Synaptic modification rules ?

NMF nm_fournier at
Tue May 18 22:14:10 EST 2004

> modification takes place not exactly when spikes occur at the same
> instant in time but rather during a certain time window, dont quote me
> on this but something like 50 ms or so ? unless of course, one accepts
> 50 ms as occurring at the same time.

The results that you are discussing here refers to an old study  dating from
97 by Markram and colleagues.  They showed that long-term synaptic
potentiation or depression can be induced depending on the arrival time of a
backpropagating action potential (BAP) to the activated synapse.  (One thing
that has been known for a considerable length of time is that the generation
of an action potential also produces a passively conducting backprogating
wave throughout the dendritic arbor as well.).  Typically, if the measured
postsynaptic EPSP occurred after BAP arrived, depression is induced
(weakening of the synaptic connection); if the EPSP preceded the BAP,
potentation occurred (strengthening of the synaptic connection).  The
typical range is that of the coincidence of BAP and EPSP is on the order of
10 msec.  So if the BAPs occurred 10 msec before the EPSP, the amplitude
observed was reduced. Obviously, if the BAPs occurred 10 msec after the EPSP
the amplitude observed was increased.

One reason for the amplitude attenuation has to do with the types of ionic
channels that are activated.  Without getting too technical, the end result
if the BAP occurs before the EPSP, is that the relative segment of the
dendrite at the portion will be subsequently hyperpolarized.  If BAP occurs
after EPSP, there is evidence of suprasummation of calcium entrance,
possibly due to the unocclusion of the Mg2+ blockade ( hat Matt mentioned
for the NMDA receptor,) or the subsequent release of extra calcium through
internal stores (or perhaps the activation of L-type voltage dependent
calcium channels).

So the process you have mentioned suggests that the temporal coding of BAP
and EPSP has been likened to the important coincidence detector that Hebb
remarked in his work.  The spike-dependent (Hebbian plasticity) process is
considered important for producing long lasting changes in synaptic
potentation.  Moreover, due to the rather compartmentalized aspect of the
dendritic tree (i.e. branches, necks, spines, and complex geometry of
dendrites), one would find the potential for many different local
coincidence detection mechanisms, i.e. the NMDA receptor, itself, or even
the ryanodine-sensitive internal stores might all serve as coincidence
detection apparatus.

More information about the Neur-sci mailing list