What's new
TerraForums Venus Flytrap, Nepenthes, Drosera and more talk

Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

March (or scurry) of the cyborgs

The Ratbot: Researchers unveil first robot with "biological brain"
Posted: August 13, 2008, 9:50 PM by Brad Frenette
<object width="425" height="344"></object>
Robot technology has been making steady advances in recent years, but the announcement today of a robot with a rat brain seems to be a game-changer.

Researchers at University of Reading have built a robot with a "biological brain". This brain is made up of 300,000 cultured rat neurons, which are placed on a dish of electrodes. The electrodes pick up the electrical currents of the cells, which then compel the movements of the robot. When approaching an object, the brain signals the robot's wheels, which then avoid the object.
No remotes, no humans, this brain is the only thing that controls this bot. And as the robot returns to the same terrain, it's hoped that the brain will remember where the objects are, and avoid them, from memory.

Even though it sounds like an Isaac Asimov script, the point here is purely scientific, not science fiction (although somewhere, someone is surely dusting off Izzy's Three Laws of Robotics, just in case).
One of the researchers, Professor Kevin Warwick, talks about the application of a learning robot:
This new research is tremendously exciting as firstly the biological brain controls its own moving robot body, and secondly it will enable us to investigate how the brain learns and memorises its experiences. This research will move our understanding forward of how brains work, and could have a profound effect on many areas of science and medicine.
The team hopes those areas will include "a better understanding of development and of diseases and disorders which affect the brain such as Alzheimer's Disease, Parkinson’s Disease, stoke and brain injury."


Ratbot: A Q&A with a member of the research team
Posted: August 14, 2008, 12:56 PM by Brad Frenette Randomness, Robots<input name="ctl00$Main$WeblogPostTagEditableList1$ctl01" id="ctl00_Main_WeblogPostTagEditableList1_ctl01_State" value="value:%3Ca%20href%3D%22%2Fnp%2Fblogs%2Ftheampersand%2Farchive%2Ftags%2FRandomness%2Fdefault.aspx%22%20rel%3D%22tag%22%3ERandomness%3C%2Fa%3E%2C%20%3Ca%20href%3D%22%2Fnp%2Fblogs%2Ftheampersand%2Farchive%2Ftags%2FRobots%2Fdefault.aspx%22%20rel%3D%22tag%22%3ERobots%3C%2Fa%3E" type="hidden">

Yesterday, we told you about Ratbot, the world's first robot built with a biological brain.
Turns out Ratbot's already got a name, and it's much less exciting.

We reached one of the project members, Dr Lucy Chappell at the University of Reading to get more details on the robot:

Q: How long has this project been in progress?
A: The project has been going on for approximately 2 years.

Q: Please explain how the biological brain robot will learn.
A:The robot is learning by the habit of doing things. This strengthens the neural pathways in the brain. A major part of the research is learning what electrical and chemical stimulation to apply to strengthen these neural pathways.

Q: Can you expect that the brain's functions will go beyond movement?
A: Yes, movement is the only thing the researchers have allowed it to do yet, but we expect that it will go beyond movement as the project progresses.

Q: Ok, so have the researchers given the robot a name?
A: The researchers have named the robot Gordon after the University of Reading Vice Chancellor, Gordon Marshall.
 
Seems they're downplaying the aspects I can envision with this technology, virtual immortality or at the least extended longevity. If they can make it work with a rat brain on a simple machine why not someone near death (from say terminal illness or an accident) who's brain / consciousness is not yet destroyed be placed into a anthropomorphic silicone and metal machine? See the book The Electric Church by Jeff Somers for a pretty good description of the constructs I'm thinking of but basically silicone rubber skin, metal skeletons and a nervous system of cables controlled by a human brain. A purely "nervous system". Then after more research if that same brain / consciousness could be transferred to silicon which is basically "immortal" (until a better storage medium is found) humans too could achieve immortality albeit in a roundabout, eventually inorganic manner. I find this possibility amazing to think about.

1) If immortality is possible in this manner, what does that do to religion?

2) What does that do to the alcohol & drug industries? Could a silicon brain get "high" or "wasted"? Would there be codes that could give a silicon brain a psychedelic experience?

3) Could a silicon consciousness be killed by a malicious computer virus? Could a silicon consciousness be copied / downloaded / backed up?

4) At first mating would have to be done while people were still in their "organic" phase but perhaps later two silicon consciousness' could be able to combine in a "sex drive" or something and create an entirely silicon consciousness that never had an organic pre-history. Is that then still a "person"?

Are these possibilities interesting or frightening to you?
How do you answer the questions I posed?
What questions would you add to the list above of things to ask before you got made into a cyborg?
If stage one cyborg conversion ("brain in the machine") were possible in the next 20 years, it only took them 2 to do the ratbot, would you do it as your "time" was running out?
Would you only do it if it could be entirely cybernetic ("Ghost in the machine" via silicon consciousness) or would you choose an "analog death"n (i.e choose to die)?
 
I honestly believe that on a normal path of progression, biological self-evolution would dictate that we would make a conscious effort to jump from wetware to hardware.

1: Same thing alien contact would. Such an undermining concept as immortality or extraterrestrial contact would probably incite a number of riots. I can't remember what article it came from, but it said: "In all probability, when the last scientific research center in the world operating in a dystopian world on the first true god-like intelligence finally flips the switch on, the AI would awaken, saying: "I understand! The way we fix everything is to..." before being interrupted by a raving religious mob as they pull the plug"

2: Fun speculation. I did a sci-fi short story where robots worshiped PI due to its infinite nature.

3: A lot of Sci-fi books coming out right now are dealing with this subject. Cliff Pickover's "The Heaven Virus" also deals with this quite explicitly, postulating what would happen to a corrupted 'immortality chip' where people could live for all time after wetware death in a sort of psychological anarchy.

4: I think the idea of reproduction will take a much different form. A 'mix' personality as would result from sexual reproduction would probably take more of the format of individuals replicating themselves with certain basic memory patterns, and then being 'raised' by another consciousness. This gets a little further along into philosophy, but any replication that separates would be an individual being due to the effects of individual experience. So, in theory, 'populating' a silicone society would be as easy as copy-paste-run.exe

My own personal questions are consciousness-oriented. The continuity of narrative within a personal experience starts at birth and ends at death, so who knows what happens once we're able to manipulate such a nebulous and undefined concept. Although I would be comforted by knowing that my digital copy would be me in it's entirety, and my identity would be preserved, I'd be very concerned as to what happens to the 'me' itself as they drain my brain into a DVD and burn what's left of my protein self.

Most certainly. I totally plan to give myself up to whatever strange experimentation was going on at the time in the event that I was pronounced terminal. If not for 'nothing to lose' curiosity, then for my own motives, which boil down to 'i just wanna know'.

I'm not sure, I think I'd be compelled to try and maintain my original brain if for nothing else than sentimental value. I'm sure it would be something of a trend on the day such a thing in possible. While it would be an incredibly weakness practicality wise, if we're uploading people into mainframes like it's nothing at the time, I'm sure I could probably get some sort of adamantium-esque skull for my goopy companion. Also, see 2 paragraphs up.
 
C'mon people, Botanica and I can't be the ONLY dreamers / futurists here!
 
heh, post-humanist thinking isn't really a part of everyone's day, unfortunately XD
 
How did I miss this? I'll post a response tonight! I love this stuff.

xvart.
 
This doesn't seem to be very close to human-cyborg brain transplant, as far as I can tell. The techniques they're using don't sound like they're really advancing neuroelectric interfaces at all, which is what counts for hooking a mature brain to a machine. These neurons were cultured onto electrodes, and then allowed to develop; it's not really a rat brain in the robot, any more than a pile of auto parts is a car. At best, it's probably more like a rat cortex or cerebellum. Before the "brain" learned to control the robot, it probably spent a lot of it's learning ability to interface with the electrode array. The brain of a mature adult is much less plastic than freshly cultured neurons, and the hard part to wiring it up would be making the proper connections. We can't just purée some terminal patient's brain and slop it onto a petri dish (well I suppose we can, but it wouldn't do a lot of good for anyone.)
I suppose if we could clone neuronal stem cells, there might be a way to grow the proper connections... IE, implant the whole rig in this experiment, cultured neurons and all, into a mature brain and see if the stem cells bridge the gaps. We're probably several constitutional amendments away from that, though. I guess I should keep my fingers crossed for mad scientists in third-world countries.
~Joe
 
I think it's more of a matter that the fact that could even be a neurological/robotic interface at all that makes this ground-breaking. If this 'brain' could indeed learn through an external interface that is not at all its originally intended body, then, well, I'd buy stocks in these guys.

I think there's a major amount of damage that's been caused by sci-fi books/movies where absolutely spectacular things happen all the time. Then, when something even remotely amazing things happen like a bio-energy powered robot that eats sugar, or the possible discovery of a higgs-boson, or robotic assistance frames that can increase your carrying capacity by 50-200 lbs, or a pile of neurons interacting with a robotic exterior, it seems to just fade into the realm of nonchalant acceptance.

If this actually works, and they find a way to organize cultured neurons into an input-output feed that is actually capable of independent causation reasoning facilities, then, well, holy crap, Someone get this researcher a martini!
 
I don't think there's any evidence of reasoning, or even an expectation of finding it with this experiment - memorizing a map of an area and navigating it successfully is autoassociation, which is different that deduction. Calling this thing a brain is like calling the arithmetic unit in a calculator a brain. Sure, it's connected to a larger mechanism and controls it's functions, but "real brains" are more than just jumbles of neurons; they have distinct parts with specialized functions that allow them to do the things that we think of brains as doing. Neurons are surprisingly simple machines and it takes more than just a large number of them to do the really interesting stuff.
It's not a question of if it works; we already can culture all sorts of other cells onto electrodes, and we know enough about tuning neurons that we can do things like electronically controlling a rat's walking reflex. Don't get me wrong - this is a big step - but I think the headline is way over the top and the language of the article is implying things that are still a long way out. I think when they speak of learning here they're talking about weak learning - the ability to match outputs to inputs - rather than strong, or conceptual, learning. I almost see the opposite of what you describe, Botanica - I feel like the popularity of sci-fi subject matter causes press to spin up scientific progress with fantastical speculations in order to sell more news. There's no direct mention here of the problems that researchers are trying to overcome; only that it's a robot with neurons in the control circuit, where the press seem to feel it's alright to use "neuron" and "brain" interchangeably. If you look at what the researcher's hopes for the experiment are, it's to get a better picture of how neurons tune their connections, as that's the major function at hand in the progress of cognitive diseases like Alzheimer's. Nothing about thinking robots.
~Joe
 
Back
Top