The Contemporary Relevance of The Sensory Order

In the last decade or two, Austrians and other Hayek scholars have "rediscovered" Hayek's work in theoretical psychology and many of us have tried to tie that work to his larger political economy.  One of the most interesting things about The Sensory Order is how ahead of its time it was.  Its basic framework for understanding how the mind works was more or less correct as recent work in neuroscience has shown.  In particular, Hayek's general claim that the mind is a classifier and "interpreter" of sensory data, rather than just a passive receptor, is more and more accepted as an accurate picture of the process.

Evidence of this comes from lots of places, but one of the most fascinating is this article in The New Yorker about the way in which the mind can create sensations even when the physical receptors are absent and how the mind can be tricked into ending painful or uncomfortable sensations through the use of mirrors and other forms of misdirection.  (Plus it has a cool story about a woman whose scalped itched so bad, she scratched right through to her brain!)  Those who know The Sensory Order will understand what's happening in that article at a whole different level.  I particularly like the author's use of the phrase "the mind's best guess" to describe how the mind constructs what it "thinks" should be there, whatever the direct sensory inputs might be.

Another really cool piece of evidence for the basic framework of The Sensory Order is an optical illusion known as the "Charlie Chaplin Mask."  Again, the mind "fills in" what is supposed to be there based on its past experience, even though we rationally know that's not what's there.  It's "the mind's best guess" in action.  You really have to force your mind to see what IS there not what it's making you THINK is there.

Enjoy.

Comments

[Dain]

Plus it has a cool story about a woman whose scalped itched so bad, she scratched right through to her brain!

And the A-Hypochondriac-Should-Avoid-Reading-This Award goes to...

[matthew mueller]

Yes, I have been pushing this argument for months. The best treatment of Hayek's theory of psychology in terms of its relevance for economics is the work being done by Butos and Koppl. Austrians still insist that Hayek's knowledge problem argument (knowledge as diffuse) remains a sufficient counter to Keynes' objection to the self-correcting mechanisms of free markets. Butos and Koppl say that this is not true. The real problem is not the coordination and utilization of knowledge, but of the generation of knowledge. And for this, they point to Hayek's "sensory order" book./ In short, the market is not a "discovery process", but a "knowledge-generating process." Of course any Shackle enthusiast can find much to recommend this claim. But I think this is where Austrians should be turning. Kirzner is a brilliant economist, but don't let his emphasis on entrepreneurial discovery arrest the development of Austrian economics from recognizing the importance of the market as a process of knoweldge generation.

Here are some references that I would encourage students on this blog to read:

Butos (2001) "Garrison and the Keynes Problem" QJAE.

Butos (2003) "Knowledge Questions" RAE

Butos and Koppl (1993) "Hayekian Expectations" Constitutional Political Economy.

Butos and Koppl (1997) "The Varieties of Subjectivism" History of Political Economy.

Koppl and Butos (2001) "Confidence in Keynes and Hayek" Review of Political Economy

[Rafael Guthmann]

Matthew, I think that what you call "generation of knowledge" is what Hayek called knowledge. I could say that the world is not in general equilibrium because the knowledge generation capacity of the individual human mind is weak compared to the total amount of knowledge that need to be generated. If the human mind could perceive everything (generate the knowledge) in an infinitesimal amount of time (infinite knowledge generation capacity), then general equilibrium would be reality and socialism would work.

In your language, I could say that the market is a mechanism that unifies the capacity of knowledge generation of the individual minds, generating order in an economic system much more complex than a single individual mind could. Kirzner's theory of the entrepreneurial discovery process is a precise theory about the workings of this mechanism.

Also,I think that Keynes's theory is inconsistent with subjectivism and methodological individualism. With makes his theory irrelevant to the Austrian understanding of the market economy.

[Rafe]

Steve, with all due respect, this Kantian take on the active mind is critically important but also old hat. At the level of sensory perceptions consider the "phantom limb" effect that has been well documented for ages, at least since the war between the states. There is a longstanding literature from Gestalt psychology on pattern recognition and the search for pattern in the environment. For example very young babies can modify their rate of sucking on a bottle to "suck" a pattern into focus that is projected in front of them.

It was the publication of Hebb's book "The Organization of Behavior" (1948?) with the same major thesis as Hayek that promoted F A to complete his ms. I appreciate that the oundatio was laid in the 1920s but gestalt psychology was active at the time and the active, problem-solving mind was being explored before WWI by the Wurzberg school led by Kulpe. He might have achieved a status in his field comparable to Menger but he died in 1914 at the early age of 53 with only one volume published of a four-volume project. His two closest associates also had their careers cut short, Selz was murdured by the Nazis in the 1930s and Buhler was forced to leave Austria shortly after, a move which effectively ended his major project with only one of three proposed books in print. For a comparison, one is tempted to speculate how Menger's ideas would have survived without the advocacy of Mises and Hayek through their long and active lives.

For more on that line of thought in psychology and the philosophy of science, especially Buhler. http://www.the-rathouse.com/K_and_C_Buhler.html

Karl Buhler would have been the most likely reference for Hayek because he worked in Vienna up to 1938 when he was taken into "protective custody" by the nazis. He then walked over the border to start a new life in the US. Like Mises, he lost his library and papers but unlike Mises he did not manage to get a fresh start in his career and he ended up keeping the records of his wife's counselling practice in Hollywod. Charlotte Buhler pioneered the detailed study of child development that later made Piaget famous. She was also a leader in the Third Force or Humanistic Psychology movement in the US.

Buhler's book on the crisis in psychology (circa 1929) attempted to overcome the disunity in methods in psychology by recognising the results of different methods and seeking for ways to combine or unify them. This can be compared with Menger's effort to achieve integration between the different methods of the social sciences. Buhler's book should have started a debate on methods in psychology but apparently it fell dead from the press. Buhler was also a foundation scholar in the study of language and semiotics but he was written out of the history of psycholgy by rivals who were better placed during the latter part of his life.

The stir of interest in the active mind at present is partly a reflection of new technologies which make some things more visible to people who never noticed them before, and partly a reflection of the domination of psychoanalysis and behaviorism in the mainstream of psychology so that better ideas that were well developed decades ago are not common knowledge. The comparison with the Austrian school comes to mind. This is an article on the way that Buhler was written out of the history of the field.

http://findarticles.com/p/articles/mi_qa3711/is_199407/ai_n8726847/pg_1#

A man close to the history of these ideas has advised that it only touches the surface of the vandalism and mendacity of the villains of the piece. Again some critics of Austrian ideas come to mind!

[Steve Horwitz]

Rafe,

I certainly wasn't trying to argue that Hayek was somehow alone in making these arguments, nor that TSO was itself responsible for them. Of course Hebb's book is central, as Hayek himself recognized. And while the Kantian ideas have been around a long time, we now understand that the "categories" are not a priori but tied to the history of the person and the species.

Hayek was part of a larger intellectual current for sure. That said, TSO remains one of the earliest efforts to elucidate what is now fairly commonplace, as neuroscientists who have read it would acknowledge.

[Roger Koppl]

Rafe: I think Hayek discovered Hebb's work only after he had returned to the 1920 manuscript. I think that is what he says in TSO. Do you recall the evidence for your idea that it was the publication of Hebb's work that induced him to write TSO?

[Rafe]

Roger, yes, the ms was drafted in 1920 or 1921, Hebb's book induced him to return to it and complete the book for publication.

Steve, if you want to explore the significance of TSO in relation to Hayek's program you probably need to see what Jack Birner has written on that topic.

"we now understand that the "categories" are not a priori but tied to the history of the person and the species." Something along those lines can be found in Human Action, that is the evolutionary approach that later became known as "evolutionary epistemology" in the hands of Lorenz, Popper, Campbell and others. Radnitzky and Bartley edited a particularly helpful collection of essays on the topic.
http://www.the-rathouse.com/revradbart.html

[Roger Koppl]

Rafe:

I know what you meant, but your statement contradicts Hayek's explicit claim in TSO that he found out about Hebb only late in the process. He was working on TSO, planning to publish it, and then read Hebb's Oganization of Behavior and almost scrapped the project. He decided, however, that he (Hayek) had done a better job of drawing out the underlying principles and, especially, their philosophical consequences and that, therefore, there was room for him to go ahead and finish TSO in spite of Hebb's priority. Anyway, that's how I remember Hayek's comments in TSO.

You said something quite different, namely, that reading Hebb induced Hayek to return to the his old MS and turn it into a publishable book.

If you are right, Hayek is wrong and vice versa So who is right? It might be you! Hayek's statement in TSO could be a confabulation or something. Thus, I want to know your source. How do you know what you claim? Perhaps you don't recall. I get that; this is a blog site and no one is writing for a peer review process. But if you do recall your source, please let me know what it is.

[Roger Koppl]

Hayek’s central point in TSO was that memory precedes perception, not the other way around. That’s quite radical, quite fundamental. Thus, while he drew on, for example, Gestalt theory, it would be an error to say he was merely restating it. I don't think Rafe was saying otherwise, but I thought it might be worthwhile to give an explicit reminder of Hayek's central thesis in TSO.

I think Matt Mueller is right to suggest that Bill Butos and I have pushed the argument a bit further than “dispersed knowledge.” If I know what he means about knowledge *generation*, credit goes to Bill and Thomas McQuade. I can’t take any credit for that particular bit of the argument. Even without the Butos/McQuade point, however, I think Bill and I made some progress by articulating a theory of expectations that was not quite present in Hayek and that gives (I think) a pretty fair answer to Keynes’ Chapter 12. And I think that theory does avoid the view that somehow the “dispersed knowledge” is comparable to so many pebbles on the beach, objectively “out there” from the start. I think Kirzner’s 1982 article “Uncertainty, Discovery, and Human Action: A Study of the Entrepreneurial Profile in the Misesian System” shows that Kirzner recognized perfectly well the role of creativity, but I also think he was weak on this point prior to 1982 and easily gives you the knowledge-as-pebble idea.

Rafael: I think it’s clear that the old “hydraulic” Keynesians violated both subjectivism and methodological individualism. Can we really say that same of Keynes himself?

[Rafe]

Thanks Roger, if you have checked then I accept your account because I have not read TSO lately and I don't have any other source. BTW I always like to cut people some slack on blogs, the first comment can be treated a bit like conversation but as soon as it gets serious the references need to be correct.

[Sinclair Davidson]

If I could give Julie Novak's monograph "Sensory order and economic order" a plug.

https://www.sslcis.org/cart/index.php?main_page=product_info&products_id=9

[Rafael Guthmann]

"I think it’s clear that the old “hydraulic” Keynesians violated both subjectivism and methodological individualism. Can we really say that same of Keynes himself?"

(note that my english is not very good)
I think that we can say that. Keynes did have arguments in his theory that were grounded on methodological individualism and subjectivism, but had other arguments that violated these principles. His theory was contradictory, and the parts that were grounded on methodological individualism contradicted the rest of the theory.

I think that hydraulic keynesianism is the result of the process of formalization that "straighted out" Keynes's theory. This process of formalization jettisoned the parts of the theory that were grounded on subjetivism because they contradicted the rest of his theory.

[Steven Horwitz]

Let me second Sinclair Davidson's endorsement of Julie's monograph. I just read that in the last week and it's one of the best introductions to TSO and Hayek's social thought around.

And Rafe: I have explored the TSO/social thought link myself, and perhaps before Birner did:

http://www.gmu.edu/rae/archives/VOL13_1_2000/horwitz.pdf

[Roger Koppl]

Rafael: I think that's reasonable. I think you're talking about the old "Keynesian Cross" with Y=C+I+G. Personally, I prefer to give the model -- as it appears in Keynes -- an individualist and subjectivist reading. C is a stable function of Y because consumers are passive in his system, acting on habit and current income. I is volatile and unpredictable precisely because it depends on the expectations mechanisms of chapter 12. G is neither volatile nor stable; it is the policy instrument. This macroeconomic theory is not the one I prefer! But I think it is fair to say 1) it is in Keynes and 2) it respects subjectivism and methodological individualism.

[Paul Lewis]

Rafael (if I may?),

As Roger has pointed out, there are many different readings of Keynes. One that you might find interesting, as it does justice to Keynes's subjectivist tendencies, is provided by Tony Lawson in the following papers:

Lawson, T. (1985). ‘Uncertainty and Economic Analysis.’ Economic Journal, 95: 909-27.
__________ (1987). ‘The Relative / Absolute Nature of Knowledge and Economic Analysis.’ Economic Journal, 97: 951-70.

Hope they're useful.

[Bogdan Enache]

I have two questions regarding your comments, prof. Koppl:

1. I'm (vaguely) familliar with Gestalt psychology theory, however, I don't know to what extent can it be said that Hayek went beyond it ?

2. How does the large emphasis on aggregation in Keynes squares with interpreting him as compatible with both subjectivism and metholodological individualism - at least the "Austrian" varieties?

[Roger Koppl]

Hi Bogdan,
1. Gestalt is descriptive TSO is explanatory. The Gestalt theory says that our perceptual wholes generally determine the parts and not the other way around. Hayek said memory precedes sensation. So Gestalt is *describing* something fundamental about our sensations, whereas Hayek is *explaining* where those sensations come from.
2. I don’t know how much emphasis on aggregates there was in Keynes. He did use those aggregates, C, I, G, it is true, but I don’t know how much he emphasized them. Anyway, there is nothing about subjectivism and methodological individualism telling us not use aggregates. It is just that any aggregates we use should have a sensible relation to understandable human actions. We use aggregates all the time. Supply and demand are aggregates, including money demand. No one objects to doing so. We object only when the aggregates are imagined to move in ways that are not sensibly related to what we know about people.

The cost-push theory of price inflation says that wages rise because of higher prices, which stimulate further wage hikes in a wage-price spiral. Methodological individualists object that this story implicitly assumes that people do not care about the size of their real cash balances. We know, however, that people do care about the sized of their real cash balances. The methodological individualist’s objection is not to the use of aggregates; he or she objects that the behavior of one aggregate (real cash balances) is not consistent with our reasonable sense of what real people would really do in the posited situation.

I hope those remarks are helpful.

[Roger Koppl]

Hi Bogdan,
1. Gestalt is descriptive TSO is explanatory. The Gestalt theory says that our perceptual wholes generally determine the parts and not the other way around. Hayek said memory precedes sensation. So Gestalt is *describing* something fundamental about our sensations, whereas Hayek is *explaining* where those sensations come from.
2. I don’t know how much emphasis on aggregates there was in Keynes. He did use those aggregates, C, I, G, it is true, but I don’t know how much he emphasized them. Anyway, there is nothing about subjectivism and methodological individualism telling us not use aggregates. It is just that any aggregates we use should have a sensible relation to understandable human actions. We use aggregates all the time. Supply and demand are aggregates, including money demand. No one objects to doing so. We object only when the aggregates are imagined to move in ways that are not sensibly related to what we know about people.

The cost-push theory of price inflation says that wages rise because of higher prices, which stimulate further wage hikes in a wage-price spiral. Methodological individualists object that this story implicitly assumes that people do not care about the size of their real cash balances. We know, however, that people do care about the sized of their real cash balances. The methodological individualist’s objection is not to the use of aggregates; he or she objects that the behavior of one aggregate (real cash balances) is not consistent with our reasonable sense of what real people would really do in the posited situation.

I hope those remarks are helpful.

[Anonymous]

"General Theory" is full of faulty reasoning. But, it is constructed in a quite logical and defendible way, and constructed from an individualist standpoint. The problem is that it's full of holes.

[TGGP]

Hayek's theory would be considered "connectionist". Connectionist theories have been losing ground to "comutationalism". Steven Pinker (an adherent of the latter school) discusses the divide in some of his books like "How the Mind Works".

[Rafael Guthmann]

"It is just that any aggregates we use should have a sensible relation to understandable human actions."

Roger, but I think that these aggregate statistical measures (Y,G,I,C) abstract from the intricacies of the economic system. And from the perspective of methodological individualism and subjetivism the most important macroeconomic problems are caused by the massive complexity of the process of coordination of millions of individual plans. An approach to macroeconomic problems that is firmly based on subjetivism and methodological individualism (like Hayek's approach) focus on macroeconomic problems from this perspective, and from this perspective, an increase in G cannot be used to help to coordinate plans that were discoordinated because of a unanticipated fall in I, since shocks such as an increase in G and an unanticipated fall in I have disproportional effects on the relative prices in the economic system, with means that an increase in G cannot be used to cancel the effects of a unanticipated fall in I.

Since Keynes did not appear to have perceived this type of problem I think that he didn't base his though firmly on methodological individualism and subjetivism, even though he had some influence from this approach to economics.

[Roger Koppl]

TGGP: Everything depends on what you think "computationalism" or "connectionism" entail. The two are not necessarily at odds. If computationalism requires us to think of the mind as a Turing machine, then I think the work of Tito Arecchi pretty much redeems a Bergsonian or Husserlian view against computationalist models. But I don't think we have to chose one or the other; I think they are complements.

[Roger Koppl]

Rafael,

But now you are explaining why we should prefer capital-based macro to hydraulic Keynesianism. Right! But your argument is not that "aggregates are methodologically taboo." Your argument is that the particular aggregates chosen by Keynes obscure relative price effects that matter. I agree! But that's a dispute in economic theory, not methodology.

[Anonymous]

After reading Rafael's post I'm a bit more in sympathy with that point of view.

Criticism of old-style Keynesianism goes beyond capital. If the capital stock of the world were necessarily fixed then that Keynesianism would still be incorrect because Say's law is essentially true.