Temporal neural activity - what is it?
jonesmat at ohsu.edu
Mon Nov 8 13:43:00 EST 1999
In article <ynuV3.8974$Q5.352632 at news1.rdc1.ab.home.com> Kyle Doerksen,
>I'm interested in brain patterns as varying with time. From what little I've
>read, it seems that new reserach is being conducted that indicates that the
>patterns (time series) of neuronal excitation are important in overall brain
>function. Could you please advise me as to where I could look to find more
>information about this, and what relevant terminology I could use for www
This is currently a very controversial subject. One point of view is
that the precise pattern of timing of action potentials (spikes) carries
a lot of information. Among people holding this view, some emphasize
correlations in spiketimes over populations of neurons, and suggest that
these correlations (especially synchronous oscillations) help to "bind"
information from different sensory modalities together into a coherent
percept. A recent issue of the journal Neuron had a whole section focused
on this so-called "binding problem". Definitely worth reading. Another
emphasis is placed on the patterns of spikes in individual neurons, and
work in this area also suggests that spike timing carries a lot of
information, in fact, individual spike times can convey quite a lot of
information in some experiments.
Naturally, there's a whole 'nother camp which claims that spike timing
doesn't really matter, because the brain is too noisy a place for single
spikes or even coherent oscillations to make any difference, and instead
it's the -average- rate of firing that contains all the information.
For excellent reviews of the "binding" problem, see the September 1999
issue of Neuron. For a thorough and thought-provoking explanation of the
information content in patterns of spike firing, see the book "Spikes:
Exploring the Neural Code" by Rieke et al.
More information about the Neur-sci