[Hplusroadmap] An update and some notes on attention (was Fwd: Your Plastic Brain)

[Bryan Bishop]

On Wednesday 21 May 2008, Dan Bolser wrote:
> Perhaps off topic but (I say perhaps because I lost interest and
> didn't read the whole email) ...

That's no problem, any time you think I've been babbling, just call me 
out on it and ask me for summaries or elaboration in some context.

> Isn't wavering attention an important part of human intelligence?
> i.e. "dam! I'm doing this boring task *again*, surely there is some
> way I can speed up this task?" Hofsdater calls this 'in the system'
> (doing the task) vs. 'out side the system' (thinking about how to
> avoid doing the task).

Yes.

> I know you need to focus to perform some tasks, and being focused
> more easily would make performance on those tasks more efficient, but
> doesn't too much attention on one task lead to a kind of uncreative
> robotism?

Nah, I can't see that as being true. When you sit down to program, you 
can sustain attention and focus, but at the same time, and without 
resorting to pseudopsychoscience (the most terrible kind of 
pseudoscience :-)), it seems that top-down management still works. So 
if you're getting bored, if you're noticing that there should be a 
better way to get something done, then *go get it done*. That's the 
whole point. :-)

> Anyway, for 'highlevel' neurobioloigcal models of attention, I think
> you can find some interesting stuff from Susan Greenfield - she puts
> schitzophrenia and OCD on one 2D neurochemical spectrum that controls
> attention and 'error checking' mechanisms in the brain.

http://www.pharm.ox.ac.uk/academics/greenfield

> Measurement of dopamine release ‘on-line’ from brain slices using fast
> scan cyclic voltammetry.  

http://en.wikipedia.org/wiki/Cyclic_voltammetry

Looks like an electrode-setup for measuring ions and charge. So this is 
somewhat like my neuropod idea, or the idea of combinatorial 
experimentation with organotypical neurotissue slices for the massive 
combinatorial possibility space of neuropsychopharmacology.

> Real time monitoring of release of a protein from the brain 

http://brain-maps.org/

Hm. Any chance of noninvasive deep-brain protein reading? The Allen 
Institute is funding brain-maps for humans now, it's just 
slice-and-scan of brains, and then profiling the mRNA that was present 
in those locations, so no time-dependent data is being acquired.

> Electrophysiological recording from brain slices in vitro and 
> substantia nigra in vivo in both physiological and pathological 
> conditions   

http://heybryan.org/docs/neuro/
http://heybryan.org/mediawiki/index.php/Brain_implants

- Bryan
________________________________________
http://heybryan.org/