Animal Cognition

Ed:

Does a beaver "choose" to build a dam? Does a spider "choose" to spin a web? Are there such things as web-weaving spiders -- who "choose" not to? No.

Dam building is a general behavior of beavers that is preprogrammed. But if a beaver is going to build a particular dam, it still has to make many choices. Is it the right place? What is that sound? A predator approaching? A rival? Should it flee (where?) or attack? Etc. Similar for the spider, although the repertoire of the spider is of course much more limited and more automatic than that of the beaver; it is much more sphexish to use a term coined by Douglas Hofstadter (Metamagical Themes), after the rather stereotypic behavior of the Sphex wasp.

So Cal, you are one of the Brutes Abstract Pack. Do you have evidence pertaining to this conclusion of yours? If so, post it. Let's examine it together. You will find that I can be very reasonable, when the situation calls for it.

You mean about the "only perceptual" being a simplification? For that I'll have to dig into the literature, which I don't have at hand now, so that'll take time. But perhaps you have evidence to the contrary you can show us?

This does seem intuitively true, I'll give you that. When "others" make unpredictable choices, then we do ascribe a free will to them -- but there is another way to get to evidence of free will, namely, introspection.

But it's equally true for introspection. When you're confronted with a choice between several alternatives, you "see" these before your mind's eye, you can weigh the pros and cons of each alternative and reach a decision. So you end up with a particular choice. You think this is a free choice while you can't predict in advance what it will be. Your introspection can only look as far as your conscious thoughts, but everything "under the hood", where the deterministic machine is grinding away, is invisible to you. So all the possibilities seem still to be open. But inexorably you'll arrive at that one particular choice that you were destined to make.

Post 61

Saturday, January 7, 2006 - 5:50pm

Ed,

Can a spider choose how to spin its web? Or does a given species of spider have the exact same web, from spider to spider, generation to generation -- woven perfectly on the very first try?

That's pretty low on the awareness scale, but still, the spider decides when to rest, when to work, when to eat, where to put the web, when it is large enough and so forth. It makes decisions, albeit within limitations. All of these decisions involve its survival mechanisms.

Free will is not either-or. Some acts are instinctual and others are freely chosen. I have no problem with that, since I observe it all around me.Why do you?

But going up the awareness scale a bit, wouldn't you be extremely grateful for the volitional capacity of a lion if you were unarmed and came up against one and it decided not to eat you at that moment? Or would you prefer the predictable programmed beast?

Michael

Post 62

Sunday, January 8, 2006 - 5:32am

MSK:

That's pretty low on the awareness scale, but still, the spider decides when to rest, when to work, when to eat, where to put the web, when it is large enough and so forth. It makes decisions, albeit within limitations.

How do you know that? What is your evidence?

Certainly the spider sometimes rests and sometimes works but how could you possibly know that these were choices? A computer program could be said to make "decisions" but it doesn't choose.

Beware of anthropomorphism!

Post 63

Sunday, January 8, 2006 - 9:25am

So when Kasparov makes a move, this is a choice between several possible moves, and when Deep Blue makes a move it's a decision between several possible moves? Strange...

Post 64

Sunday, January 8, 2006 - 11:24am

Rick,

LOLOLOLOL...

How would you define "to choose" then? Something only humans can do?

Michael

Post 65

Monday, January 9, 2006 - 7:58pm

You asked ...

Choose:
1 a : to select freely and after consideration

Decide:
1 a : to arrive at a solution that ends uncertainty or dispute about <decide what to do>

Instinct:
1 : a natural or inherent aptitude, impulse, or capacity <had an instinct for the right word>

2 a : a largely inheritable and unalterable tendency of an organism to make a complex and specific response to environmental stimuli without involving reason

b : behavior that is mediated by reactions below the conscious level

Ed
[m-w.com]

Post 66

Monday, January 9, 2006 - 11:13pm

Ed,

I stand by doggy.

Doggy is sleepy (alternative 1). Doggy sees and smells food (alternative 2). Tail thumps (consideration). Doggy decides he is more tired than hungry, so he closes eyes and drifts off to sleep (choice).

Here are two definitions for you (American Heritage Dictionary, Second College Edition):

Choose - To select from a number of possible alternatives; decide upon and pick out.

Free will - 1. The ability or discretion to choose, free will. 2. The power, attributed especially to human beings, of making free choices that are unconstrained by external circumstances or by necessity.

I see animals doing that. I don't see dice or algorithms doing that. (Dice merely fall where they and may and algorithms fall within preprogrammed randomness rules.)

Also, notice that the phrase given by the dictionary was "attributed especially to human beings," not "attributed exclusively to human beings."

Michael

(Edited by Michael Stuart Kelly on 1/09, 11:14pm)

Post 67

Tuesday, January 10, 2006 - 10:35am

Michael:

I see animals doing that. I don't see dice or algorithms doing that. (Dice merely fall where they and may and algorithms fall within preprogrammed randomness rules.)

Algorithms may choose without randomness rules, just by weighing the pros and cons of the alternatives. Why should Deep Blue not be able to choose? Only while we know that it is a deterministic machine? After all it has passed the chess Turing test with flying colors (at a certain moment Kasparov couldn't believe that there wasn't some human player involved in DB's moves)! If we wouldn't be prejudiced by our knowledge about its interior, we wouldn't hesitate to conclude that it could choose very well. Pure discrimination of machines!

And if dice and coins can't choose, why do we use them sometimes just for that purpose? No, we don't expect a choice that considers all the pros and cons of the different alternatives, but that isn't always the best way to choose, sometimes a random choice may be the better strategy. And they're better at it than we.

Post 68

Tuesday, January 10, 2006 - 11:07am

On guard, Michael! [dictionary wars]

m-w.com ...

Main Entry: free will
Function: noun
1 : voluntary choice or decision <I do this of my own free will>

2 : freedom of humans to make choices that are not determined by prior causes or by divine intervention

Also, notice that the phrase given by the dictionary was "freedom of humans," not "freedom of all living beings."

Marshalling evidence-without-argument is better than nothing, but when evidence-without-argument is refuted with a mere marshalling of contradictory evidence (as I have done here) -- then evidence-without-argument is not any better than nothing, anymore (you've been successfully, objectively, refuted now).

So, the dictionary battle is at least a truce, if not a victory for me. On to your cute example ...

Tail thumps (consideration). Doggy decides he is more tired than hungry, so he closes eyes and drifts off to sleep (choice).

Tail thumps --> "consideration"?

That is not a sound inference. You're ascribing a particular mental activity to the dog, one that humans perform. What is needed for proof of this unseen mental activity is an EEG and a PET-scan of the dog's brain -- to check for brain activity congruence. If this is not yet done, then your assertion is, entirely, unfounded.

Doggy decides he is more tired than hungry --> "choice"?

Hell, even I don't always "decide" to fall asleep (I'm just so damned tired that I crash). So if humans don't always decide when to fall asleep, then this example is a bad example -- it has an internal inconsistency (with regard to falling asleep, the "wide" concept of choice doesn't even refer to reality).

Ed
[does a Venus Fly-Trap "choose" to capture prey?]

Post 69

Tuesday, January 10, 2006 - 11:30am

Dragonfly,

A person who makes a choice by using dice or chance is making a choice not not choose. He has chosen to leave it up to wherever chance falls. Deep Blue is certainly a magnificent creation, but it still lacks the basic essential of being alive and valuing. Deep Blue is unaffected on a survival basis - it neither gains nor loses anything - by the outcome of any particular game because the "it" is only a machine. It can't die. It can only be unplugged.

Ed,

What the dictionary thing shows is that this is not a settled issue according to widespread usage. That would favor a non-exclusive for humans approach. I didn't check, but I would wager that your dictionary has more than one definition, also.

In your analysis of going to sleep, we are not talking about doggy being dog-tired, so much so that he crashes from exhaustion, but doggy being too lazy to get up. Big difference. He doesn't so much choose to sleep. He chooses to not get up and get the food and let sleep come.

My evidence is observation. If you want to run EEG and PET-scans of the dog's brain to check for whether or not he is choosing, I would suggest you first come up with what one of those things looks like for a human being choosing so as to set a standard.

But this is getting silly. I am not going to discuss whether the world is round anymore. The world is round. And animals and other living creatures with a brain make choices. (Plants don't, by the way, because they don't have a brain. They merely react to sensations.) A universe where human beings are disconnected completely from the principles governing other life forms is too Kantian and arbitrary for me. (Sorry, but I have far more important things to do than go around and around about the obvious...)

Michael

(Edited by Michael Stuart Kelly on 1/10, 11:37am)

Post 70

Tuesday, January 10, 2006 - 11:59am

Fine then, Michael.

Go ahead and turn away from my (and some experts') evidence and reasoning -- because you have a "feeling" about the subject matter ...

Pavlov on the conditioned reflex method and its limitations. Am J Psychol. 1995 Winter;108(4):575-88.

The physiological approach (a) avoids anthropomorphizing or speculations about the dogs' subjective experiences, and (b) permits the explanation of observed phenomena which the subjective method is not capable of doing. Pavlov realized that the conditioned reflex method has a limitation; it cannot be used in the study of human subjects because their thinking interferes with experimental results.

Post 71

Tuesday, January 10, 2006 - 2:26pm

Good, Ed - am surprised Pavlov hadn't been brought in brfore.

Post 72

Tuesday, January 10, 2006 - 2:52pm

Ed,

I like you a lot. (A whole lot.)

but...

(yawn)

Next thing you know, people will be claiming that animals don't really sleep because we don't have EEG and PET-scans of their brains.

(btw - I think my parent's dog is dreaming about me right now. I guess I'm just an incorrigible anthropromorphizer.)

Michael

Post 73

Tuesday, January 10, 2006 - 3:07pm

Michael:

A person who makes a choice by using dice or chance is making a choice not not choose. He has chosen to leave it up to wherever chance falls.

But is the reason for making a particular choice essential for the definition of choice? Suppose you ask me to choose between a number of alternatives. You don't know that I've no particular reason to prefer one alternative over the other ones. I could use a coin or a die to make my choice, but then it would be obvious to you that I really didn't care, so I try to make a random choice in my head, by thinking of a simple algorithm to get a particular number. It shouldn't be too difficult to calculate that number in my head, but it shouldn't be so easy that I could predict in advance the outcome. I could for example decide to choose the 10th digit of the decimal expansion of pi (I still know a few dozen digits from the expansion I memorized when I was a student long ago). Then I make my choice. I suppose you wouldn't say then: hey, that's no choice! You left it to chance!, because you just can't know how I arrived at that particular choice. Is there any essential difference between this scenario and the scenario where I use the die or the coin to make my choice? And what about a mixed case, for example I dismiss options A, C and E but have no preference between B and D, so I throw a mental coin to decide? How can we know whether someone's choice is a "real" choice or just some random decision? I think it's therefore not useful to include the method of choosing into the definition of "choice". You don't hesitate to grant a spider the possibility of choosing (neither do I), but do you know how it makes its choice?

Deep Blue is certainly a magnificent creation, but it still lacks the basic essential of being alive and valuing. Deep Blue is unaffected on a survival basis - it neither gains nor loses anything - by the outcome of any particular game because the "it" is only a machine. It can't die. It can only be unplugged.

Now you define choosing as something that can only be done by living beings. Why? For the choices in a chess match it doesn't matter one whit whether Deep Blue can die or not. Just as I can disguise my random choosing, DB can be disguised, for example by playing at a distance, via a computer terminal. In that case you can't distinguish its "pseudo-choices" from "real choices". Of course DB's choices are limited to chess moves only, but why should the range of possible choices be included in the definition of choice? I think it's more meaningful to omit all references to the mechanism behind choosing from the definition of "choice". That is also the strong point behind the idea of the Turing test, it doesn't make any implicit assumptions about the intelligent object that is tested, it looks only at the way this performs. The proof of the pudding is in the eating.

Post 74

Tuesday, January 10, 2006 - 3:23pm

MSK,

I like you a lot. (A whole lot.)

:-))

Next thing you know, people will be claiming that animals don't really sleep because we don't have EEG and PET-scans of their brains.

Ahh, that's not fair, you don't need conceptual awareness to sleep (but I do say that you need it to "choose").

(btw - I think my parent's dog is dreaming about me right now. I guess I'm just an incorrigible anthropromorphizer.)

Incorrigible anthrop[r]omorphizer, uh? Alright. Basically, what this means, is that I'll need hit-you-over-the-head evidence then -- which may take some time to marshall. Well, until then ...

Ed

Post 75

Tuesday, January 10, 2006 - 3:29pm

Dragonfly,

With the dice, I was referring o the astounding allegation from before on another thread that the dice themselves had free will and made choices, not the people throwing them. I can't say with 100% precision what anyone thinks without something like ESP, and that is a dirty word with hard-core Objectivists. (I am not claiming whether it exists or not.) All I know with 100% certainty is that they make the choice to throw the dice. Why they do is speculation.

The two positions being debated around here are: (1) inanimate objects have free will, and (2) only human beings have free will. Context is completely dropped (like having a brain) and the subject becomes either-or according to an arbitrary criterion, not a reality-based one.

On Big Blue, how can I say this? Can you imagine it having a hissy fit and losing on purpose like people do? No. It is programmed to always take the best solution among alternatives according to calculations that ultimately reduce to O and 1. Technically, by selecting one move from several alternatives (or even one strategy from several), it is choosing, but the concept is not the same as exercising free will like a living being does. It is a program that is running, not a living being that needs to act to survive.

Michael

Post 76

Tuesday, January 10, 2006 - 4:51pm

Michael:

With the dice, I was referring o the astounding allegation from before on another thread that the dice themselves had free will and made choices, not the people throwing them.

Well, I didn't say that, dice are objects that in themselves can't make choices, but they can be part of a system that makes choices, whether that system is a human being or a machine.

I can't say with 100% precision what anyone thinks without something like ESP, and that is a dirty word with hard-core Objectivists. (I am not claiming whether it exists or not.)

It doesn't exist.

All I know with 100% certainty is that they make the choice to throw the dice. Why they do is speculation.

But if they throw the dice mentally? Their choice would be as random as if they'd thrown real dice, but you wouldn't know it.

The two positions being debated around here are: (1) inanimate objects have free will, and (2) only human beings have free will. Context is completely dropped (like having a brain) and the subject becomes either-or according to an arbitrary criterion, not a reality-based one.

Is a machine an inanimate object? It does have a purpose. But would it only be able to choose if the ability to choose would be necessary for its survival? That seems also an arbitrary criterion to me.

On Big Blue, how can I say this? Can you imagine it having a hissy fit and losing on purpose like people do? No. It is programmed to always take the best solution among alternatives according to calculations that ultimately reduce to O and 1.

I'll repeat my question; why should the range of possible choices be included in the definition of choice? And can't we reduce human thought also ultimately to 0 and 1 (to fire or not to fire a neuron)?

Technically, by selecting one move from several alternatives (or even one strategy from several), it is choosing,

Ah, I've finally established a beachhead...

but the concept is not the same as exercising free will like a living being does. It is a program that is running, not a living being that needs to act to survive.

The need to survive may be important for the kind of choices you make, but not for the possibility of choosing.

Post 77

Tuesday, January 10, 2006 - 5:43pm

Dragonfly,

One question. Why do wish there to be no difference at all between the acts of a living organism and a non-living one?

If you don't like the word, choice, certainly there must be a word that fits the concept I am talking about.

(This is different than the human/animal division, where the issue is not kind, but degree of use of volitional consciousness stemming from percept/concept integrating capacities.)

btw - I HATE candy-striped posts. I go into a catatonic non-volitional state reading them. Focused arguments are much better for me. (I don't mean that aggressively - it's just that I really do HATE that.)

Michael
(Edited by Michael Stuart Kelly on 1/10, 5:45pm)

Post 78

Wednesday, January 11, 2006 - 9:03am

Michael, it seems you have a tiger-complex (or is it a tabby-complex?), so I'll give an unstriped reply especially for you.

I don't say that there is no difference at all between the acts of a living organism and a non-living one, only that the difference is a matter of degree. I think it's also more useful to look for unifying principles if you want to understand such issues as free will and choice, than to separate them as something uniquely human (or animal in your case), because that is then end of the scientific story (religious people no doubt like that). After all humans have evolved from other animals, which ultimately have evolved from non-living matter, so unless you believe in supernatural creation or in miracle mutations, things like human free will and choice must have evolved from simpler versions in other mechanisms.

Who's afraid of reductionism? There is an enormous difference between a human being and a bacterium. Nevertheless they are based on the same DNA/RNA system and they have a common ancestor. What could be more different than a pebble I throw and the moon moving along the sky? Yet the movement of both is described by one single law, as Newton realized. And what is the difference between an ant and a machine? Sure, ants can reproduce (although worker ants can't, but never mind), and that's a trick machines still have to learn, but what's an ant else than a very sophisticated machine that still has to obey the laws of physics and chemistry, without any miracle stuff? The ultimate goal of the ant is to propagate its genes to a next generation, and that is a difference with the average machine, but isn't the possibility to choose more fundamental than the particular reason (survival of the genome versus other reasons) for making choices?

Post 79

Wednesday, January 11, 2006 - 11:42am