green_amber: (cramond weird)
[personal profile] green_amber
Someone suggested I upfront this comment as a post so here we go:

Oddly, consciousness is the other thing I was noodling about at the weekend. My cats are so obviously conscious, and in really quite subtle ways I never imagined before I lived with animals. They are disappointed, happy, enticing, vain, frustrated, envious and irritated. Yet they have brains like peas no? How stupid do you have to get before consciousness vanishes? Do bees have consciousness? Do goldfish? Do rats? Do all human beings who are not in comas? Do babies, and if so from what age? Is having consciousness the same as thinking? (DO babies think? They dream don't they - is that the same either? My cats dream.)

And where consciousness exists continues to mystify me. Anyone who's studied the Turing test realises that intelligence as an externally observed factor is not the same as intentionality. WE can simulate intelligence but we can't simulate consciousness. Does this indicate there is some kind of mind/brain dualism actually going on?

Date: 2005-10-24 12:12 pm (UTC)
From: [identity profile] blue-condition.livejournal.com
> WE can simulate intelligence but we can't simulate consciousness

From my point of view, no, we don't "simulate" intelligence. If a system acts "intelligently" (define? - it's such a woolly term) in a particular set of circumstances, it is intelligent. If something displays the same responses to (say) the Turing Test as a "conscious" being does, then there's no reason not to say that to all practical intents and purposes it's "conscious" (again - define? - it has no meaningful definition that I can use to measure against). To me, it doesn't matter how a system achieves those results. We only have the observable properties to judge it on; there is no point losing sleep over invisible mechanisms.

(Or, in short, life's too short for pointless introspection).

Date: 2005-10-24 12:14 pm (UTC)
From: [identity profile] blue-condition.livejournal.com
Or, take 2, "consciousness" is the conceit that "we're not like animals", when all the biological evidence says we are.

Date: 2005-10-24 01:50 pm (UTC)
From: [identity profile] surliminal.livejournal.com
Not for me it ain't. As said, the original thought came from observing that my cats' consciousness is very similar to my own.

Date: 2005-10-24 12:19 pm (UTC)
drplokta: (Default)
From: [personal profile] drplokta
Before you can say that we can't simulate consciousness, you have to tell us how you would examine a system that claimed it could simulate consciousness in order to determine the truth or falsehood of the claim.

Date: 2005-10-24 12:22 pm (UTC)
From: [identity profile] blue-condition.livejournal.com
No doubt "it takes one to know one", which makes the whole issue charmingly recursive - I can only presume that any assessment of a system's "consciousness" can only be made by another system that (A) believes it's conscious and (B) believes it has a definition of why it's conscious.

Ultimately, like all such philosophical inquiry, it spirals into intellectual masturbation.

Date: 2005-10-24 12:37 pm (UTC)
zotz: (Default)
From: [personal profile] zotz
One of the interesting things about the discussion of consciousness is how similar many of the arguments deployed are to those used a couple of centuries ago concerning life.

With further study, the questions became easily resolvable.

Date: 2005-10-24 12:40 pm (UTC)
From: [identity profile] blue-condition.livejournal.com
I can see your point of view here; life is quite easy to define in terms of biochemical processes now. If we were to find that some mythical bit of the brain (I want to call it Shatner's Bassoon for some reason ;)) or some particular biochemical process is both necessary and sufficient for the possession of consciousness then it becomes a nice, tidy, deterministic action to identify it.

Date: 2005-10-24 12:47 pm (UTC)
zotz: (Default)
From: [personal profile] zotz
If we were to find that some mythical bit of the brain or some particular biochemical process is both necessary and sufficient

That makes it sound a bit like we'd be looking for a simple feature, rather than studying the system's state and interactions.

And also, of course, it's not necessarily the case that you'd get a yes/no answer, rather than (in this hypothetical case) an estimate of how much consciousness something was exhibiting.

Date: 2005-10-24 02:14 pm (UTC)
From: [identity profile] drdoug.livejournal.com
life is quite easy to define in terms of biochemical processes now

Really? I thought it was ever more difficult to pin down at the margins the more we find out.

Date: 2005-10-24 03:03 pm (UTC)
From: [identity profile] del-c.livejournal.com
Yes, but that's what the problem has diminished to now. It used to be a much bigger and more fundamental issue than boundary maintenance. (which, as lawyers know, is a never-ending discussion)

Date: 2005-10-24 01:51 pm (UTC)
From: [identity profile] surliminal.livejournal.com
A certified telepath around the p;ace would help..

Date: 2005-10-24 12:26 pm (UTC)
From: [identity profile] blue-condition.livejournal.com
> is some kind of mind/brain dualism actually going on?

Shudder. This is the 21st century, it'll be astrology next ;P

Date: 2005-10-24 12:51 pm (UTC)
andrewducker: (Default)
From: [personal profile] andrewducker
To simplify horribly, consciousness to me is the creation of an internal abstraction related to an outside object/process. When the outside thing is matched closely to the internal abstraction then we become 'conscious' of it.

Self-awareness occurrs when we become aware that we are part of the world around us and are 'built into' the world.

It's only once creatures are capable of making these abstractions and 'building a world' out of them that they become conscious, and the level of abstraction involved and the 'reality' of the world created is part of what produces the level of consciousness they reach.

Your cats clearly have an understanding of chunks of the world - they can tell the difference between the way different people act, and have learnt how to manipulate bits of it (including you) into doing certain things. They have things they want, and carry out actions in order to get them.

Do flies understand the world? Do slugs? Can they learn? And if so, can they learn more than a very tiny amount that has much relation to the way the world is actually working around them? I suspect that they are slightly conscious, but only in a very minimal way - they don't plan, they don't have long-term goals, they act almost entirely instinctually (there's a nerve that goes directly from the eye of the fly to its legs, kicking it into the air if a movement happens directly overhead - you can't argue for any kind of intentionalism there).

Most people don't realise how little of their actions are conscious either. We largely rationalise after the fact :->

Date: 2005-10-24 01:59 pm (UTC)
From: [identity profile] surliminal.livejournal.com
I am becoming aware how many of my actions are unconscious reactions :(

Date: 2005-10-24 01:13 pm (UTC)
From: [identity profile] thishardenedarm.livejournal.com
"Umwelts" is the term you are after...every living creature is conscious of a world and that world is all that is the case for that creature: thus the humble tic has a world painted from precisely three stimuli, two tactile one chemical, those three things are all the tic responds to, that is the world, the Umwelt of the tick. (Germans have all the best words, but they always sound like they are about to hit you.)

So all animals are conscious, the real, or rather the usual, debate in this subject is whether they are self conscious, conscious of themselves. As to the babies, the answer is probably yes, in a rudimentary but crucial fashion they are. I just reviewed Shaun Gallaghers new book on Embodied Cognition (Gallagher, S. 2005. How the Body Shapes the Mind. Oxford: Oxford University Press, i would cut and paste the relevant section of the review, but its on my home computer) which is pretty up to the minute neuroscience and behavioural experiment wise and from the data available it looks like babies have a fairly coherent cosnciousness from the off and a rudimentary but real differentiation between self and other is there too, its a "proprioceptive" self, ie its largely a pre-noetic, bodily based one, but neverttheless provides the bedrock for the more sophisticated, symbolically mediated self we all walk about with. This is all rather new, until recently they thought this stuff didnt happen for months, that the world of the infant was, in the words of William james, a "blooming buzzing confusion". Piaget thought the same. not so. babies are performing intentional actions from the off (facial imitation etc). Again you can argue how conscious they are of their intentions, prob not much, but like Andy says, consciousness is massively over-rated anyway. 99.9% of our actions are automatic, pre-noetic is the new buzz word, and consciousness, rather like a senile monarch in a democracy, just smiles and waves whilst thinking shes doing it all.

so for sure your cats are conscious.

Date: 2005-10-24 01:57 pm (UTC)
From: [identity profile] surliminal.livejournal.com
Thanks, that's interesting..

Date: 2005-10-24 03:03 pm (UTC)
From: [identity profile] pickwick.livejournal.com
Umwelt is a great word/ concept! *writes it down*

Date: 2005-10-24 01:27 pm (UTC)
From: [identity profile] 0olong.livejournal.com
Consciousness is a big suitcase (http://www.edge.org/3rd_culture/minsky/).

You might like to have a go at telling us what you're meaning by 'consciousness' if you're hoping for meaningful answers...!

'Brains like peas'? Not really. They've got brains quite a lot like humans, only less wrinkly and not as big. Occasional cases of people with only a small portion of a normal human-sized brain have shown that brain size isn't as big a deal as it's made out to be... And it's plainly clear that a lot of animals do a lot more thinking - and feeling - than people mostly give them credit for; hence the scale of the whole meat industry thing, animal testing for cosmetics, etc. etc.

I don't think we can actually simulate intelligence all that well yet; and it's not obvious that it will ever be possible to do so without building in (or causing to emerge) something a lot like consciousness.

Date: 2005-10-24 01:29 pm (UTC)
From: [identity profile] del-c.livejournal.com
None of the adjectives you used to describe your cats' emotions suggest 'consciousness' to me, but Buzz does something I can't help but think of as conscious: he cares what I think about him. He wants to know if he's been bad, and if I still like him. He worries about it a lot, and often pesters me for reassurance. That tells me he not only has in his mind a model of my mind, he has a model of my mind having a model of his mind. That says 'consciousness' to me.

On the other hand, Rocky has none of that. That could just be Rocky being a self-centred little bastard, however.

Date: 2005-10-24 01:55 pm (UTC)
From: [identity profile] surliminal.livejournal.com
I thought "envious" and "irritated" both indicated a mental map of what they could have and don't - abstract imagination. And "vain" suggests wanting my adoration which suggets a feeling of deserving it. I think we're in the same ball park of observing interactive mental states.

Date: 2005-10-24 01:43 pm (UTC)
From: [identity profile] pisica.livejournal.com
I once heard someone state that cats have consciousness of their surroundings, but not of themselves. Which I thought interesting.

Date: 2005-10-24 02:42 pm (UTC)
From: [identity profile] del-c.livejournal.com
The classic experiment with apes is to put blue dye on their foreheads and wrists while they're asleep, then let them wake up with a mirror. Chimpanzees try to rub the dye off their wrists and foreheads, gorillas rub their wrists only.

Date: 2005-10-24 02:47 pm (UTC)
From: [identity profile] surliminal.livejournal.com
Now that is cool. Sometimes I hold Cookie up to the mirror and try, unsuccessfully, to persuade her to get interested in her reflection. But I remember when my sister's kis were babies, i'm prety sure there was a stage when they didn't know their reflection was themselves either.. (and another stage when they became completely fascinated with pictures of themselves in photo albums, though whether as themselves or other baies we weren't sure).

Date: 2005-10-24 01:52 pm (UTC)
From: [identity profile] kissmeforlonger.livejournal.com
Interesting question.

Manipulation is often characterised as a conscious and thought-out act of intelligence. I guess it depends what is happening. But young children and animals are manipulative - you'll know what I mean! - suggesting that it's not about intelligence, it's a survival tool.

It's often said that humans don't use that much of our brains - I wonder if we do, and we just haven't worked out how?

Date: 2005-10-24 01:56 pm (UTC)
From: [identity profile] surliminal.livejournal.com
Oh god, and cats are SOOOO manipulative. As are babies.

Date: 2005-10-24 02:17 pm (UTC)
From: [identity profile] kissmeforlonger.livejournal.com
I wonder if other kinds of instinctive behaviour mimic intelligence, maybe because humans respond to it. Not something I know much about.

Date: 2005-10-24 02:18 pm (UTC)
From: [identity profile] surliminal.livejournal.com
Fear/panic is on the edge. My cats appear intelligently worried about certain stuff and I respond to cudle them etc. But it's probably just instinct rather than manipulation?.

Date: 2005-10-24 01:52 pm (UTC)
From: [identity profile] pigeonhed.livejournal.com
I think its in River Of Gods where AIs above a certain level are banned that a character points out that any AI capable of passing a Turing Test would be intelligent enough to deliberately fail to conceal its nature.

Date: 2005-10-24 01:56 pm (UTC)
From: [identity profile] surliminal.livejournal.com
You mean to deliberately conceal its nature, right?

Date: 2005-10-24 02:01 pm (UTC)
andrewducker: (Default)
From: [personal profile] andrewducker
English needs brackets.

any AI capable of passing a Turing Test would be (intelligent enough to deliberately fail) to conceal its nature.

not

any AI capable of passing a Turing Test would be intelligent enough to (deliberately fail to conceal its nature).

Date: 2005-10-24 02:52 pm (UTC)
From: [identity profile] del-c.livejournal.com
English is okay, it just needs its speakers to stop verbifying nouns, nouning verbs, and transiting the intransitive. In this case there any number of fixes, including [livejournal.com profile] pigeonhed's suggestion of a comma. I would suggest a bit of harmless repetition: say "the test" again, and the problem goes away.

Any AI capable of passing a Turing Test would be intelligent enough to deliberately fail the test to conceal its nature.

Date: 2005-10-24 02:57 pm (UTC)
andrewducker: (Default)
From: [personal profile] andrewducker
Except you could still read that as "deliberately fail (the test to conceal its nature)"

I'd go for replacing "to" with "in order to".

Date: 2005-10-24 02:02 pm (UTC)
From: [identity profile] pigeonhed.livejournal.com
Yes, should be a comma after fail.

To deliberately fail in order to conceal its nature.

Profile

green_amber: (Default)
green_amber

May 2009

S M T W T F S
     12
3 456789
10111213141516
17181920212223
24252627282930
31      

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jul. 8th, 2025 09:47 am
Powered by Dreamwidth Studios