Scientists find way to map brain's complexity(reuters.com)
reuters.com
Scientists find way to map brain's complexity
http://www.reuters.com/article/2011/04/10/us-brain-model-idUSTRE7392KU20110410
6 comments
This should make Ray Kurzweil happy.
http://news.ycombinator.com/item?id=1611007
http://news.ycombinator.com/item?id=1611007
Combine with http://news.ycombinator.com/item?id=2421597 and perhaps sentience gets achieved in a Google cluster. :)
Is Tom Mrsic-Flogel, the lead researcher, any relation to Janko Mrsic-Flogel who is credited as programmer on some early '80s videogames and as musician on many later ones?
This was featured on a TED a while ago:
http://www.ted.com/talks/sebastian_seung.html
http://www.ted.com/talks/sebastian_seung.html
There are many approaches to the connectome. One way to differentiate the groups is by the imaging resolution at which they work. The Nature paper uses light microscopy and patch clamping a sample of neurons (caveat -- haven't had a chance to read the paper yet) while Seung's group works on methods to trace electron microscopy, which is significantly higher res than light microscopy. You can see individual synapses (e.g., the T-bars and post-synaptic densities) in EM images while the light microscopy guys need another method to establish true connectivity versus physical proximity, hence the patch clamping. Ideally, you'd be able to tell whether two neurons are connected to each other by looking at high-res images -- see the T-bar, lots of vesicles with neurotransmitters, the post-synaptic densities on neighboring neurons. The gold standard, though, is to be able to electrically establish connectivity. You just can't do that gold standard for the vast number of neurons in the brains.
True. Hadn't noticed this was a different group. I assumed it was connected with Seung's group (other than by the subject itself, of course).
A fascinating subject though, and one that might bring yet more insight into the recently reborn field of consciousness and subjective experience.
Kohonen forever.
http://www.nature.com/nature/journal/vaop/ncurrent/full/natu...