Re: Modeling axonal propagation delays

["Joshua W. Brown" <jwbrown@artsci.wustl.edu>]

Hi All,

As Randy mentioned, I am indeed developing a new pdp++ application intended
to model the real-time dynamics of neurons.  This new framework (RT++, for
"real time") has code to model synaptic delay mechanisms.  RT++ is still
pre-alpha, however, and it is not ready for public release.  (It works fine,
but there is almost no documentation, and I have only compiled it on SUN4
platforms so far.)

If you are interested, send me an email, and I will be happy to share the
code on an individual basis until public release.  If there is sufficient
interest, I will try to prepare an alpha release soon.  Details below.

Josh Brown

---------------
RT++ Details:
*Makes time, in real units of seconds or milliseconds, the backbone of a
simulation, with continuity across trials.
*Uses finite difference equations like those of Usher & McClelland (2001) to
model neural activity as a rate code.
*Allows arbitrarily complex time-sensitive contingencies on environmental
events (i.e. reward if correct but abort trial if no response within a given
time)
*Has automated training procedures to fit the timing of model output to
relevant data
*lots of other features

RT++ will be especially useful for modeling:
*response times in behavioral tasks (such as obtained in experiments with
PsyScope)
*effects of event timing within and across trials
*the time course of electrophysiological cell types obtained from
single-unit recordings in animals
*just about anything sensitive to the timing of individual events


> Josh Brown (jwbrown@artsci.wustl.edu) is writing
> a "real time" spiking unit model for pdp++ and may have already done
> this, and if not his framework would be the best place to add it.  Josh?
>