribution. Often,
interneurons have their cell bodies located in the local lamina into
which the principle neurons send their dendrites.
It turns out in most cases that the "major" or "principle" neurons are
also the ones that project axons _outside_ of the immediate circuit,
whereas the interneurons only form local connections within the circuit
that their cell body lives in. I think this is opposite to what you
stated in your question. It further turns out that most principle neurons
(i.e., with long-range projections) in the brain are glutamatergic,
excitatory neurons, whereas most interneurons are GABAergic local
inhibitory neurons. There are some obvious exceptions: Purkinje cells in
the cerebellum are the principle cells, but are GABAergic and have long
range projections. But for the most part, these cases are rare.
Finally, interneurons, by secreting GABA, cause a hyperpolarization and
also a resistive shunt. In your simulations, you would maybe implement
these as a negatively weighted synaptic input, or a divisive input in the
case of the shunt (i.e., a larger leak when the inhibitory synapse is
activated). Interneurons participate in both feedforward and feedback
inhibitory connections between principle neurons. Thus it is possible
that these might correspond to a mechanism for backpropogation of errors.
For further information, try doing a medline search for the following
authors' names:
Freund and Buszaki (a comprehensive review paper on hippocampal
interneurons).
Roger Traub (simulations of hippocampal networks, using conductance-based
models involving interneurons)
Shadlen and Newsome (J.Neurosci 18:3870-3896) (A paper where the balance
between excitation and inhibition is explored in terms of physiological
data and models using integrate-and-fire and spiking neurons)
Konig, Engel and Singer (a review featuring inhibition as a mechanism of
increasing neuronal coincidence detection)
Cheers,
Matt Jones