No.Is it possible to have the weight of a NetCon pointing to a "reference of a variable"?
This can already be done with code in the synaptic mechanism's NET_RECEIVE block.That way the weight of the NetCon can vary during a simulation (to account for use-dependent plasticity for example).
Quite the contrary. Are you familiar with the models of short-term use-dependent plasticity described byall the use-dependent plasticity has to be dealt with in the post-synaptic side in a mechanism (a .mod file for some conductance, as in NEURON book example 10.5), which conceptually doesn't represent pre-synaptic plasticity very well (e.g. synaptic depression resulting from depletion of vesicles).
Varela JA, Sen K, Gibson J, Fost J, Abbott LF, Nelson SB (1997) A quantitative description of short-term plasticity at excitatory synapses in layer 2/3 of rat primary visual cortex. J Neurosci 17:7926-40
and
Tsodyks M, Pawelzik K, Markram H (1998) Neural networks with dynamic synapses. Neural Comput 10:821-35
?
These models involve representations of presynaptic processes, and NEURON implementations of them are available from ModelDB http:www.modeldb.edu/.
NMODL also has a WATCH keyword that can be used to monitor membrane potential at the postsynaptic site for occurrence of an action potential (helpful for implementing spike timing dependent plasticity), and a FOR_NETCONS keyword so that a single instance of an STDP mechanism can deal properly with multiple input streams. See
http://www.neuron.yale.edu/neuron/news/ ... 23_06.html
where you will find an announcement of FOR_NETCONS and a link to stdp.mod which is an NMODL implementation of a synaptic mechanism with STDP.
Take a look at the short-term plasticity and STDP mechanisms that are already implemented and see if they meet your needs, perhaps after some minor modifications. If not, maybe a different approach would be necessary.Maybe this can be accomplished by having a "release" mechanisms listening to the NetCon, and all the post-synaptic mechanisms listening to that release mechanism using "setpointer", though it is sort of obsolete at this time. What do you think?