Consciousness, New Thinking About

Matt Jones jonesmat at physiology.wisc.edu
Thu Jun 6 11:04:11 EST 2002


"DJ" <DJ at hotmail.com> wrote in message news:<newscache$pui9xg$n0f$1 at maggie.netlink.com.au>...
> JGC9 <jgcasey at hotkey.net.au> wrote in message
> news:3cfeb3a5$1_1 at news.iprimus.com.au...
> >
>You could argue that these things are coincidental.  It's an
> interesting fact though that memory problems are commonly reported by people
> with epilepsy.



There's a pretty good reason for that, especially with regard to short
term memory and "episodic" memory (memory about events or sequences of
events in daily life).

The hippocampus is central to episodic memory. If you muck about with
hippocampal function, either grossly or subtly, you screw up the
ability to form new episodic memories, but generally don't affect
already existing memories.

The hippocampus is also the brain region with the lowest seizure
threshold.  If you do something likely to induce seizures, like inject
a GABA-A receptor antagonist, in various brain regions, the
hippocampus generally displays epileptiform activity before other
regions, and generally at lower doses of drug. Also, head injury tends
to preferentially damage hippocampal inhibitory neurons, even though
they're buried relatively deep, leaving neurons closer to the injury
site less damaged.

The current thinking about how the hippocampus works is that it is an
"autoassociative" memory, which means that if certain neurons are
activated together (thereby combining to "represent" the thought or
memory), then excitatory connections between these neurons are
strengthened. Therefore, the next time some of the neurons are
activated together in response to a partial cueing stimulus, they can
recruit the rest of the ensemble that was originally activated
together (this is called "pattern completion", and is a sort of
cellular analogue of being reminded of something by seeing a vaguely
similar thing).

Anyway, it's these reciprocal excitatory connections that allow
autoassociative memory storage, but they also provide a form of
"self-excitation" that is probably what makes the hippocampus so
jumpy. It is an organ poised on the verge of a seizure all the time,
and needs to be that way in order to store and recall information.
Other regions proposed to have autoassociative functions are also
prone to seizure activity.

Seizures mess up memory probably because they activate everything,
instead of allowing the normal specificity of different neuronal
ensembles to represent different patterns.


Cheers,

Matt




More information about the Neur-sci mailing list