Wednesday, October 31, 2007

The Cognitive Revolution & Objectivism, Part 1

Starting in the 1950s, the sciences of mind, brain, genes, and evolution began forming a new understanding of human nature, based on an ever growing body of empirical data. This "cognitive revolution" presents a formidable challenge to the largely speculative model of the mind implicit in Objectivism.

Here's Steven Pinker's description of the new model of the brain emerging from the cognitive revolution:
Before the revolution, commentators invoked enormous black boxes such as "the intellect" or "the understanding," and they made sweeping pronouncements about human nature, such as that we are essentially noble or essentially nasty. But we now know that the mind is not a homogeneous orb invested with unitary powers or across-the-board traits. The mind is modular, with many parts cooperating to generate a train of thought or an organized action. It has distinct information-processing systems for filtering out distractions, learning skills, controlling the body, remembering facts, holding information temporarily, and storing and executing rules. Cutting across these data-processing systems are mental faculties (sometimes called multiple intelligences) dedicated to different kinds of content, such as language, number, space, tools, and living things.... Still another layer of information-processing systems can be found in the affect programs, that is, the systems for motivation and emotion.

The upshot is that an urge or habit coming out of one module can be translated into behavior in different ways—or suppressed altogether—by some other module.... More generally, the interplay of mental systems can explain how people can entertain revenge fantasies that they never act on, or can commit adultery only in their hearts. In this way the theory of human nature coming out of the cognitive revolution has more in common with the the Judeo-Christian theory of human nature ... than with behaviorism, social constructivism, and other versions of the Blank Slate. Behavior is not just emitted or elicited, nor does it come directly out of culture or society. It comes from an internal struggle among mental modules with differing agendas and goals.

This modular theory of the mind, when combined with behavioral genetics, challenges Rand's tabula rasa view of the mind, her denial that free will is "saddled" with tendencies, and her ideal of a consciousness "in perfect harmony."

17 comments:

Jay said...

Isn't it possible that by studying things like evolutionary psychology, people can learn how to better harmonize their consciousness?

Just a thought.

David said...

What exactly is meant by "harmonizing consciousness"?

Anonymous said...

1. Either you or Pinker are wrong (I'm not sure) in identifying the cognitive revolution with the modularity theory of mind. The cognitive revolution is simply a the general trend in psychology since the mid 20th century of retreating from the behaviorism of Watson, Skinner, et al:

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

Seen from this light, Ayn Rand's criticisms of behaviorism were ahead of their time. I don't agree with all of the details of the following article, but it outlines some of the reasons for this:

http://hubcap.clemson.edu/~campber/randcogrev.html

Yes, there are many details of contemporary cognitive science that Objectivists would dispute. But then again there are many such that many psychologists would dispute. There is little if any consensus about whether any cognitive sciences have shed light on human nature.

2. Now as for the modularity theory of mind (which is just one of many competing views in contemporary cognitive science), here I think that either its pronouncements are not as well-established as you suggest, or they are and they do not contradict Objectivism.

Pinker says that traditional psychology "invoked enormous black boxes" to do its explanatory work. True, but modularity theory's response is basically just to posit quite literally lots of little black boxes. You'll be surprised by the methodology many of them employ. See, for example, Stich and Nichols' modular theory about our mindreading abilities. We have this ability, so they hypothesize about what kinds of subsystems *might* add up to create this bigger system. They create lots of charts with little "boxological" diagrams to record this speculation. Yes evidence is cited along the way, but it is extremely speculative and numerous alternative explanations are possible (and not considered). It is not as if modularity theorists are always opening up our skulls and saying, hey look, there's the Tomato-Recognition Module!

That said, even if modularity theory turns out (luckily, in my opinion) to be correct about the existence of physically distinct modules responsible for elements of our cognition, I don't see any reason why this would contradict AR's "tabula rasa" view. Any modularity theorists who argue that the existence of innate modules implies the existence of innate knowledge are confusing the potential with the actual--a mistake that has been made by nativists since Plato. Just because there is a ready-made system responsible for acquiring some cognitive content does not mean that we possess that content from birth. As for the question of "perfect harmony," I'm not sure where you're getting that idea from, but notice that even Pinker speaks of the allegedly distinct modules "cooperating to generate a train of thought or an organized action."

Now I suspect that you're correct that there will be some conflict between cognitive science's tacit determinism and AR's views of free will, but then again, this determinism doesn't arise because of any distinctive discoveries of cognitive science. It's just a philosophical presupposition that the cognitivists inherit from the behaviorists, who got it from 19th century philosophers. There is a conflict here, but it has to be settled by philosophy, not science.

Daniel Barnes said...

Anon writes:
>...The cognitive revolution is simply a the general trend in psychology since the mid 20th century of retreating from the behaviorism of Watson, Skinner, et al...Seen from this light, Ayn Rand's criticisms of behaviorism were ahead of their time...

Well, Nyquist cited Pinker:

"In this way the theory of human nature coming out of the cognitive revolution has more in common with the the Judeo-Christian theory of human nature ... than with behaviorism, social constructivism, and other versions of the Blank Slate."

Clearly The Bible was ahead of its time too...;-)

The point is not "Ayn Rand criticised behaviourism" - lots of thinkers did (eg:Karl Popper). What we're interested here is 1) were her distinctive criticisms any good? 2) were her competing theories any better established and 3) how coherently 1) and 2) sit in the context of the rest of her philosophy?

IMHO Rand's writing is, like the Bible, vague and contradictory enough to reverse engineer a rather large load of miscellaneous theories into should one be suitably enthusiastic. For example, a footnote in the Robert Campbell essay you cite mentions Barry Vacker, who argues that "Rand's visual esthetics...anticipated fractal geometry, chaos theory, and other conceptions of dynamic systems...particularly her descriptions of Howard Roark's buildings...."

See what I mean? Now she's "ahead of her time" in advanced mathematics too.

>That said, even if modularity theory turns out (luckily, in my opinion) to be correct about the existence of physically distinct modules responsible for elements of our cognition, I don't see any reason why this would contradict AR's "tabula rasa" view.

This question turns on how "rasa" was her "tabula". This, once again is very vague, and highly inconsistent within her philosophy. For example, in the intro in "We The Living" she claims that no-one is born with any kind of "talent", that all are self-created. Remarkably she even argues - on the empirical basis of absolutely nothing - that something as primal as integrating sensations into percepts is not innate, but acquired!* So we have about as radical tabula rasa position as you get by now. But then just when you think she's taken a firm stand, elsewhere she'll write something completely different. For example, the ITOE:

"A percept is a group of sensations automatically retained and integrated by the brain of a living organism. [....] Percepts, not sensations, are the given, the self-evident..."

So now they're "automatically" integrated by the brain, not acquired? For example, another way tabula rasa is often asserted by Objectivists is as a kind of "blank" mechanism, that contains no information. A blank slate, they argue, is not no slate.

But this seems to be mere verbalism. For to assert a mechanism is to inevitably assert a tendency. A water wheel does nothing absent a flow of water. However, when in the presence of water it will exhibit a tendency to turn. To assert a mechanism without a tendency is like a grin without a cat, so to speak. Rand herself even asserts a tendency to be rational in man, asserting that it is part of his survival equipment, like the tooth and claw of animals. But if we can have one inborn tendency, why can't we have another, or many others?

As Nyquist has often remarked, one of the problems of critiquing Rand's thought arises from the complexity of its confusions. Certainly things like the cognitive revolution do not necessarily contradict Objectivism, if we are prepared to make Objectivism suitably vague in the first place.

* "...to perceive the things around him by integrating his sensations into percepts (which is not an innate, but an acquired skill)..."("The Comprachicos")

Anonymous said...

You wrote:

"The point is not "Ayn Rand criticised behaviourism" - lots of thinkers did (eg:Karl Popper). What we're interested here is 1) were her distinctive criticisms any good? 2) were her competing theories any better established and 3) how coherently 1) and 2) sit in the context of the rest of her philosophy?"

A fascinating attempt at changing the subject. The *point* of the original post was to suggest that cognitive science has made new and profound discoveries that contradict Objectivism. I gave reason to think that wasn't true. You respond by saying "the point is that AR's theory might not be better than behaviorism in the right way, and might also be contradictory." No, that's a new topic!

Now as to the question of tabula rasa. You have correctly identified a tension in AR's statements about sensations. You'll find tension like this in the work of any thinker of any complexity. (If that weren't true, there wouldn't be cottage industries devoted to interpreting Plato, Aristotle, Kant, etc.) The question is, what's the significance of it?

First, suppose AR is contradicting herself here. What does this have to do with the issue of tabula rasa? The *ability* to integrate something into something else is an ability. If we do have such an innate ability, this doesn't mean we have any innate knowledge. Knowledge is the product of exercising an ability, not the ability itself. It is an actual informational state, a mental grasp of reality.

You might disagree because of what you write here:

"For to assert a mechanism is to inevitably assert a tendency. A water wheel does nothing absent a flow of water. However, when in the presence of water it will exhibit a tendency to turn."

But I have no clue about why this addresses the issue. Yes, "to assert a mechanism is to assert a tendency." So? Not every tendency is an actuality. Just because our minds have the tendency to know from birth does not mean we are born knowing. Just because a water wheel has the tendency to turn in the presence of water doesn't mean that it necessarily has water in it. Isn't that obvious?

Now as it happens, plenty of other Objectivists have noticed the tension you note between the two statements about sensations. Most have decided that she made a simple and pretty innocuous mistake, one that doesn't have much significance for her overall philosophy. They tend to agree that if there are such things as sensations, they are integrated automatically by the brain (which is, after all, her position in a treatise that is explicitly concerned with epistemology).

But there are even many among them who go further, and argue that AR may have been mistaken about whether or not there are such things as sensations to integrate. They invoke the work of J.J. Gibson who argued that perception is direct, and that there are never any point sensations in need of integrating (a view that is more of a relic of bad Humean empiricism). And it's of little consequence for AR's view, because she herself admitted in the ITOE appendix that the existence of sensations was a scientific not a philosophic hypothesis (though it was a popular one for centuries).

I myself think that there is another interpretation of her view about the integration of sensations that hasn't been considered. When she spoke of sensations, she may not have meant anything like point sensations, but sensations understood more globally: visual vs. tactile vs. auditory sensations. A percept might be understood as the integration of trans-modal sensations, particularly as an integration of the so-called "common sensibles" (qualities that can be apprehended by more than one modality).

But maybe she didn't mean it. In any case, none of this really has anything to do with the question of whether or not the cognitive revolution has dealt a blow to Objectivism.

Daniel Barnes said...

Anon:
>A fascinating attempt at changing the subject.

I was responding to your 1., didn't really mean to divert.

>You'll find tension like this in the work of any thinker of any complexity.

Of course. The point is trying to get Rand's actual position clear. This and other "tensions" in her work suggest that her position on a number of issues often wasn't clear even to her. This is the first level of significance for such clashes.

>Not every tendency is an actuality.

I do not see how this touches mine or Greg's point. Did Rand assert that "free will" was saddled with inborn psycho/biological "tendencies" or not? If she says that it has none whatsoever, then then we have a bold psychological theory of compelling interest. If she is merely claiming that it's sorta-kinda free to some unspecificed greater or lesser extent, it becomes increasingly less so.
That is, as I wrote,"..the cognitive revolution (does) not necessarily contradict Objectivism, if we are prepared to make Objectivism suitably vague in the first place."

>In any case, none of this really has anything to do with the question of whether or not the cognitive revolution has dealt a blow to Objectivism.

See my above.

BTW: Interesting and engaged replies, Anon, for which I thank you.

Anonymous said...

You wrote in response to my point that not every tendency is an actuality:

"I do not see how this touches mine or Greg's point. Did Rand assert that "free will" was saddled with inborn psycho/biological "tendencies" or not? If she says that it has none whatsoever, then then we have a bold psychological theory of compelling interest. If she is merely claiming that it's sorta-kinda free to some unspecificed greater or lesser extent, it becomes increasingly less so."

But this is changing the subject again! You originally brought up the issue of tendencies and water wheels, etc., in order to make a point about tabula rasa (presumably, to dispute it). Now you're saying the point is about free will--a distinct issue.

But if you're going to raise that new issue, fine. If it is true that we automatically integrate percepts into concepts, this does very little to affect AR's view of free will. She almost always describes the locus of volition as the choice to think or not to think, which is the choice to engage our *conceptual*, not our perceptual faculty. There's no problem for the freedom of one aspect of the mind if a distinct aspect is determined. Just like: there's a very big difference between the control we have over the lifting of our hand, vs. the beating of our heart.

Daniel Barnes said...

Anon:
>But this is changing the subject again! You originally brought up the issue of tendencies and water wheels, etc., in order to make a point about tabula rasa (presumably, to dispute it). Now you're saying the point is about free will--a distinct issue.

Firstly, this is not changing the subject. See the last para of Greg's post. Obviously the freedom of human choice is the very thing in question, as the power of inborn psychological tendencies stands as the second line of suggested restriction on such freedom (the first line and most powerful being the laws of physics, the third "learned" or cultural restrictions which are not really "inborn" but, speculatively, inculcated at an early age and difficult to escape).

But in your response you seemed to have missed my question: Did Rand assert that "free will" was saddled with inborn psycho/biological "tendencies" or not? I would say not, by definition, but then it is not very clear what that definition does and does not cover. It is extremely vague. As far as "the choice to think" goes, it is difficult to see how Objectivists could both state that "there is no 'why'" causitive to that initial choice, yet also insist on such things as the principle of causality - for of course, such a principle would assert at least a predispositon to think.(Incidentally, I recall reading somewhere that Peikoff is walking away from that position on one of his audio lectures) As to whether this is a philosophical error supposedly confusing 'potential' with 'actual' seems scarely relevant - it either is a predisposition or it isn't.

Pugsley said...

Anon wrote,
"She almost always describes the locus of volition as the choice to think or not to think, which is the choice to engage our *conceptual*, not our perceptual faculty."

Does an Objectivist who is not already thinking about the choice to think or not "choose" to think or not? It would seem that the "free will" choice to think or not could only be made by someone already thinking, and thinking specifically about the choice to think or not, and in such situations, the only "choice" would be to stop thinking, since it would be the only option of doing something that wasn't already being done automatically.

Anonymous said...

Now things are really getting confused. Yes, Greg's post raised issues about both tabula rasa and free will, but these are distinct issues.

The potential/actual distinction responds to the concern that we might have innate knowledge, not to the concern that we may have no free will. You are running them together in your last paragraph. The Objectivist answer--and it is just the same as the Aristotelian answer--is that we have an innate capacity or potential for knowledge, but not actually innate knowledge.

It's obviously true that any inborn capacity is not there by choice. It's also true that some of our inborn capacities may be actualized automatically--like our ability to perceive objects. But this is not a problem for free will as long as free will concerns a different capacity. That is why the difference between perception and conceptualization is important. *Man* may be saddled with inborn tendencies, but that does not mean free will is. He may have free will with respect to the use of some capacities, but not with respect to others.

Regarding the general question of whether free will in regard to any capacity is possible, that is a big new separate question that we could go on forever about. My point is that there's nothing cognitive science has to contribute to this question, because it is philosophical. The questions you raise about physics and about predispositions to think are not questions raised by cognitive science.

gregnyquist said...

Anon: "Either you or Pinker are wrong (I'm not sure) in identifying the cognitive revolution with the modularity theory of mind."

Pinker is using the term "cognitive revolution" to cover all the sciences dealing with human nature and the mind, including especially neuroscience, cognitive science, behavioral genetics and evolutionary psychology. There are, to be sure, no absolute, precise or perfect consensus either in or among any of these sciences. Evolutionary psychology is not as empirically well grounded as the other sciences, and behavioral genetics challenges views that only 50 years ago were dominant (and are still dominant in the social sciences). This breeds resistence. Indeed, it among those that don't like behavioral genetics and evolutionary psychology that you're most likely to find resistence to Pinker's modular theory of the mind.

Anon: "That said, even if modularity theory turns out (luckily, in my opinion) to be correct about the existence of physically distinct modules responsible for elements of our cognition, I don't see any reason why this would contradict AR's 'tabula rasa' view."

True, it would possible to develop a modular theory that is more in keeping with the tabula rasa view. That is why the challenge to Objectivism does not come from the modular theory per se, but from the theory when its combined with behavioral genetics (see my original post).

Anon: "As for the question of "perfect harmony," I'm not sure where you're getting that idea from, but notice that even Pinker speaks of the allegedly distinct modules 'cooperating to generate a train of thought or an organized action.'"

The "perfect harmony" phrase comes from Rand's Playboy interview. It's a phrase describing Rand's obsession with harmonizing emotion with thought. Now this very notion of emotions conflicting with thought is itself a bit of misnomer. Emotions are integrated with thoughts on a much deeper level, so that any perceived conflict between thought and emotion is really a conflict between rival emotions, one of which is misidentified with a thought. The modular theory of the mind, if regarded as a product of natural selection, provides a plausible grounding for this view.

It's important to note that the Pinker version of the modular theory of the mind does not represent the most critical challenge to Objectivism coming from the cognitive revolution. The principle challenge comes from behavioral genetics (which will be covered in my next post). The inclusion of the modular theory is provided merely to give people a taste of what is going on among the sciences of human nature. Perhaps in all of its details Pinker's theory will be found wanting. But since the brain itself is modular, it seems plausible that any theory of the mind that makes use of brain science is going to have to make some concessions to modularlity.

gregnyquist said...

Anon: "Man may be saddled with inborn tendencies, but that does not mean free will is. He may have free will with respect to the use of some capacities, but not with respect to others."

This is not really the point at issue, as I see it. The question, in the final analysis, has to do with behavior: that is to say, do some, most or all people have innate predispositions along with free will that causes them to more likely behave in one way than another? Cognitive science, for example, has discovered that human beings are not good at logical reasoning. Is this due to innate causes, or is it the result of Kant's influence on the culture? However this question is answered has huge implications for Objectivism.

gregnyquist said...

Jay: "Isn't it possible that by studying things like evolutionary psychology, people can learn how to better harmonize their consciousness?"

Possibly. However, what if evoltionary psychology tells us that a harmonized consciousness poses a danger to our survival? Perhaps people who are conflicted (up to a certain point) are more attentive to what's going on around them than people who aren't, which gives them leg up in darwininan competition. There are some philosophers (e.g., Nietzsche) who believee that a certain level of tension in the soul is good.

There are, to be sure, other more rigorous approaches to attaining something along the lines of "harmony." Indian philosophy (from India, not Native Americans) has its own theories (abnegation of will, and so forth) which find a western echo in the works of Schopenhauer. Spinoza found liberation from the "bondage" of contrary emotions in the disinterested cognition of truth. Santayana believed that the individual could find moments of respite in contemplating the aesthetic qualities of the data of the mind.

PhysicistDave said...

anonymous,

You refer to some Objectivists who “invoke the work of J.J. Gibson who argued that perception is direct…”

With all due respect, that view is equivalent to arguing that the sun moves around the earth or that the earth is flat!

Whatever one’s philosophical views, as a matter of physics and physiology, we know that we perceive the world through a complicated indirect process: light bounces off an object, enters our eye, and is refracted onto the retina. This induces complex electrochemical phenomena, an initial signal-processing stage near the eye, and then signals are sent on to the brain.

Forget the brain. Even before you reach the brain, the process is surely not direct!

I know that philosophers keep “direct realists” around for amusement just as circuses might like to have a kangaroo. But, as a scientist (Ph.D. in physics from Stanford), I view direct realists as a sad joke.

I also notice that your posts offer various twists and turns and strategies in an effort to defend Objectivism.

Why bother?

If the simple and obvious meaning of something Rand wrote is wrong, why not say, “Okay, she was wrong on that!” and move on.

That’s what I do as a scientist. I can think of errors made even by great scientists like Einstein. I just admit that even smart people make mistakes. I do not try to re-interpret their work to turn the mistake into a non-mistake.

I see this let’s-save-Rand-from-herself stance constantly with Objectivists. Perhaps this is a sign that something is really wrong with Objectivism?

All the best,

Dave

Robert Campbell said...

Steven Pinker is a fascinating writer, whose books I frequently recommend to nonpsychologists who want to get some sense of what the field is like.

He is also a crappy historian of psychology.

The notion of a "module" was not part of the Cognitive Revolution of the 1950s, which drew on computer science, linguistics, and information theory, as well as some pre-existing traditions within psychology.

Nor is modularity the same idea as localization of function in the brain (which goes back well into the 1800s). Modularlty didn't really come together until the early 1980s, when Jerry Fodor published a little book called The Modularity of Mind. (Fodor, as noted in some comments to the second installment up above, has since had second thoughts about what he was advocating).

The massively modular stuff that Pinker favors (though he has moderated it somewhat since his book How the Mind Works) is just the latest version of evolutionary psychology. There has been one sort of evolutionary psychology or another ever since The Descent of Man was published; i.e., for virtually the entire period since psychology became a separate discipline. Pinker is correct in complaining about the widespread acceptance of enviromental determinism during part of the 20th century (particularly in the form of behaviorism and culturally relative anthropology), but he seriously exaggerates its sway over cognitive psychology in recent decades.

Robert Campbell

PS. One reason for skepticism concerning Pinker's variant of evolutionary psychology is his intellectual dependence on Noam Chomsky, one of the most anti-evolutionary thinkers of the past century.

Robert Campbell said...

Dave,

Objectivists, as you've noted, sometimes dismiss thinkers whose works they've never troubled to read, on the basis of a snap philosophical diagnosis.

I hope you're not falling into this kind of trap with regard to J. J. Gibson.

Gibson was one of the greatest psychologists of the past century.

He was, of course, aware of what is going on in the retina and the optic nerve and (later in his career) in the primary visual cortex and so forth.

His claim about the directness of perception pertained to relationships between aspects of the environment for perception and aspects of perceptual experience, not to the underlying neurophysiology.

The position he objected to was one that claimed that we (somehow) perceive a "proximal stimulus" (in visual terms, a "retinal image") in the form of 2-dimensional snapshots that must then be informationally enriched with depth and a temporal dimension and indications of what actions are possible with regard to them, etc. etc. etc.

Gibson and his students discovered quite a few visual phenomena previously unknown to psychology and came up with new explanations of other previously known phenomena.

There are some problematic aspects to Gibson's framework. One useful discussion of them is in a little book titled On the nature of representation: A case study of James J. Gibson's theory of perception by Mark Bickhard and D. Michael Richie.

But writing off Gibson as a relic or a crank is not recommended.

Robert Campbell

Robert Campbell said...

Anonymous,

You say:

"Any modularity theorists who argue that the existence of innate modules implies the existence of innate knowledge are confusing the potential with the actual--a mistake that has been made by nativists since Plato. Just because there is a ready-made system responsible for acquiring some cognitive content does not mean that we possess that content from birth."

What is your notion of cognitive content here?

Robert Campbell