[Neuroscience] Re: Q's about the distribution and roles of spiking
neurons in the brain
m.kirkcaldie at removethis.unsw.edu.au
Mon Jul 3 19:27:44 EST 2006
In article <1151930873.264993.68570 at h44g2000cwa.googlegroups.com>,
borg at swirve.com wrote:
> However, what I haven't
> been able to discover with my preliminary reading is whether specific
> types of spiking neurons are localized to specific areas of the brain
> or whether all types are distributed evenly. In addition, I have not
> been able to find anything definate about specific functional roles of
> the various types of spiking neuron behavior in the brain, either
> singularly or in combination.
The only stuff I've read along these lines is the work of Kawaguchi and
Kubota - in the 90s these guys painstakingly isolated, recorded and
characterised a range of different inhibitory neurons in the rat brain.
They found that, by and large, different types of inhibitory neurons had
different spiking patterns. Every neuron will behave in a
characteristic way if you stimulate its inputs or inject current across
the membrane, and the responses of these cells fell into several
categories which corresponded with their shapes and the types of calcium
binding proteins they used. Details are a bit hazy but I remember that
basket cells, which provide lateral inhibition by suppressing activity
of the cell bodies of the large "pyramidal" neurons, were capable of
sustained, extremely fast bursts of spiking, which the authors suggested
might be due to high levels of parvalbumin inside them - the idea being
that it's normally quite dangerous for a neuron to fire too fast for too
long, but the parvalbumin was protective because it kept calcium levels
manageable. Other types of cells had different CBPs and different
firing patterns as well as different shapes and connection preferences.
All these types are more-or-less uniformly distributed in the cortex,
they don't cluster in one place or another.
Intuitively I've always thought of cortex as a mix of cell types which
respond in a variety of ways to the same input, and from this variety
the local circuits are able to settle down into the "best" way to handle
that particular input. It's like if you had a diverse bunch of people
in an office and every time a new task was brought in, everyone tackled
it their own way until someone seemed to be getting on top of things,
whereupon the office would get together and finish the job. I have no
evidence for that analogy though.
More information about the Neur-sci