Jul 06 2012
Robot Legs and Central Pattern Generators
You are currently browsing comments. If you would like to return to the full story, you can read the full entry here: “Robot Legs and Central Pattern Generators”.
10 Responses to “Robot Legs and Central Pattern Generators”
Leave a Reply
You must be logged in to post a comment.
Dr. Novella.
You touched on this briefly, though it’s a bit off-topic, but I wanted to ask you about brain-machine interfaces.
There is equipment that allows the blind to see images by implanting electrodes into the brain, but have you heard/read anything about going the other way, i.e. taking images from within the brain, the so-called mind’s eye, and displaying them?
I would check myself, but I am no longer in school and have limited access to online journals (and that’s another story altogether).
Thanks,
-Adam
Adam: “have you heard/read anything about going the other way, i.e. taking images from within the brain, the so-called mind’s eye, and displaying them?”
I’m going to hazzard a guess and say that this would not be possible. There are no images in the brain. Different aspects of vision are processed in different parts of the brain. They do not come together as image in the brain.
It would not make sense for the brain to distribute the visual image on the retina to different parts of the brain for processing only to bring it all back together again to form an image just like the one it started with on the retina.
Somewhat off topic, but where does learned automation lie in the hierarchy — as as when an elite athlete is able to repeatedly duplicate the optimal mechanics of a movement without conscious thought? In the wake of the New Orleans Saints bounty scandal, I even read that incentives can degrade these types of muscle memory by causing the athlete to over think a movement that would otherwise be automatic. Does the control point or process change from the time the athlete is consciously learning an action to the time when he or she has committed it to muscle memory?
Adam and Billyjoe7:
I heard somewhere (maybe it was here?) that scientists had produced a robotic camera eye, and a mathematical algorithm that converted the pixel display from the camera into crudely similar images that could be sent down the optic nerve and the brain could ‘understand’ as it were… Wish I could remember where I have seen it… Sure its at low resolution but I’m sure the technology will increase
– I just cant wait for when they reverse that conversion and I could plop a little chip onto my optic nerve and make me summon a browser window in my left eye field, having internet access at any time would be pretty handy!
BillyJoe
Sorry, what I was going for would effectively be a way for other people to see what someone else sees in their mind or, effectively, a way to see what people dream of. My thought process was that if images are taken in through the eyes and processed by the brain, could the process be reversed?
Hopefully that makes more sense.
I have really vivid dreams and I guess I was just thinking how awesome it would be to see them played back!
“My thought process was that if images are taken in through the eyes and processed by the brain, could the process be reversed?”
I suppose it would be possible, but imagine how much it would take to perform that action.
They would need to identify every area of the brain involved in visual perception, interpret the neuronal level activity in those areas and, from that, somehow deduce what the brain “sees”. There are different modules for shapes, boundaries, colour, contrast, and many more features of vision. And what a person sees depends on what they saw a moment before and is also influenced by their present emotional state and past experience.
And there’s another level of complexity. Only our central vision is sharp and colourful but, as a result of flitting our eyes around the room, we seem to see the whole room as being sharp and colourful. So we would need a time aggregated image. Otherwise what we would see is a sharp, colourful central spot surrounded by an area that is increasingly blurred and colour desaturated towards the periphery. The brain also compensates for the effects of body and eye movement which would otherwise produce a jerky and motion blurred image.
I think this won’t happen in our lifetime.
Heptron, see Reconstructing Visual Experiences from Brain Activity Evoked by Natural Movies.
We had a discussion of that paper at philosophy of brains in this thread. I discussed some important caveats there, and the obvious directions this research could go.
“We are not consciously aware of most of the processing that is needed to produce smooth and coordinated movement. If we desire to walk across the room, for example, we guide our movements to accomplish that desire, but we are not aware (thankfully) of the many components that go into the astonishing balancing act of walking.”
Anyone who has ever tried to play QWOP knows how horribly difficult (and hilarious) attempting to coordinate a pair of legs via conscious processes can be.
http://www.google.com.au/search?q=qwop&hl=en&safe=active&rls=com.microsoft:en-au&tbm=vid&prmd=imvns&source=lnms&sa=X&ei=jmf6T_fOO-i1iQfo56DKBg&ved=0CBIQ_AUoAw&biw=1117&bih=586
I worked in this area in a different lab. It’s bittersweet to see someone else get there first but at least someone got there.