This site is not maintained. Click here for the new website of Richard Dawkins.

← Robot with a Biological Brain: new research provides insights into how the brain works

Robot with a Biological Brain: new research provides insights into how the brain works - Comments

DamnDirtyApe's Avatar Comment 1 by DamnDirtyApe

Ah, Kevin Warwick... I love that crazy guy. Looks like his folks are doing some interesting work.

Hmm.. what kind of sensors are they on that little robot? Are they ultrasonics or infrareds or something like that? Its very cool how they're able to get feed from real sensors hooked into real live neurons these days.

Sun, 24 Aug 2008 13:56:00 UTC | #223727

Serdan's Avatar Comment 2 by Serdan

Ape,

The science guy said sonar sensors.

Sun, 24 Aug 2008 14:11:00 UTC | #223737

DamnDirtyApe's Avatar Comment 3 by DamnDirtyApe

Ah. Damn my feeble mass of neurons. :p

Sun, 24 Aug 2008 14:41:00 UTC | #223773

OhioLen's Avatar Comment 4 by OhioLen

My question: if "cultured neurons" have the ability to learn, does that include the ability to replicate in order to accommodate increasingly complex neural relationships?

Sun, 24 Aug 2008 14:57:00 UTC | #223785

Nova's Avatar Comment 5 by Nova

I got a tremendous sense of awe when they actually showed the robot moving, to think that robot there is controlled by a massively simplified version of how I think.

Sun, 24 Aug 2008 15:01:00 UTC | #223787

J Mac's Avatar Comment 6 by J Mac

As long as they have a robot body to stick my brain in before I get too old.

Sun, 24 Aug 2008 15:36:00 UTC | #223794

HappyPrimate's Avatar Comment 7 by HappyPrimate

Fscinating. I hope they continue to get more positive results with this research.

Sun, 24 Aug 2008 15:47:00 UTC | #223799

NewEnglandBob's Avatar Comment 8 by NewEnglandBob

. Comment #236454 by Nova:

... to think that robot there is controlled by a massively simplified version of how I think.


This is too easy of a target, so I will pass and let your minds wander on what I could have said here.

Sun, 24 Aug 2008 16:00:00 UTC | #223803

JimmyGiro's Avatar Comment 9 by JimmyGiro

Captain cyborg rides again. I bet his research students and post-docs were all hiding in the corridor trying desperately to stifle their guffaws when ever the mighty Warwick drummed up more research funding flimflam.

Sun, 24 Aug 2008 16:15:00 UTC | #223805

bslatner's Avatar Comment 10 by bslatner

Reminds me a bit of TIM from The Tomorrow People. Next we'll all be jaunting around the country side with stun guns.

Sun, 24 Aug 2008 16:34:00 UTC | #223812

Border Collie's Avatar Comment 11 by Border Collie

I was thinking poor, little retarded robot ... only a few neurons to think with ...

Sun, 24 Aug 2008 17:13:00 UTC | #223830

J Mac's Avatar Comment 12 by J Mac

"I was thinking poor, little retarded robot ... only a few neurons to think with ... "

Yet IT wont demand equal time in a class room to rant about how it was designed by an intelligent creator.

Sun, 24 Aug 2008 17:21:00 UTC | #223837

dr joneZ's Avatar Comment 13 by dr joneZ

All brains are merely the substrate on which computations are run. A few neurons evolutionarily speaking is that much better than bugger all neurons. Future brains in this kind of robot will doubtless have neurons where all of us presently have but bum fluff

What is really being constructed here is MIND, everybody. A huge experiment with an outcome surely of interest on the scale of the LHC. Whether the mind turns out to be human or not will be interesting. Better media in the future will mean better substrate and more neurons

We are fast approaching the singularity - in the Kurzweillien sense. The moment when we understand mind just in time to see it accelerate past us

The mind runs on patterns of recognition. It's like rain falling on a landscape or hot globs of ink falling on a jelly mould. A depression or impact site will then have a channelling effect for all subsequent impacts. Memory is a landscape of patterns that actually govern not only what a conscious creature does (as in the robot reacting to its environment) but in what the conscious creature SEES (as in "error" and the learning effect of the mistake.)

Very soon there will be - probably as a consequence of this research - a memory chip the size of a molecule - that can be injected into your brain - will accelerate your thinking up to a trillion times.

What sort of world will that give birth to?

Sun, 24 Aug 2008 19:02:00 UTC | #223878

Rawhard Dickins's Avatar Comment 14 by Rawhard Dickins

And there are those who think our brains alone are not capable of producing our behaviour.

Mon, 25 Aug 2008 00:29:00 UTC | #223939

Laurie Fraser's Avatar Comment 15 by Laurie Fraser

Very soon there will be - probably as a consequence of this research - a memory chip the size of a molecule - that can be injected into your brain - will accelerate your thinking up to a trillion times.


Excellent, dr joneZ! I vote we inject the first one into Joe Morreale - it might get him up to about spider monkey level.

Mon, 25 Aug 2008 01:28:00 UTC | #223948

DamnDirtyApe's Avatar Comment 16 by DamnDirtyApe

I wonder what point sentience occurs. I wonder if there is something like a kind of 'critical mass' where bam, it's got the type of connectivity and overall processing power to make complex decisions about itself, and gain the loosely defined feeling of self awareness.

I do worry about the quasi-religiosity of the singularity, but I've no doubt that making an artificial sentient being is going to happen. And soon.

We just have to make sure that people don't start worshiping them as gods. That would be so disappointing. Voltaire might have been amused though.

Mon, 25 Aug 2008 01:57:00 UTC | #223953

Billy Sands's Avatar Comment 17 by Billy Sands

here's one for the theologians to waste time on - do robots with biological brains have souls?

Mon, 25 Aug 2008 02:09:00 UTC | #223957

Laurie Fraser's Avatar Comment 18 by Laurie Fraser

I don't think they can, Billy, as they were created by imperfect, material humans, as distinct from a perfect, metaphysical being that, er, doesn't exist. (And that's a prerequisite for having a soul, apparently.)

Mon, 25 Aug 2008 02:31:00 UTC | #223963

skip's Avatar Comment 19 by skip

This was probably the most interesting article I have seen on this site. I just love robots and to think we can use brain cells to direct them is utterly amazing.

Good story

Mon, 25 Aug 2008 03:31:00 UTC | #223976

Byrnie's Avatar Comment 20 by Byrnie

Do you think it'll dream of electric sheep?

Mon, 25 Aug 2008 05:17:00 UTC | #223999

Ishruul's Avatar Comment 21 by Ishruul

We're one step away from Robocop it seem.

Mon, 25 Aug 2008 05:20:00 UTC | #224001

Fanusi Khiyal's Avatar Comment 22 by Fanusi Khiyal

This'll end well.

Don't these guys watch enough science fiction? The next thing you know they'll wheel in Davros.

Mon, 25 Aug 2008 05:22:00 UTC | #224002

Ishruul's Avatar Comment 23 by Ishruul

Better yet! Add some nanotechnology and we'll get Borgs.


Edit: Resistance is futile, just like the Muslim's motto.

Mon, 25 Aug 2008 05:47:00 UTC | #224005

wonder's Avatar Comment 24 by wonder

the most interesting part for me would be a more in-depth discussion of this one critical part of the video where he talks about how the neurons communicate with the sensors:

"that electrical activity goes through fairly complex processing steps..." to control the robot.

what i'm mainly wondering is, what elements of the "processing steps" give rise to goals? what do the neurons "work towards"? not high-level goals, of course, but in order to do useful, directed work and "learn" the neurons would need to organize their signals around some goal, avoid or seek a pattern of input of some kind. i wonder how that's implemented. fascinating stuff.

Mon, 25 Aug 2008 07:59:00 UTC | #224038

Adam Morrison's Avatar Comment 25 by Adam Morrison

This is just the first step to ... ROBO NIXON!!!

Dun dun duuuuuhhhhh

Mon, 25 Aug 2008 08:43:00 UTC | #224056

Ishruul's Avatar Comment 26 by Ishruul

24. Comment #236744 by wonder


what i'm mainly wondering is, what elements of the "processing steps" give rise to goals? what do the neurons "work towards"? not high-level goals, of course, but in order to do useful, directed work and "learn" the neurons would need to organize their signals around some goal, avoid or seek a pattern of input of some kind. i wonder how that's implemented. fascinating stuff.


Well it isn't that complicated, the brain is just like a really advance computer hard drive (keep in mind not a perfect one) and all the interaction of the neurons give birth to somewhat of a software program.

The brain got a 'DOS' program wich manage bodily fonctions and also some self-writing-learning programs to asset the environnement in wich we interact.

The only real challenge neuro-biologist-cyber-engineer really got to solve before we can all be terminator-like-cyborg (sci-fi imagery if you like) is to crack the neurons language. Because it is conflicting with the binary used by modern computer.

Still I could just be babbling crap as usual :)

Mon, 25 Aug 2008 09:14:00 UTC | #224079

eellerto's Avatar Comment 27 by eellerto

I just read this article in New Scientist which goes into a little more detail. What I understood from this is that when the robot's ultrasound sensor is activated with an object, like a wall, it sends this signal, via Bluetooth (the culture has to be kept in a sterile incubator), to the culture. This electrical signal stimulates the "brain" culture at a particular voltage. If the same cluster of cells repeatedly fire when they receive this signal over and over, then the firing of that cluster of cells can eventually, through a feedback loop, direct the robot to do something else when it senses a wall, like turn. It's pretty mindblowing. If they could somehow figure out how memories are stored from all of this, I'd be happy.

Mon, 25 Aug 2008 11:04:00 UTC | #224118

J Mac's Avatar Comment 28 by J Mac

"I wonder what point sentience occurs."

It doesn't.

That's like tracing the history of abiogenesis and evolution and asking at what point life occurs.

You can define life or sentience as some arbitrary qualifying characteristics so you can draw a line somewhere, but there is no significant change between the thing right before the line and the organism right after the line. It's just arbitrary.

Mon, 25 Aug 2008 11:13:00 UTC | #224123

popecorkyxxiv's Avatar Comment 29 by popecorkyxxiv

Deus Ex Machina! The future has come. I cannot wait to go all Ghost in the Shell, clinical immortality is almost here. Can you imagine what the world will be like after the post-human revolution.

Mon, 25 Aug 2008 11:20:00 UTC | #224127

Ishruul's Avatar Comment 30 by Ishruul

I'll be on the robot's side!

Hail the All-Metal Lord!!!!

Mon, 25 Aug 2008 11:59:00 UTC | #224141