Toilet instinct

[Jeff Corey]

Dr. Peven Stinker of Haavahd Uni . has recently proved that humans have an instinct to micturate and defecate in certain places, viz, the loo or behind some shrubbery. Our reporter, Jeff Coprophage covered the story.

Dr. Stinker, what proof do you have that infants have this potty instinct?

Well, Brian, it's not just a potty instinct, it includes other things, such as
swimming pools. Humans evolved in such a fashion that they have hard wired modules in some mysterious part of the brains whereby they instinctively knew they just couldn't micturate and defecate all over the place. What would the neighbors say?

And your research in this area?...

Well, Brian, it takes me into some very exotic climes. Once, in Bororoaveria, I happenened upon a tribe of microcephalics that insisted on using ducks' necks to wipe their bums.

Never read any Rabelais, did they?

Never, not a bit. So it must have been innate.

Well, then there you have it, Dr. Stinker.
Now we will take some calls from our viewers

[Dymanic]

Dr Stinker, how would you answer claims that humans exhibiting such behaviour are simply responding to the call of nurture?

[Jeff Corey]

Dr Stinker, how would you answer claims that humans exhibiting such behaviour are simply responding to the call of nurture?
Pardon, I am not Dr. Stinker. He is a dead person. Lucky for you, I can channel him through my spirit guide named Bobby.
ZZZZZZZZ Ach du lieber.
Bobby sez that Dr. Stinker nurtures no explanations that don't involve nature.

[Cuddles]

Well, most animals show some kind of toilet instinct, and if you've ever tried teaching a cat to do what you tell it it should be fairly obvious that they can't have learnt it.

[Geckko]

I once saw an interesting program on our revulsion instinct, which must be related. It showed thorugh some excellent experiments how people find certain sensory cues revolting. These were commonly had associations with hygene (rank food etc.)

[robinson]

The debate, well, fist fight really, over the existence of instincts in humans, is a long bloody battle. I don't know why, but it sometimes seems like a religious war in the level of hostility the participants exhibit.

Rather than discuss evidence, research, experience and critical thinking about these matters, the debate reaches the utter low of personal attacks and venom one often sees in the gutter brawls over matters of faith.

[Piggy]

Yeah, I bit. Y'all know I have a few hot buttons on me.

But I respect Dr. Pinker a lot, and what he's done.

This thread attempts to portray him as a crank who merely slaps an "instinct" label on things and tries to pass it off as an explanation.

That has nothing to do with reality.

[EternalSceptic]

Well, most animals show some kind of toilet instinct, and if you've ever tried teaching a cat to do what you tell it it should be fairly obvious that they can't have learnt it.

They _know it_

My two cats _always_ use that part of my garden where I have the most fragile seed as their toilet. And I am sure if I had a valuable carpet the would use it. duh

[Piggy]

The "original" thread suffers the innate failing of many an online discussion, heavy with opinion and light on scientific evidence and reasonable discussion, with that ever present tendency to repeat something, as if sheer repetition will beat down the opposition with nothing more than overwhelming fatigue and boredom.

Personally, I don't find that necessarily a failing. The fact that all the back-and-forth over terms and turf is happening there at all is evidence that some common ground must be sought somehow.

I think enough research has indeed been posted there to support the claim -- NSL, Yang's work, Pinker's work, infant phoneme discrimination studies, overview of the development of transformational grammar theory, and so forth.

The trouble is, we keep getting derailed onto arguments over who's using the proper terms for what, and high-level battles over linguistics v. general learning theory or behaviorism.

Most of the repetition has been an effort, seems to me, to reword things in terms the other side can understand and accept. And also, the bumps and tweaks have helped me correct faults in how I saw the issue at first.

I've proposed a pretty simple atheoretical approach toward coming to a consensus -- or agreeing that we disagree -- but so far haven't been taken up on it. :(

[Piggy]

Ahh, well that little factoid probably escaped most of us, not being up on it all. A surgical strike perhaps, aimed at one specific target, most of us blissfully unaware of it all.

I see. But yeah, that's what that strange OP is up there. A snipe at one of the most respected, productive, influential researchers in the field.

[robinson]

Have you done any research on Pinker? There is a plethora of well reasoned criticisms aimed at his work, and even at the man himself. Are you unaware of this?

[Piggy]

Have you done any research on Pinker? There is a plethora of well reasoned criticisms aimed at his work, and even at the man himself. Are you unaware of this?

Oh, sure. I'm not a dedicated Pinkerista. Personally, I find him rather overdedicated to a particularly narrow "evo-bio" model.

I would be the last person to simply accept his work -- or anyone else's, for that matter -- just b/c it has his name on it.

But even if a lot of what he's proposing turns out to be wrong in the end, he ain't no hack. And he's not guilty of what JC is accusing him of, imo.

Even Einstein, peace be upon him, was dead wrong about some really important stuff. But that doesn't mean he was a fool, as JC obviously takes Pinker to be.

[JoeTheJuggler]

The debate, well, fist fight really, over the existence of instincts in humans, is a long bloody battle. I don't know why, but it sometimes seems like a religious war in the level of hostility the participants exhibit.

Rather than discuss evidence, research, experience and critical thinking about these matters, the debate reaches the utter low of personal attacks and venom one often sees in the gutter brawls over matters of faith.

Surely you're not saying this can stop being a kosher debate?

[Skeptic Ginger]

Of course we don't live among $h!+ piles, it's a God rule. (http://www.thebricktestament.com/the_law/camp_defecation/dt23_14.html)

-

[robinson]

But that doesn't mean he was a fool, as JC obviously takes Pinker to be.

Jeff,

Do you think Pinker is a fool?


Surely you're not saying this can stop being a kosher debate?

I don't even know what that means!

[Jeff Corey]

Jeff,

Do you think Pinker is a fool?
No, never said so. And I think there was a scatload of overreaction to a bit of satire.
Makes me no never-mind.

[Piggy]

No, never said so. And I think there was a scatload of overreaction to a bit of satire

Ok, so what's the point of the satire? What exactly are you lampooning?

In the other thread, you said I wasn't making enough effort to understand your meaning, and perhaps that's true.

I don't want to make this personal -- I don't think you posted this for my benefit -- but I would like to know... why? What's the point?

Especially since this is posted in the Science forum rather than Humor.

[Jeff Corey]

It was a lampoon of nativist linguists who describe language acquisition without being in the trenches, with kids that don't get it as easily as my kids did. I'd come home from working with slow, really slow, children and see my kids picking stuff up rapidly, and think that I could now see what was going on in full forward after seeing slo-mo.
Anyone who doesn't think that verbal behavior in infants is not shaped by reinforcing successive approximations hasn't observed kids.
One problem is that the critics think that reinforcement is what they think is a "reward".
Nope. It can be any stimulus which occurs after a response which increases the probability of that response. Rubbing my cat on the head after he stands up works. So does a paycheck after you work. So does hugging your kid after he says , Da", in England.
With verbal behavior, there are a multitude of reinforcers.
Self reinforcement, from feedback. External, from people around you. So my point here is that a lot of people, like Dr. K, don't know scat about any of this and put down "simple S-R behaviorism" without any scatological idea of what their talking about.

[Skeptic Ginger]

How are you guys distinguishing the difference between language and verbal/vocabulary skills? Obviously some are better at verbal/vocab and genetics and learning both play a part. How can you distinguish between the two?

[Jeff Corey]

How are you guys distinguishing the difference between language and verbal/vocabulary skills? Obviously some are better at verbal/vocab and genetics and learning both play a part. How can you distinguish between the two?

I can't. Buy I can teach kids to talk no matter what the cunning linguists say.

[Mercutio]

FWIW, I have been avoiding Piggy's thread because he has explicitly asked for it not to spiral into a behaviorist debate. There have been several posts that were, as he so vividly put it, "the flag before the bull". If I were to respond there, I would (perhaps) be chastised for bringing the behaviorist argument where it explicitly was not wanted. In this thread, I see that if I bring the argument elsewhere, I may be seen as forfeiting that thread. Lose-lose. (IOW, had Corey posted his OP in that thread, would it not be attacked as inappropriate there?)

DrK claims that studies show that language arises even in the absence of reinforcement, but also demonstrates, proudly, that his conception of reinforcement is decades out of date. Shall we accept his claim? Can he know that language develops without reinforcement if he does not know what reinforcement is? I posted a study (out of date, of course) that showed the vast difference in interpretation (was X an example of reinforcement or not?) between learning theorists and other linguists. It is quite possible that DrK's attack on behaviorism is like Kleinman's on Natural Selection--utterly ignorant.

If so, he would (sadly) not be the first. From Castagnaro (2006), in the Journal of Applied Linguistics, "Audiolingual Method and Behaviorism: From Misunderstanding to Myth", the abstract:This article contends that the modern descendant of B. F. Skinner's experimental analysis of behavior, ‘behavior analysis,’ and as well his 1957 masterwork Verbal Behavior, have rarely if ever been seriously contemplated by applied linguists for possible contributions to the field. Rather, a pat literature of dismissal has developed that justifies itself on (a) a fictitious link between the audiolingual method and undifferentiated behaviorism, and/or (b) a demonstrably erroneous notion that operant psychology is too simplistic to effectively take up language issues. In reality, behavior analysis is alive, well, and making significant contributions in applied language settings, but not typically in the second language area. I had hoped that DrK would not be so willfully dismissive of evidence, but so be it. His merger of Radical Behaviorism with other schools of behaviorism is quite wrong, but he does not see it. I don't admire that in Kleinman, I can't bring myself to admire it in Dr. K, no matter how much I admire the man in other respects.

[Piggy]

Hi, Merc. I'm gonna do my own little hijack here, just for a mo -- hey, you owe me. ;) I'm actually outta this thread, but saw you'd posted, so....

FWIW, I have been avoiding Piggy's thread because he has explicitly asked for it not to spiral into a behaviorist debate. There have been several posts that were, as he so vividly put it, "the flag before the bull". If I were to respond there, I would (perhaps) be chastised for bringing the behaviorist argument where it explicitly was not wanted.
Nah, I welcome all sides, as long as we were talking about the question at hand, but y'know, folks get to arguing more generally about the validity of their fields, and then suddenly we're talking about something else.

But anyway, it looks like that thread may have run its course. Thanks for the cites, btw, I'm gonna read them over the weekend.

DrK claims that studies show that language arises even in the absence of reinforcement, but also demonstrates, proudly, that his conception of reinforcement is decades out of date.
And see, this is what I mean.

Who gives a damn what we call it? Why do we have to argue over terms?

I mean, Ichneumonwasp and I, for example, were both on board saying, sure, there's gotta be an environment which is rich in what a behaviorist (as we understand it) would call reinforcement. If not, we'd all automatically speak the same language, like all spiders of one species spin the same web. But -- as Pinker says clearly, btw, and which I quoted him as saying in the thread -- it's an interaction. The little kernel programs that run in our brains require the environment to produce results. Different linguistic environments produce kids who speak different languages. It's not all built-in, and the bit that is built-in is not sufficient to produce a result. Genie and Victor are proof of that.

The question is, do we agree about what's happening? I suspect we agree more than the various domain languages may permit us to see.

I'm not going to address the crux of your post regarding behaviorism per se -- which I'm glad you brought here, and thanks for that, I do appreciate it -- because as I've said, I'm not specifically interested in that debate.

I just wish that the various camps could stop squabbling over who calls what which, and approach a question by trying to agree on some terms, and deciding what the pertinent issues really are, and then examining the fact and seeing if we agree, or don't, about what's going on. And if we don't, ask why.

But that's just for my little thread. Seems this one is open to a higher-level debate about the validity of the underlying bases of the various theories. And that's fine.

I won't be following it, but best of luck.

Thanks again for the cites.

[JoeTheJuggler]

Surely you're not saying this can stop being a kosher debate?


I don't even know what that means!

It was a pathetic attempt at a joke--"kosher" meaning Jewish laws of cleanliness and "kosher" figuratively mean proper or decorous.

[robinson]

It was a pathetic attempt at a joke--"kosher" meaning Jewish laws of cleanliness and "kosher" figuratively mean proper or decorous.

OK, I understand Kosher, I just couldn't wrap my mind around a debate about the toilet instinct to be considered "Kosher" on any level.

heh

Sometimes it seems there is an instinct to crap all over other peoples ideas and thoughts. Maybe I should start a thread about instincts. We can bash each other about the head and shoulders with blunt instruments, all the while crying about social injustice and thread hijackings.

:wackylaugh:

[Piggy]

We can bash each other about the head and shoulders with blunt instruments, all the while crying about social injustice and thread hijackings.
And bellyaching about other people's posting styles. ;) :)

[Ichneumonwasp]

Well, Jeff, I found it amusing.

Self reinforcement, from feedback.

I know this belongs in the other thread, but that is completely bogged down in ridiculousness now. This is Piggy's point actually (at least the way I interpret it), that language serves as a self-reinforcer (and that is what I tried to argue with Dr. Kitten). Exposure to language is what serves as the impetus for change, yes, yes. That is what is different about those 'instincts' (I continually put the word in square quotes because in the human I don't think it makes sense to speak of much behavior independent of nurture and I think too many people are caught up on that term and think those arguing for a slight nativist position view 'instinct' in human as absolutely hard-wired) like learning language, standing, fear of snakes, etc. There are simply certain pathways that permit ease of learning, in part, because the stimulus itself serves as reinforcer (it's a bit more complicated for the fear of snake story, but I hope the point is obvious). For language I think this is largely the case for the reason above that you cite -- it is self-reinforcing. Learning to drive is not self-reinforcing. Getting there serves as a reinforcer, but not the exposure to driving. Piggy expressly said that learning had to be a part of the process (I can't imagine anyone who would think otherwise). But there must be something other than a general problem solver in place for language acquisition to occur the way it does. That wiring can be incomplete or 'broken' in some people so that they have difficulty acquiring language and need to use more explicit instruction (we know that folks with altered FOXP2 cannot learn language the way that those with the more frequent allele --in humans-- do). From Kant on we have recognized that there must be categories in place for learning ever to occur. The weight of evidence simply suggests that the 'category' for language acquisition has this aspect -- learning is very easy for most people and language itself seems to serve as the reinforcer. I agree wholeheartedly with this. At some point, however, the self-reinforcing aspect of language seems to turn off, probably represented by the inactivation of some input or down regulation of some receptor (? dopamine receptors on the neurons responsible for the learning process?) so that language acquisition becomes more difficult.

Like I said in that other thread, I think you guys want more emphasis on the learning aspects. I have no problem with that at all. I don't see any way to separate out nature/nurture issues, so I think it is wiser for us, I think, to simply accept that both are necessary. No nature without nurture and no nurture without nature.

[Ichneumonwasp]

If I can use another analogy, even though I know many people hate to think of humans having any instincts and we are definitely not ducks, but think of the imprinting process with ducks.

First caveat, I do not mean this to represent any aspect of language except the critical period and how I think the self-reinforcement of langauage might work.

Now, I don't recall all of the details in the duck imprinting story, but from what I do recall there seem to be two inputs necessary for a duck to imprint quickly on a 'parent' -- something that moves (visual input) and a species specific call (auditory input) during a narrowly confined critical period. There is a certain brain area in chicks (generally in the area of the striatum if memory serves and generally on the left side) that is necessary for imprinting to occur. Damage the area and no imprinting is possible. Apparently GABA is released during the imprinting process and is necessary for the imprint to occur. Give GABA blockers and the process cannot occur. Sometime shortly after a chick imprints on something, GABA receptors on the cells in this area near the striatum are down-regulated (a genetic switch is thrown), so no further imprints are possible.

I don't see why language acquisition can't be explained in an analagous fashion -- upregulate dopamine receptors on the cells responsible for learning language so that language itself serves as a self-reinforcer. Then down-regulate these receptors later so that we need explicit instruction for learning a second language later in life. This would certainly explain the critical period and would also explain how it is that we can still acquire language later -- we only turn off its self-reinforcing aspect but leave behind any other reinforcers that we can use and the wiring for learning language is still intact.

[Jeff Corey]

Self reinforcement is only part of the picture. Early language in the form of mands gets us reinforcers ("Juice, please") and attention from other people is social reinforcement. And "Differential reinforcement resulted in accelerating frequency of production of spontaneous foreign language use and better performance on both production and comprehension tests"(p.430) in 2 and 3 year olds.
Whitehurst, G.J. and Valdez-Menchaca, M.C.(1988) What is the role of reinforcement in early language acquisition? Child Development, 59, 430-40.
Russ Whitehurst was a trained behavior analyst who knew how to reinforce verbal behavior.
The less said about his later career the better.

[Ichneumonwasp]

Self reinforcement is only part of the picture. Early language in the form of mands gets us reinforcers ("Juice, please") and attention from other people is social reinforcement. And "Differential reinforcement resulted in accelerating frequency of production of spontaneous foreign language use and better performance on both production and comprehension tests"(p.430) in 2 and 3 year olds.
Whitehurst, G.J. and Valdez-Menchaca, M.C.(1988) What is the role of reinforcement in early language acquisition? Child Development, 59, 430-40.
Russ Whitehurst was a trained behavior analyst who knew how to reinforce verbal behavior.
The less said about his later career the better.

Oh, I agree completely, but I think the self-reinforcing aspect explains much of what we see from the 'nativist' side of the argument. I didn't mean what I said to exclude other learning processes, only that I see more middle ground in this debate than I am hearing.

[robinson]

The issue of "human instincts" is probably one of the most interesting topics. But damn, can people get hot under the collar quick over it. The first thing I notice, when "instincts" are brought into a conversation/debate/fight-to-the-death, is that the DEFINITION of "instinct" becomes of overwhelming importance. There is no doubt that the definition of a word, in this case, "instinct", is of absolute importance, because the very heart of the matter, is a fight over whether "it" exist or not. "It" being "an instinct", leading to the very real, but still ironic, situation, where we are fighting over something that is called "instinct", which is a word that means something, but what that something is, is defined by the people fighting over whether it exist or not. While some may resist the obvious call for "define your terms", before the conversation goes forth, it seems more than a semantic point in this case.

If you go with the view that the use of a word determines its meaning, and enough people agree on a meaning, then a word does mean what the group says it does, if you go with that view of things, (wikiality), then "instinct" means whatever somebody wants it to mean, which is just about the dumbest thing I can imagine happening.

A common understanding of terms is essential for scientific progress, (as well as business in general), and language is often a barrier to communication, rather than a great force to bring us together. This is not by chance. Specialized languages, within a common tongue, are used as much to prevent outsiders from understanding what is being said, as they are a sign of being in-the-know, an essential factor in many human enterprises, both business as well as military in nature.

Still with me? Maybe I missed it, but it seems in the discussion about language, instincts, learning, etc etc, a salient point would be the learning/use of a language, in written form. Of course this might just be clouding the waters, but language seems to be limited to speech and cognition, while ignoring the more complex, and very important task, or learning a language, in the written form.

[Ichneumonwasp]

The issue of "human instincts" is probably one of the most interesting topics. But damn, can people get hot under the collar quick over it. The first thing I notice, when "instincts" are brought into a conversation/debate/fight-to-the-death, is that the DEFINITION of "instinct" becomes of overwhelming importance. There is no doubt that the definition of a word, in this case, "instinct", is of absolute importance, because the very heart of the matter, is a fight over whether "it" exist or not. "It" being "an instinct", leading to the very real, but still ironic, situation, where we are fighting over something that is called "instinct", which is a word that means something, but what that something is, is defined by the people fighting over whether it exist or not. While some may resist the obvious call for "define your terms", before the conversation goes forth, it seems more than a semantic point in this case.

If you go with the view that the use of a word determines its meaning, and enough people agree on a meaning, then a word does mean what the group says it does, if you go with that view of things, (wikiality), then "instinct" means whatever somebody wants it to mean, which is just about the dumbest thing I can imagine happening.

A common understanding of terms is essential for scientific progress, (as well as business in general), and language is often a barrier to communication, rather than a great force to bring us together. This is not by chance. Specialized languages, within a common tongue, are used as much to prevent outsiders from understanding what is being said, as they are a sign of being in-the-know, an essential factor in many human enterprises, both business as well as military in nature.

I agree. Most debates eventually devolve into an issue over the meaning of a word or words. The reason, again, why I put the word 'instinct' in scare quotes is because I don't think we have a good definition that serves the purpose for human interactions. Part of the issue, I think, is that since we are always the result of an interaction between nature and nurture it is very difficult for us to piece out which contributes where. There are plenty of folks who see the word 'instinct' and immediately think -- absolutely programmed behavior with no variation. That isn't the way that William James used it precisely and he is the guy who suggested that humans have more instincts than other creatures, which is why we have more scope for action -- we have to choose among the various possible 'instinctual' options.

If you look at the duck imprinting issue, learning is absolutely necessary to the process. There is no clear divide between learning and 'instinct'. For language, I think one of the supreme ironies may be that the 'instinct' may result in the peculiar way that language is learned. Nothing in the system is absolutely hard-wired, but there does seem to be a certain underlying architecture. It's obvious that it must be very plastic. But there are aspects of language that allow us to pick it up easily (because of the way the system is wired), and it may be something simple like the expression of a receptor or two on a particular group of neurons that have a particular tendency for action (one that remains very plastic as is most of the human nervous system during early development). It becomes a bit difficult to stick to any strict definition of the word 'instinct' in this situation. I thought Piggy and I were fairly clear about how we used the term in that other thread, though. I, at least, meant it as a tendency in certain nervous architecture to facilitate the learning of language. I generally put it in those scare quotes because I realize this is not the typical way the word is used (like, say, for the way a cockroach acts in the presence of a change in light or movement), but it is used for this purpose by many, so I stuck with it just for ease of communication (while still trying to explain more fully what I meant).

We have the same problems in the nearly continuous mind-body debates because our language is hopelessly dualistic.

[Jeff Corey]

As an undergrad, I learned that instincts were 1. Species-specific 2. Complex behavior sequences 3. common to all members of the species (or of all members of one sex of that species) 4. that occurred without any learning involved 5. triggered by specific situations.
The fertilized female digger wasp finds a caterpillar, paralyzes it, lays her eggs in the caterpillar,digs a hole in the sand, and buries it.
The scientific term a half century later is "fixed action patterns".
Humans don't got any.

[Ichneumonwasp]

Still with me? Maybe I missed it, but it seems in the discussion about language, instincts, learning, etc etc, a salient point would be the learning/use of a language, in written form. Of course this might just be clouding the waters, but language seems to be limited to speech and cognition, while ignoring the more complex, and very important task, or learning a language, in the written form.

And I think that process highlights the differences in learning that Piggy was trying to bring into focus in the other thread. Learning to read and write is different from learning to speak or sign. Reading and writing require quite a bit more teaching input. I think this is because the system for learning to read and write is not self-reinforcing the way that learning a first language for speaking or signing is.

[Ichneumonwasp]

As an undergrad, I learned that instincts were 1. Species-specific 2. Complex behavior sequences 3. common to all members of the species (or of all members of one sex of that species) 4. that occurred without any learning involved 5. triggered by specific situations.
The fertilized female digger wasp finds a caterpillar, paralyzes it, lays her eggs in the caterpillar,digs a hole in the sand, and buries it.
The scientific term a half century later is "fixed action patterns".
Humans don't got any.

That's fine. But we also use the term 'instinct' colloquially to refer to the imprinting process. I don't have any problem with another term being used in its place as it relates to humans. As I have said over and over it is the internal wiring that makes the process of language acquisition easier that is the important point. I don't really care what word is used. We do need the proper terms to be able to communicate without ambiguity, I fully agree, but does the point come across? I'm afraid that we don't all possess the behaviorists' lingo.

I suppose we could dig up William James and yell at him a bit for introducing the idea.;)

[Jeff Corey]

If we are going to yell at anyone, it should be Jacques Theroux.

[Ichneumonwasp]

OK, that fine with me. Theroux was an ass.

I think the really exciting bit is that we are finally piecing out the contributions from genes and realizing that even speaking about a "gene for x behavior" is just stupid talk that reporters in the lay press get wrong. It's discussions like this that, I think, help the process along.

Do we have a good substitute word for 'instinct' in the way we have been using it or do we need to use a longer descriptor? I only used 'instinct' because that is what most of the lay writing about this stuff uses.

[Piggy]

As an undergrad, I learned that instincts were 1. Species-specific 2. Complex behavior sequences 3. common to all members of the species (or of all members of one sex of that species) 4. that occurred without any learning involved 5. triggered by specific situations.
Seems to me that all those conditions apply to the process driving preverbal phoneme discrimination in infants, for example.

But the kicker is #4. The phoneme discrimination process, of course, is an element of a larger process in which learning is undoubtedly occuring.

But what the infant's mind is doing in order to make that more complex learning process start up, continue, and eventually shut down -- that runs automatically. And it's the same in everybody.

Again, it comes back to there having to be 2 aspects in order to guarantee that all "normal" members of the species will learn a human language -- that is, the language of their community, which itself conforms to universal human grammar.

But that's just me.

[Ichneumonwasp]

And it's the same in everybody.




Slight correction -- it isn't the same in everybody because there are people who never learn language or who have significant problems with it. It depends on one's genetic heritage.

[Piggy]

Slight correction -- it isn't the same in everybody because there are people who never learn language or who have significant problems with it. It depends on one's genetic heritage.

True. That's why I added the term "normal" in quotes to the line below. But that's the same for any animal instinct, too -- a deformed spider doesn't spin the web right -- so I didn't think it was a concern. Seemed axiomatic.

[Piggy]

Slight correction -- it isn't the same in everybody because there are people who never learn language or who have significant problems with it. It depends on one's genetic heritage.

Oh, maybe I mistook your meaning. I don't mean that it's identical. But the process does the same thing in all of us. Back to the spider, the webs aren't exactly identical, but the programming has the same result for all members. You don't see them spinning just any old web.

Anyway, I think we're in agreement, as usual.

[Ichneumonwasp]

Yep

[Jeff Corey]

Seems to me that all those conditions apply to the process driving preverbal phoneme discrimination in infants, for example.

But the kicker is #4. The phoneme discrimination process, of course, is an element of a larger process in which learning is undoubtedly occuring.

But what the infant's mind is doing in order to make that more complex learning process start up, continue, and eventually shut down -- that runs automatically. And it's the same in everybody.

Again, it comes back to there having to be 2 aspects in order to guarantee that all "normal" members of the species will learn a human language -- that is, the language of their community, which itself conforms to universal human grammar.

But that's just me.
You just reminded me of another criterion for "instinctive" behavior. The sequence is not able to be modified by learning.

[Piggy]

You just reminded me of another criterion for "instinctive" behavior. The sequence is not able to be modified by learning.
Good point. So that's where the borderline would lie between those parts of the process that we can slap the ole "instinct" label on and those we can't.

In the case of phoneme discrimination, the process of learning to distinguish the phonemes -- the "program that's running" to make it happen -- in preverbal infants is not modified by learning. It does its thing til the auto shutdown sequence stops it.

On the other hand, the infant's capacity to distinguish phonemes is indeed modified over time, and each individual's resulting ability in this regard is slightly different, depending on the environment. (But as has been noted, the ways in which this ability is allowed to be different are not determined by environmental reinforcement alone, but are also constrained by the working of the automatic device itself. We know this because the observed variances in fundamental and purely linguistic aspects of language do not encompass the entire range of possibilities, but are limited to a set range of options which are nowhere implicit or explicit in the environment.)

But the learning part, the modification, would not occur if the instinctive part -- the universal program that's just doing its thing -- weren't operant, if it didn't bootstrap itself, run the same way in all of us, and shut itself down.

And this little program isn't built for general smarts. It's built to do exactly this, with the specific goal of language acquisition.

So again -- as I and Ichneumonwasp and Pinker also are saying -- the entire process of "learning to talk" appears to require some core components that can aptly be called "instinctive" and which are specific to the task, others that are more generalized, and interaction with the environment which is not instinctive.

One of the reasons this particular example may not be, at first, intuitive is that, unlike the spider spinning its web or the dirt dauber mudding up its nest, we don't see what's going on. After all, this is happening inside a preverbal infant's mind. Once we make the effort to measure what's going on -- by using differential attention studies, for example -- it becomes clear that something extraordinary is going on inside that moony little head. Once the baby begins to babble and produce its own proto-linguistic sounds, a heck of a lot of work has already been done.

[Jeff Corey]

Good point. So that's where the borderline would lie between those parts of the process that we can slap the ole "instinct" label on and those we can't.

In the case of phoneme discrimination, the process of learning to distinguish the phonemes -- the "program that's running" to make it happen -- in preverbal infants is not modified by learning. It does its thing til the auto shutdown sequence stops it...
And they said that Skinner treated people as if they were machines. Of course they are, but I was feeling a bit risible, like the frotteur rubbing the comedian.
To your point, do you have any data that phoneme discrimination is not modified by learning? The well known phenomenon of babbling drift shows that infants gradually shift their production of phonemes to match their verbal environment with the first year. How do you measure their actual discrimination?

[Piggy]

To your point, do you have any data that phoneme discrimination is not modified by learning? The well known phenomenon of babbling drift shows that infants gradually shift their production of phonemes to match their verbal environment with the first year. How do you measure their actual discrimination?

Well, I think this actually speaks to the difference among the various aspects of the process. As has been noted, the current linguistic consensus is that there's an interplay between the "automatic" components, the plastic components, and environment -- the resulting drift/learning results from their interaction.

So it's actually not correct -- as I'm sure you already agree -- to say that "phoneme discrimination is not modified by learning". If it were not, it's difficult to imagine how it could happen at all, especially given the fact that babies learn the language that's being spoken around them; they don't all simply start speaking some universal human language.

The process as a whole certainly is modified over time, and depending on conditions. So it's crucial to keep in mind that the standard model does not claim that the entire process is what could be called "instinctive". Rather, the research indicates that there are "automatic" or "hard-wired" or "instinctive" components involved, and that these are necessary and specific to language acquisition rather than general-purpose.

Some cites, which will answer the "how do you measure" question better than I can, and give some context. Not all these are fully available by link, but you'll have the citations anyway if you want to find hardcopy at the library.

Journal of Cognitive Neuroscience: Common Neural Basis for Phoneme Processing in Infants and Adults (http://jocn.mitpress.org/cgi/content/abstract/16/8/1375)

In this article, we review event-related potential (ERP) results obtained in infants during phoneme discrimination tasks and compare them to results from the adult literature. The striking similarities observed both in behavior and ERPs between initial and mature stages suggest a continuity in processing and neural structure. We argue that infants have access at the beginning of life to phonemic representations, which are modified without training or implicit instruction, but by the statistical distributions of speech input in order to converge to the native phonemic categories.

Another abstract (http://science-mag.aaas.org/cgi/content/abstract/298/5600/2013), more concerned with methodology:

Human infants begin to acquire their native language in the first months of life. To determine which brain regions support language processing at this young age, we measured with functional magnetic resonance imaging the brain activity evoked by normal and reversed speech in awake and sleeping 3-month-old infants. Left-lateralized brain regions similar to those of adults, including the superior temporal and angular gyri, were already active in infants. Additional activation in right prefrontal cortex was seen only in awake infants processing normal speech. Thus, precursors of adult cortical language areas are already active in infants, well before the onset of speech production.

Here's a short paper (https://intranet.psychology.bangor.ac.uk/dspace/bitstream/123456789/8708/2/) (PDF) heavy on the details. From the conclusion:

Insofar as ERP differences are confined to MMN like modulations, it may be assumed that the engagement of attention by familiar words in 11-month-olds is an automatic process which starts very early in the course of auditory processing, i.e. within 250 ms.

And, fwiw, Pinker's introduction to "Language Acquisition" (http://users.ecs.soton.ac.uk/~harnad/Papers/Py104/pinker.langacq.html), which can be searched and provides header-marked sections in a wider context, along with references.

[Jeff Corey]

I started to read Pinker, but I was stopped in my metaphorical tracks when he described Comsky's review of Verbal Behavior and characterized it as trying to use "a few simple laws of learning" to explain the acquisition of language. A two semester graduate course is required to partially master "a few simple laws of learning".
Apparently neither Pinker nor Chomsky ever read the book. But I'll give it a shot.

[Piggy]

I started to read Pinker, but I was stopped in my metaphorical tracks when he described Comsky's review of Verbal Behavior and characterized it as trying to use "a few simple laws of learning" to explain the acquisition of language. A two semester graduate course is required to partially master "a few simple laws of learning".
Apparently neither Pinker nor Chomsky ever read the book. But I'll give it a shot.
:D

Thanks for giving it a shot. You might want to skip around to the relevant subsections in Pinker. As robinson pointed out earlier, there's plenty to criticize in Pinker.*

Of course, if the topic here is Pinker v. behaviorism, I'm sure you'll find plenty of fodder there.

For the more narrow consideration of what exactly is going on in the process of "learning to talk", and whether or where there's a boundary point where we can meaningfully stick an "instinct" label on any area of the model that describes what's happening during that process, it's probably best to focus on the more detailed stuff being done in the field and lab rather than his gloss.


* The reason I usually found myself yelling at the page when reading Pinker was his (to my mind) pointlessly dogged adherence to a specific model of evolutionary biology as the tool of choice for whatever came across the bench. This is especially annoying when his approach, while able to conjure up a quite reasonable explanation for why a phenomenon may have come to be, did not appear sufficient to propose any means of falsifying that explanation or searching for others. It also seemed, at times, to give short shrift to alternative tools of modeling, such as systems theory and AI.

[Jeff Corey]

More of what Steven Jay Gould called "Just so stories" (with reference to Wilson's Sociobiology)?

[Piggy]

More of what Steven Jay Gould called "Just so stories" (with reference to Wilson's Sociobiology)?

You mean my objection to some of Pinker's stuff? Yeah, there's some just-so to it. Not that I object to people coming up with those kinds of explanations, because they can lead to productive avenues of research. But when we do arrive at those kinds of hypotheses, it's crucial never to forget what they are.

Some posts on another thread, in fact, were reminding me of one such evo-bio explanation of the universality of gossip in human culture (a topic, btw, that plays a central role in my daily work) which turns out to be related to the language issue.

The hypothesis is that the kind of reciprocity tracking we see in, for example, chimpanzees -- which can be "gamed" by misrepresentation -- was made even more difficult when language evolved, because people could simply speak words and lie without having to go through machinations to game the system.

As a result, humans developed a keen interest in gossip, in order to compare people's words with news they gathered about their actions, and so we are naturally attracted to it. And this -- among other things -- accounts for why the life of Paris Hilton, for example, is followed by millions of people who have never met her. And, perhaps, it accounts for the evolution of the blush response etc etc.

Well, ok. Interesting idea. True? Maybe.

But in Pinker's writing I felt he was too quick to accept that sort of evo-bio explanation of some phenomena and move to the conclusion that, since we now have an evo-bio model for it, there's not a whole lot of value in applying other theoretical lenses to the issue. Kindof a FIDO attitude.

[Jeff Corey]

"FIDO"? Unfamiliar acronym to me. Is there a rule #8 there?

[Piggy]

"FIDO"? Unfamiliar acronym to me. Is there a rule #8 there?
"F*** it, drive on" = nevermind what just happened or what's going on around you, just keep going

[Jeff Corey]

Reminds me of Pete Seeger- "Waist deep in the Big Muddy (but the big fool said to push on)"
Surprized it doesn't any air play nowadays.

[blutoski]

You just reminded me of another criterion for "instinctive" behavior. The sequence is not able to be modified by learning.


Two questions about this, though:

1. Is there a distinction between instinct versus reflex actions? Sword swallowers learn how to modify their gag reflex.

2. Animals that are capable of learning seem to be able to suppress certain instincts. eg: domesticated Canus and Felinus spp learn to suppress their instinct to mark all of their territory. They learn to limit it to the areas approved of by the human master. A friend of mine works at the Vancouver Aquarium studying octopi, and she indicated that instinct-suppression is evidence of learning.

Much to Roy Horn's dismay.

[Jeff Corey]

Two questions about this, though:

1. Is there a distinction between instinct versus reflex actions? Sword swallowers learn how to modify their gag reflex.

2. Animals that are capable of learning seem to be able to suppress certain instincts. eg: domesticated Canus and Felinus spp learn to suppress their instinct to mark all of their territory. They learn to limit it to the areas approved of by the human master. A friend of mine works at the Vancouver Aquarium studying octopi, and she indicated that instinct-suppression is evidence of learning.

Much to Roy Horn's dismay.
1. Reflexes are simple S->R relationships. Puff of air to cornea->eye blink. Fixed action patterns (might as well use the technical term) are more complex and involve the whole organism. Nest building.
2. I don't these are fixed action patterns.

[Ichneumonwasp]

Do any mammals possess any instincts, so defined?

[Piggy]

I don't [think] these are fixed action patterns.
By that definition, if I understand it correctly, I'd say the challenge is to find out what exactly is going on and then determine whether, or to what extent, anything like that is happening.

Certainly, on the external/macro level, we're not dealing with fixed action patterns.

But because of the features of language acquisition that were noted on the other thread (e.g., boostrapping, grammar wizard, raindance effect, automatic shutdown sequence) there's reason to believe that fixed action patterns are part of the internal operation of the brain which gives rise to the macro-level learning phenomenon we actually observe.

It may be the case that these patterns don't require a whole lot of processing power -- which would be a good thing -- but that, like kernel operations, the programming would not be able to function without them.

At this point, it's safe to say that no one knows enough about how the brain works to describe exactly what's going on. But going back to the coffee pot analogy, when two machines make coffee, but only one of them can turn itself off and on, it's clear there's a device operating in one that is not operating in the other. And since some of the key features of language acquisition appear to be specific to that process, and not to have analogs in processes such as learning to use a spoon, it seems to me meaningful to posit an "instinctive" component somewhere in the core, and continue research to discover the nuts and bolts of it all.

[blutoski]

Do any mammals possess any instincts, so defined?

That's why I was asking about reflexes. There are more complex 'reflexes' that involve the whole organism. A specific example is something called "the simian response." This is shared by simians, including humans. When you drop a neonate on his back, he flails his arms and vocalizes.

In terms of nest building, this is clearly a more complex behavior. Many monkey species build nests even when raised in isolation (they don't learn to build nests), but they can also learn to build better nests if they're part of a social group with such technology. There were experiments in the 1960s with more advanced genera such as Pan which showed that some basic behaviors such as nest building are not learned, but that they style of nest building can be learned.

Another likely instinct in Pan is the obsessive power-jockeying that comes with physical maturity. Chimps raised in human households become unmanageable because their instinct is to challenge and test the hierarchy to exhaustion. They don't grow up with the same socializing instincts that human children do, and become much more independent, dangerous, and violent as adults. Human neoteny furnishes us with a more childlike disposition in our postadolescence.

Nevertheless, the techniques used to challenge authority are specific to the community. For example, members of Goodall's tribes developed and passed on the technique of rolling around empty oil drums as a way to terrorize challengers. This skill was passed from individual to individual. But the underlying urge to threaten is certainly instinctive, since - as mentioned above - it is demonstrated by individuals raised in isolation from other Chimps.



Conversely, we can identify things that are not instincts by observing they are not universal, or that societies 'lose' them. Fire is such an example: the Tasmanian aborigines 'lost' the technology after they migrated from the main continent. A pretty good argument that this is not an instinctive skill. They did not, however, lose language, and I'm pretty sure there is no example of a human community that has lost this skill. Inductively, language 'basics' does look like a candidate for instinct.

[Jeff Corey]

Do any mammals possess any instincts, so defined?
Not that I know of. I googled a bit and one source said fixed action patterns are
'extremely rare" in mammals, but curiously, didn't describe any.
Wikipedia starts off with a good definition and then screws up by saying people have them. The examples they give don't fit.
I'll check with my friend in Bio, who teaches ethology, tomorrow.

[blutoski]

I'm still having trouble distinguishing between reflex and instinct.

Is suckling not a classic mammalian instinct?

Mating?

Protection of infants? (and recognition of such by their 'cute' appearance?)

I'm confused.

[Piggy]

I'm still having trouble distinguishing between reflex and instinct.

Is suckling not a classic mammalian instinct?

Mating?

Protection of infants? (and recognition of such by their 'cute' appearance?)

I'm confused.
I'm leaving for vacation in the morning, so this may be my last post here, but I'd like to chime in on this one, just to offer a reminder, that when we ask "What qualifies as instinct?" we're no longer asking a question about what actually happens in the world, but rather about when it's appropriate to use a term.

Fwiw.

[Jeff Corey]

Bluto,
Sucking in infancy is a reflex. It turns into a voluntary behavior later. So do the palmar (grasping) and Babinski (toes curling when the sole of the foot is stimulated, that goes at about 6 months and is only seen in adults with certain brain traumas) reflexes.
As to the others, the behaviors are too variable and have too much a strong voluntary component to be called fixed action patterns, which is the term that has replaced "instinct".
The reason that "instinct" isn't used anymore is that, at one time, is was used to explain everything. Life instinct, death instinct, maternal instinct, social instinct, antisocial instinct. It's almost as if those scientists at that time had an instinct instinct.
Now scientists are more likely to talk about "drives" that motivate goal-directed behavior. The mating instinct is gone and the sex drive is in. And much is known about the areas of the brain that are key to expressing those drive.
Take my hypothalamus, please.

[robinson]

... I'd like to chime in on this one, just to offer a reminder, that when we ask "What qualifies as instinct?" we're no longer asking a question about what actually happens in the world, but rather about when it's appropriate to use a term.


hmm... I seem to remember something about this very situation. Oh yes, I think it was -

The issue of "human instincts" is probably one of the most interesting topics. But damn, can people get hot under the collar quick over it. The first thing I notice, when "instincts" are brought into a conversation/debate/fight-to-the-death, is that the DEFINITION of "instinct" becomes of overwhelming importance. There is no doubt that the definition of a word, in this case, "instinct", is of absolute importance, because the very heart of the matter, is a fight over whether "it" exist or not. "It" being "an instinct", leading to the very real, but still ironic, situation, where we are fighting over something that is called "instinct", which is a word that means something, but what that something is, is defined by the people fighting over whether it exist or not. While some may resist the obvious call for "define your terms", before the conversation goes forth, it seems more than a semantic point in this case.

If you go with the view that the use of a word determines its meaning, and enough people agree on a meaning, then a word does mean what the group says it does, if you go with that view of things, (wikiality), then "instinct" means whatever somebody wants it to mean, which is just about the dumbest thing I can imagine happening.

:hb:

[Piggy]

hmm...

Damn, robinson, don't hurt yourself there, buddy.

You keep doing that every time somebody offers a reminder, you're gonna give yourself brain damage.

Oh, wait....

[robinson]

Have a nice vacation.

[robinson]

It really is a most interesting subject. It combines semantics, understanding of meaning, communication, cultural and religious beliefs, scientific uncertainty, all combined with trying to discuss the issue, using the very tools, (words, phrases, definitions, language if you will), that are being debated.

Try to think of it as dancing, rather than fighting. That way, every now and then, you can try and cop a feel.;)

[Jeff Corey]

Or feel a cop.

[blutoski]

I'm leaving for vacation in the morning, so this may be my last post here, but I'd like to chime in on this one, just to offer a reminder, that when we ask "What qualifies as instinct?" we're no longer asking a question about what actually happens in the world, but rather about when it's appropriate to use a term.

Fwiw.

Ah. I misunderstood. I thought it was a scientific discussion. You're saying it's a definitional debate.

OK - I'll duck out. I'm a biologist, so have nothing to contribute here.

[drkitten]

DrK claims that studies show that language arises even in the absence of reinforcement, but also demonstrates, proudly, that his conception of reinforcement is decades out of date.

I stand my my writing.

If you wish to refute me, I will give you a simple scenario. Define "reinforcement" and explain how it it applies to the following situation without invoking mental states.

I tell a number of students that Harry is six foot three inches tall, while John is two hundred five
centimeters tall. I tell them that a meter is 39.37 inches long. I then tell them that if Harry is taller than John, they should draw a star on a piece of paper and hand it in (for credit). If Harry is shorter than John, they should draw a triangle and hand it in. They have never before recieved this problem and have no prior knowledge of the solution.

How does "reinforcement" explain the observation that the overwhelming majority of papers I receive have the correct symbol? How does it explain that not all of the papers I receive have the correct symbol?

I am perfectly happy rejecting any definition of "behaviorism" that denies the existence of mental states with propositional content without explaining how propositional content can emerge from behavior. Until that flaw is addressed, there's no need for my reading to be "up to date."

Behaviorism is astrology, with less artistic merit.

[Jeff Corey]

"for credit" might be a clue.

[drkitten]

"for credit" might be a clue.

No more so than "for credit" would permit them to burn a candle in a vacuum. Desire not to fail cannot overcome a physical impossibility.

And solving that problem without mental states with propositional content is physically impossible.

[Jeff Corey]

You said they solved it, so apparently not.

[drkitten]

You said they solved it, so apparently not.

Yes. It's impossible to solve in the absence of mental states with propositional content.

They solved it. Therefore, mental states with propositional content exist.

Behaviorism denies the existence of mental states (with or without propositional content).

Therefore, behaviorism is incorrect.

If you have an actual answer to this sorite, I invite you to post it. Given that no one since Skinner has ever been able to address this sorite, you'll be the first.

I stand by my writing. Behaviorism is astrology, without the artistic merit.

[Jeff Corey]

Try this http://skeptically.org/skinner/id9.html

[drkitten]

Try this http://skeptically.org/skinner/id9.html

Tried it. "Patent nonsense" would be putting it kindly.

Stored facts -- such as "John is 205cm tall" are not behavior.

"Rule-based" behavior cannot describe purely abstract behavior like solving math puzzles; rules can only operate in an environment where they have facts and axioms upon which to operate. In the case of a pure abstraction, the "fact" that "John is 205cm tall" must be stored somewhere. Since it's not physical storage, it can only be stored as a mental state.

.... which behaviorism denies.

I had a longer response, but the forum ate it. Basically -- behaviorism is still astrology, but without the artistic merit or the factual accuracy of its premises.

[robinson]

What is Behaviorism?

http://en.wikipedia.org/wiki/Behaviorism

Is that an accurate description?

[Jeff Corey]

The part dealing with Skinner's radical behaviorism (behavior analysis) looks OK, after a quick skim.

[robinson]

I'm thinking that if we can't agree on terms, there is no chance of a scientific discussion. This seems to be a huge problem in dealing with human, as well as animal studies. There isn't a standard for a lot of things.

Words themselves are in contention, or the meaning is, and while this can be dismissed as "semantics", it doesn't change the problem a bit.

[Jeff Corey]

I just checked with my friend in Bio. He confirms that they encourage their students avoid using vague terms like instinct in favor of more specific terms such as reflexes and fixed action patterns (FAPs).
As to FAPs occurring in mammals, he could not think of any. He is interested enough to contact his mentor's mentor, who studied with Conrad Lorenz.

[robinson]

No FAPs occurring in mammals? wtf? What are the behaviors called then?

[Jeff Corey]

What specific behaviors? They could be drive related or reflexes.

[drkitten]

As to FAPs occurring in mammals, he could not think of any.

This sounds implausible to me, unless you're using FAP in a term-of-art sort of way that excludes a lot of what laymen would call "instinctive."

For example, (intact) male cats mark their territories by spraying urine, even if they're indoor cats that have never met another intact adult male cat in their lives. If you're suggesting that this is learned behavior -- from whom did they learn it? If you're suggesting that this is unlearned behavior but not "instinctive," then I suggest that your definition of "instinctive" is too narrow.

I believe that raccoons raised in isolation still "wash" their food, as another example (although I'm less confident of this).

For that matter, wouldn't the "return to 'mother' when you hear a startling loud noise" (Harry Harlow's surrogates) count as "instinctive"? If not, from whom did the monkeys learn it?

[robinson]

FAPs are specific, innate behavioral sequences initiated by specific stimuli that are not a result of gene-environment interactions (Alcock, 1997). These “hard-wired” behaviors are typically under genetic control, may be species specific, go to completion upon initiation, are not regulated through feedback loops, and are not reflexes but complex, coordinated behaviors (Weber and Spieler, 1994).

Ahh... so "instinct" has been replaced with FAP. That clears everything up.

[robinson]

http://en.wikipedia.org/wiki/Fixed_action_pattern

http://en.wikipedia.org/wiki/Instinct

http://en.wikipedia.org/wiki/Language_acquisition

Fascinating. The stuff you learn from these Internets.

http://bio.research.ucsc.edu/~barrylab/classes/animal_behavior/LEARNING.HTM

There is a lot of stuff out there about these things, whatever you call them.

[Jeff Corey]

Ahh... so "instinct" has been replaced with FAP. That clears everything up.
Also see post # 32.

[Jeff Corey]

Tried it. "Patent nonsense" would be putting it kindly.

Stored facts -- such as "John is 205cm tall" are not behavior.

"Rule-based" behavior cannot describe purely abstract behavior like solving math puzzles; rules can only operate in an environment where they have facts and axioms upon which to operate. In the case of a pure abstraction, the "fact" that "John is 205cm tall" must be stored somewhere. Since it's not physical storage, it can only be stored as a mental state.

.... which behaviorism denies.

I had a longer response, but the forum ate it. Basically -- behaviorism is still astrology, but without the artistic merit or the factual accuracy of its premises.
You didn't read the article. I am sure of that.
Cognitive science is as much an oxymoron as creation science or christian science. You all believe in imaginary entities. There is no such thing as mental. There is no ghost in the machine.