Neural Representations of Trajectories

Harry Erwin harry.erwin at sunderland.ac.uk
Sat Jul 21 15:36:28 EST 2001


Lance Sherman <lancesherman at home.com> wrote:

> "Harry Erwin" <harry.erwin at sunderland.ac.uk> wrote in message
> news:1ewuuak.1ynor76163al0wN%harry.erwin at sunderland.ac.uk...
> 
> > I'm seeing evidence that trajectories of moving objects are represented
> > 4-dimensionally in the brain.
> 
> >Has anyone thought about the topology of neural representations?
> 
> at the risk of sounding flip, i feel compelled to reply, "i think i have"
> 
> i took one year of general topology thirty years ago. i'm obviously unsure i
> understand your interesting question. i will be looking for further replys
> to your post and further posts from you. i looked at your link to your www
> page and see (perhaps) some similarity to my ideas.

I won't be posting much until I've run it past my collaborators. I went
4 years towards a PhD in algebraic topology in 68-72, so I'm thinking
about the behavioral data in that context.

> 
> unfortunately i have produced no writing myself  and am hampered by lack of
> a collaborator.
> 
> for what it's worth, my ideas are in terms of time, processing into
> sequences, and breaking of symmetries.
> 
> i have tried to create directed graph models for control systems, not so
> different from the neural net models that have become popular. but i begin
> with structured stimuli that are music-like.
> 
> the primary components in my models are nodes (with the switching,
> prioritizing and integrating properties that are common in artificial neural
> nets)  and delay lines that connect them.
> 
> 
> at this point it is probably too clear why i have not produced any
> riting  - and why i have not found a collaborator.


-- 
Dr. Harry Erwin, harry at dherwin.org, http://world.std.com/~herwin




More information about the Neur-sci mailing list