Conditioning vs. Learning

[blike]

Last night i was watching this show on animal planet about the top 10 "smartest" animals. Topping them all off was this parrot who could verbally identify shapes, colors, numbers, and material. It was pretty interesting. He could even tell you how many squares were on the table. But does the parrot really know what he's doing? I can't help but think these are just conditioned responses in order to get food.

 

But if this is the case, what is real learning? I mean, I learn things in class because I have to make good grades. I need good grades to get into medical school. I would like to go to medical school to be a doctor, and I want to be a doctor for various reasons that, when it comes down to it, benefit myself and my future family. So while it may be way more complex than the parrot's situation, are we not just being conditioned for our future benefit?

[greg1917]

One could take the view that both are paths to the same end; to acquire knowledge you could as you say drum it into your head with no real understanding of its significance or purpose, whether as a reward based system or otherwise. On the other hand you could be presented with data and evidence that allowed you to conclude the same facts but in this case with an underlying knowledge of the principles behind them.

 

I wouldnt say we are undergoing conditioning in the sense you describe. Certainly in modern education it seems like conditioning because syllabi are standardised and often the simplest way to pass an exam is to memorise a set of learning outcomes. but the option to investigate these should be open to you - real learning is to arrive at the correct conclusion via investigation and reasoning. I could sit down with a dictionary and increase my vocabulary the hard way, or learn latin and understand more of the principles of language thus widening my knowledge. id say thats real learning.

[spacemanspiff]

the parrot mimicks the situation that you can find in some schools. you get the right answer but no understanding. the parrot isn't really that smart. it has just been conditioned. granted it takes a bit of brain power to even be conditioned. it does not mean the parrot has any idea what the stimuli that it is conecting mean.

 

example. pigeogns have been trained to tell the difference between piacaso and monet paintings. how much do you think the pigeon really understands? asides from (after much training) knowing the low level visual cues. of course you could ask the same thing about art buffs.

[DocBill]

Originally posted by spacemanspiff

the parrot mimicks the situation that you can find in some schools. you get the right answer but no understanding. the parrot isn't really that smart. it has just been conditioned. granted it takes a bit of brain power to even be conditioned. it does not mean the parrot has any idea what the stimuli that it is conecting mean.

 

example. pigeogns have been trained to tell the difference between piacaso and monet paintings. how much do you think the pigeon really understands? asides from (after much training) knowing the low level visual cues. of course you could ask the same thing about art buffs.

 

Is that not education does? This would go far in explaining why so many HS grads are "not really that smart."

Bill

[Ryoken]

Beacuse school work is not really that interesting, involving, challenging, and everything else intelectuals find sacred.

 

Inteligence versus (conditioned) knowledge.

[Glider]

Depends on how you look at it I 'spose. Threre really isn't any difference between conditioning and learning apart from volition. Classical and operant conditioning are just different forms of learning. They both require the laying down of associations (in these cases between stimulus > response or action > outcome, respectively). Scholarly learning also requires the laying down of associations (traces), in this case between new information and previously learned information. It also requires a degree of abstraction, which the parrot may or may not have (I don't really know), though it seems unlikely.

 

I've seen it argued that scholastic learning can be seen as a form of operant conditioning, where study is the action and good grades are positive reinforcement, bad grades are punishment and increased understanding (particularly when confronted with exams) is negative reinforcement. Of course, I think this argument depends on the student giving a toss in the first place.

[Radical Edward]

and then there is just general interest. I learn alot of stuff just because it interests me, not because it will have any effect on the outcome or path of my life whatsoever.

[Glider]

Exactly, and in this case, reward is inherent in the learning process.

[blike]

What I didn't see on the video was the parrot applying what it "knows" to different and new situations. As humans, we do this all the time. We learn a variety of things at different times and situations, yet we can apply them to new situations unlike anything we've ever seen before. I'm not sure if any apes(or animals) are known to apply concepts to new situations.

 

Also, are any animals self-aware? I've heard that dolphins may be aware that they are individuals, and I would guess that some apes would understand...

[DocBill]

I did. Second year of Cog Pscych.

 

Anywho, the volition of education for it's own sake helps explain why I am still a student. Now if only my wife would buy that..........

[Glider]

Originally posted by blike

What I didn't see on the video was the parrot applying what it "knows" to different and new situations. As humans, we do this all the time. We learn a variety of things at different times and situations, yet we can apply them to new situations unlike anything we've ever seen before. I'm not sure if any apes(or animals) are known to apply concepts to new situations.

That's the 'degree of abstraction' I mentioned. To be able to abstract newly acquired information and to mentally make the link between it and a novel situation. This is what makes the significant difference between 'tool users' and 'tool makers'. Tool users learn to use tools by watching others. Tool makers can't learn by watching others (the tool doesn't exist), so, when faced with a novel problem, they have to abstract the problem and apply learned (abstract) principles to it in the 'virtual reality' world of their mental workspace.

 

Also, are any animals self-aware? I've heard that dolphins may be aware that they are individuals, and I would guess that some apes would understand...

Hard question to answer, mainly because of the difficulties we have in agreeing a precise definition of self awareness. There was a case a little while ago, where researchers stuck a little self-adhesive red dot onto the forehead of a chimp, and then showed it a mirror. Most animals will either ignore a mirror, or treat their reflection as another animal and attack it, or run away from it or whatever. This chimp however, looked into the mirror, and reaching to its own forehead, removed the red dot. The researchers argued that this was 'proof' of self-awareness. However, others argue that recognition of of one's image as a reflection in a mirror does not necessarily indicate self awareness on the philosophical "cogito ergo sum" kind of level, i.e. as the implicit understanding of one's-'self' as an independent entity.

 

Some people argue that the ability to recognise yourself in a mirror equates to self-awareness, and others argue that true self-awareness must include an understanding of the (somewhat abstract) concept of death. For in order to be aware of self - that one exists - one must also be aware that there was a time prior to one's existence, and therefore that there will be a time after one's existence. Thus, implicit in awareness of self, must be awareness of mortality. How you'd demonstrate that in a chimp though...beats the hell out of me.

 

A lot of this problem stems from debate surrounding the nature of consciousness. We consider ourselves self-aware, but that (say some) is only by tacit agreement. Given (they say) that we can only know consciousness in ourselves, nobody else has any real proof that we are conscious, and visa versa. You know you are conscious, but what do you know about those around you? They react to stimuli (so do venus fly traps), they will, if asked, concur with you when you ask them what colour a particular flower is (the parrot can do that). The truth is, there is no way we can know what is happening in the minds of others, we can only make inferences based upon observed behaviours. Consciousness is really only a mutually agreed state, known in ourselves and assumed in others.

 

By the way, has anybody ever heard of 'the Nashville Leather Company'? I'm trying to replace a leather waistcoat ('vest' in the USA). I had a really good one, brown nubuck with satin lining, made by the N.L.C., but paramedics cut it off me while I was unconscious after smashing my Harley just before Christmas. I can't find one to match it ANYWHERE! Dagnabbit! They cut it to PIECES! It wasn't even necessary...Bastards!

[greg1917]

its so hard to determine dolphins are self aware its subjective and can be argued either way. one study i saw that interested me was where marine biologists placed a large mirror in a tank. most animals ignore or avoid the mirror as they think its another animal, but when the biologists painted a dot on the dolphin it went to the mirror and looked in it to examine the new pattern, turning this way and that to look at it. being aware of the reflection is a sign of intelligence in itself, but the dolphins then adopted a slightly different social pattern where those with dots seemed a little more confident in terms of every day interaction with their keepers. Knowing little about dolphin's social patterns to begin with i wasnt sure what to make of this in the first place.

[Radical Edward]

I wouldn't be suprised if there are quite a few animals that are aware of mortality, for example look at the way than many great apes, and even elephants treat the deceased. this can even be applied to dogs, who will often pine away when their master leaves or dies.

[Glider]

And whales.

 

But are they aware of their own mortality?

[Radical Edward]

interesting.... are any animals known to care for sick/injured members of the group/herd/pod etc?

[Glider]

Dolphins and some other toothed whales have been observed supporting sick/injured members of their pod (particularly mothers towards offspring). They swim close by, maintaining body contact and supporting the sick/injured individual at the surface (presumably so it can breathe). Elephants have been observed helping infants to their feet if they show reluctance or that they are having problems, and are reluctant to leave any individual that cannot move. Dolphins (and other whales), elephants and primates all show reluctance to leave their dead (particularly offspring). What can actually be inferred from these observed behaviours though is a question for debate.

[Radical Edward]

I would guess that looking at the way animals treat the dead is not quite as useful as how they treat the living. the way they treat the dead could be interpreted sa "not understanding why the other elephant isn't moving anymore" rather than mortality itself, wheras looking after the sick and ill seems to me to show a more in depth understanding of illness, and possibly death. Of course this is just conjecture, but it is what comes to mind. As you pointed out earlier, we can only even assume that other humans posess consciousness and so on, because as yet we don't have an understanding of what consciousness is, because the current scientific model is still rather crude.

[blike]

Dolphins and some other toothed whales have been observed supporting sick/injured members of their pod (particularly mothers towards offspring). They swim close by, maintaining body contact and supporting the sick/injured individual at the surface (presumably so it can breathe).

 

I saw this on discovery channel. It would lead me to believe that a orca (i think it was) was aware that if its young calf could not breathe it would die. Of course, this does not indicate self-awareness, but it is pretty interesting. I did some reading and I found a paper that said the 'holy-grail' test for self-awareness is suicide. Committing suicide implies that you understand your own mortality.

[Radical Edward]

Originally posted by blike

Committing suicide implies that you understand your own mortality.

 

cool, lemmings are self aware. I knew it.

[Glider]

Originally posted by blike

I did some reading and I found a paper that said the 'holy-grail' test for self-awareness is suicide. Committing suicide implies that you understand your own mortality.

That's true. But then a scorpion attaked by fire-ants will sting itself to death. There was similar thing kids used to do in Africa, they would scrape a circle in the dirt, pour something like lighter fuel into the ring, put a scorpion into the middle and light the fuel. The scorpion would circle around looking for a way to escape, and finding none would sting itself to death.

[spacemanspiff]

But then a scorpion attaked by fire-ants will sting itself to death. There was similar thing kids used to do in Africa, they would scrape a circle in the dirt, pour something like lighter fuel into the ring, put a scorpion into the middle and light the fuel. The scorpion would circle around looking for a way to escape, and finding none would sting itself to death.

 

that's frickin awesome.

 

while very interesting. is it possible that this is just another example of an animal engaging in a behavior but not really having any deep understanding. kind of like an instinct.

 

wouldn't comiting suicide and avoiding death be pretty much the same thing in terms of understanding that certain actions lead to death. in one case you avoid them and in the other you seek them out. there are plenty of animals that avoid death. wouldn't the logic of suicide=self aware also dictate that avoiding death implies that you are self aware?

[Glider]

Originally posted by spacemanspiff

that's frickin awesome.

 

while very interesting. is it possible that this is just another example of an animal engaging in a behavior but not really having any deep understanding. kind of like an instinct.

Exactly right. Of course the scorpion doesn't sting itself in any "Goodbye, cruel world" kind of way, but I think it does show how difficult it can be to avoid making assumptions based on behaviour. We can observe behaviours, that's simple enough, but we have a problem when it comes to working out the intent behind them. We can't help but draw conclusions based on second level reasoning, where 1st level reasoning is "I think" and second level reasoning is "I think he/she thinks", based on watching what they do (this is one of the things people with autism lack). So when we observe a behaviour, we can't help infering an intent from it. This is very useful within our own species, but less helpful when we're trying to work out how other species 'think'.

 

wouldn't comiting suicide and avoiding death be pretty much the same thing in terms of understanding that certain actions lead to death. in one case you avoid them and in the other you seek them out. there are plenty of animals that avoid death. wouldn't the logic of suicide=self aware also dictate that avoiding death implies that you are self aware?

Well, you'd think so, wouldn't you? But that has problems too, in that all animals work to avoid death. Even the humble amoeba when detecting changes in temperature or pH that are noxious or threaten it, will alter direction and speed to avoid it. Clams will close up if they detect something near them, sea slugs will squirt noxious or sticky substances to deter predators, flys avoid being swatted, worms will retreat down their hole if disturbed, cockroaches will run on exposure to light or vibration and so-on and so-on. Therefore, avoiding predation in and of itself can't imply an understanding of death per se. The simplest explanation is that most animals, rather than avoiding death, are avoiding the predator. Given that generally speaking, you only get caught once by a predator, this can't be the result of learning or experience, so it must be an instinctive response to try to avoid anything that's running after you.

[Skye]

That's a good point Glider, the next thing is to figure out an experiment to get around it You could teach a chimp a hand signal or picture that means death and see if it can apply it to itself. Chimps have a bit of a problem with syntax though, and of course it would be sad teaching a chimp about it's own death.

[Glider]

You're right there, though thinking about it, I can't even begin to devise an experiment that would test a chimp's understanding of death. And again you're right, in that it would be a sad thing to do to a chimp.

 

I'm sure there must be a better way to test for self-awareness. I think what's needed is to devise a test for an explicit awareness of affective state. Most animals, when (for example) angry, or frightened are simply behaving in a manner concordant with their affective and autonomic state. If we could get an animal (it'd have to be higher primate, a Chimp or Orang Utan) that could express in some way an explicit knowledge of its emotional state, that would probably indicate self-awareness. It would be compelling evidence at least.