skaggs at bns.pitt.edu
Tue Mar 25 15:43:42 EST 1997
Stephen Dunn <stephend at cogs.susx.ac.uk> writes:
> Can anyone direct me to some good references that discuss the
> differences between neuronal functioning in vertebrates and
Gordon Shepherd's "Neurobiology" testbook would probably be a good
place to start.
> So that you get an idea of what I am after, here's an example of the
> an issue: Learning and plasticity in invertebrates have often been
> associated with changes in intrinsic properties of the neurons as well
> as changes in synaptic strength. Vertebrate learning is for the most
> part assumed to be due to changes in synaptic strength alone.
It's not too hard to see how you might get this impression, but it
isn't really true. Many connectionist-type models of learning require
both synaptic changes AND changes in intrinsic properties. It's just
that, with a typical neuron having 10,000 synapses or so, the amount
of information that can be stored in modifiable synapses vastly
exceeds the amount that can be stored by changing any simple intrinsic
property (such as, say, spike threshold). Also, synaptic changes are
naturally associative, which is a desirable property in many models.
Thus, synaptic changes tend to get most of the attention -- but
there's no assumption that other types of changes do not occur.
-- Bill Skaggs
More information about the Neur-sci