[HN Gopher] "The Mind in the Wheel" lays out a new foundation fo...
___________________________________________________________________
"The Mind in the Wheel" lays out a new foundation for the science
of mind
Author : CharlesW
Score : 116 points
Date : 2025-05-15 18:36 UTC (1 days ago)
(HTM) web link (www.experimental-history.com)
(TXT) w3m dump (www.experimental-history.com)
| almosthere wrote:
| Just do stuff. To fix the mind, you have to keep busy. Don't
| focus on yourself, focus on projects. The long history of
| humanity, the last thing people did until recently was think
| about themselves. Stop. Do things.. basically touch grass.
| bbor wrote:
| In the interest of staying within the rules of the site, I'll
| just say that this is a clueless, offensive, harmful comment.
| No, you don't know more than a 125+ years of psychologists
| studying mental health disorders because you make decent money
| right now and feel ok.
| d3ckard wrote:
| I would like to provide a piece of anecdotal evidence: what
| he suggests actually worked for me best.
| yeahsure wrote:
| Just another N=1 anecdote: It didn't work for me when I was
| dealing with depression a few years ago.
|
| I also lost a dear friend to suicide and he was very
| succesful and active on his field when it happened. Nobody
| saw it coming.
|
| It's just not that simple.
| kayodelycaon wrote:
| I truly hope it does but I have seen plenty of people use
| that strategy until something broke.
|
| As I see it, you can either deal with a problem on your own
| terms or you can let it eventually deal with you on its
| terms.
| yeahsure wrote:
| This comment comes across as quite naive. If the solution were
| truly that simple, we wouldn't see so much people suffering
| from depression, anxiety, addiction, etc. around the world.
| Trasmatta wrote:
| Just "doing things" will not heal trauma. Doing things can be a
| way to try to suppress the pain of trauma, but it will still be
| there, just obscured and affecting the person's life in ways
| they can't seem to understand. And alternatively, trauma can
| also be a major blocker for actually being ABLE to do things.
| perching_aix wrote:
| Nothing quite like treating the symptoms instead of the root
| cause, am I right?
| huijzer wrote:
| I have done a MSc in CS and PhD in psych dept. Sometimes I had
| the feeling that the system didn't want me to get anything done.
| staunton wrote:
| Can you elaborate? It's hard to take much away from your
| comment without any more details.
| huijzer wrote:
| Well peer review primarily is a very old system where nobody
| seems to care about speeding up the process. Why also are
| journals allowed to make huge amounts of money without paying
| the academics and without providing much services either.
| ravenstine wrote:
| Maybe I'm missing the point of the article, but the application
| of cybernetics to psychology was already proposed (albeit not by
| a psychologist) at least as far back as 1960 in the book _Psycho-
| cybernetics_. This "new paradigm" doesn't sound particularly
| new.
|
| This sentence also puzzled me:
|
| > Lots of people agree that psychology is stuck because it
| doesn't have a paradigm
|
| Psychology might not have a _grand unifying_ paradigm, but it 's
| been highly paradigm-driven since its inception.
| profstasiak wrote:
| exactly. I cannot believe they mention cybernetics and
| psychology in one sentence:
|
| > The science of control systems is called cybernetics, so
| let's call this approach cybernetic psychology.
|
| but they never mention psycho-cybernetics
| floxy wrote:
| I always thought someone should make a second Matrix movie.
| psunavy03 wrote:
| https://xkcd.com/566/
| kogir wrote:
| Another example for all you computer folks out there: ultimately,
| all software engineering is just moving electrons around.
| But imagine how hard your job would be if you could only
| talk about electrons moving around. No arrays, stacks,
| nodes, graphs, algorithms--just those lil negatively charged bois
| and their comings and goings.
|
| I think this too easily skips over the fact that the abstractions
| are based on a knowledge of _how things actually work_ - known
| with certainty. Nobody in CS is approaching the computer as an
| entirely black box and making up how they _think_ or _hope_ it
| works. When people don 't know how the computer actually works,
| their code is wrong - they get bugs and vulnerabilities they
| don't understand and can't explain.
| gchamonlive wrote:
| > When people don't know how the computer actually works, their
| code is wrong - they get bugs and vulnerabilities they don't
| understand and can't explain.
|
| While this is true, we're usually targeting a platform, either
| x86 or arm64, that are incredibly complex pieces of
| engineering. Unless you are in the IoT or your application
| requires you to optimize at the hardware level, we're so
| distant from the hardware when we're programming in python for
| instance that the level of awareness required about the
| hardware isn't that much more complicated than the basic Turing
| machine.
| quadhome wrote:
| Physical latencies in distributed systems design. Calibration
| for input devices. Block storage failures and RAID in
| general. Monitor refresh rates. Almost everything about
| audio. Rowhammer.
| inglor_cz wrote:
| For me, a computer is at best semi-transparent.
|
| I can rely on a TCP socket guaranteeing delivery, but I am not
| very well versed in the algorithms that guarantee that, and I
| would be completely out of my depth if I had to explain the
| inner workings of the silicon underneath.
| jtbayly wrote:
| Plenty of programmers know nothing about electrons. Think kids.
|
| _Most_ programmers never think once about electrons. They know
| how things work at a much higher level than that.
| DontchaKnowit wrote:
| Yeah? So what. Theyre still using abstractions that were
| created by people who know about electrons.
| bluGill wrote:
| That only works because some EE has ensured the abstractions
| we care about work. You don't need to know everything, you
| just need to ensure that everything is known well enough by
| someone all the way down.
| manugo4 wrote:
| No, red is an abstraction that is not based on knowledge of how
| colors work.
| ysofunny wrote:
| it is an abstraction based on how our biological eyes work
| (this implies "knowledge" of physics)
|
| so it is indirectly based on knowledge of how color works,
| it's simply not physics as we understand it but it's
| "physics" as the biology of the eye "understands" it.
|
| red is an abstraction whose connection to how colors work is
| itself another abstraction, but of a much deeper complexity
| than 'red' which is a rather direct abstraction as far as
| abstraction can go nowadays
| xboxnolifes wrote:
| There is absolutely no knowledge needed for someone to
| point to something that is red and say "this is red" and
| then for you to associate things that roughly resemble that
| color to be red.
|
| Understanding the underlying concepts is irrelevant.
| Finbel wrote:
| Except I could think they mean the name of the thing, the
| size of the thing or a million other things. Especially
| if i have no knowledge of the underlying concept of
| colors.
| lo_zamoyski wrote:
| "How colors work" is dubious.
|
| In physics, color has been redefined as a surface reflectance
| property with an experiential artefact as a mental correlate.
| But this understanding is the result of the assumptions made
| by Cartesian dualism. That is, Cartesian dualism doesn't
| prove that color as we commonly understand it doesn't exist
| in the world, only in the mind. No, it _defines_ it to be the
| case. _Res extensa_ is defined as colorless; the _res
| cogitans_ then functions like a rug under which we can sweep
| the inexplicable phenomenon of color as we commonly
| understand it. We have a _res cogitans_ of the gaps!
|
| Of course, materialists deny the existence of spooky _res
| cogitans_ , admitting the existence of only _res extensa_.
| This puts them in a rather embarrassing situation, more
| awkward that the Cartesian dualist, because now they cannot
| explain how the color they 've defined as an artefact of
| consciousness can exist in a universe of pure _res extensa_.
| It 's not supposed to be there! This is an example of the
| problem of qualia.
|
| So you are faced with either revising your view of matter to
| allow for it to possess properties like color _as we commonly
| understand them_ , or insanity. The eliminativists have
| chosen the latter.
| jchanimal wrote:
| Love your perspective. It reminds me of this argument I'm
| working on about Turing machine qualia. Maybe my argument
| is just Searle in disguise? https://x.com/jchris/status/181
| 5379571736551923?s=46&t=8A60w...
| ajross wrote:
| There's no definition for "color" in physics. Physics does
| quantum electrodynamics. Chemistry then uses that to
| provides an abstracted mechanism for understanding
| molecular absorption spectra. Biology then points out that
| those "pigments" are present in eyes, and that they can
| drive nerve signals to brains.
|
| Only once you're at the eye level does anyone start talking
| about "color". And yes, they define it by going back to
| physics and deciding on some representative spectra for
| "primary" colors (c.f. CIE 1931).
|
| Point being: everything is an abstraction. Everything
| builds on everything else. There are no simple ideas at the
| top of the stack.
| lo_zamoyski wrote:
| > There's no definition for "color" in physics.
|
| This is unnecessarily pedantic. Your explanation
| demonstrates that.
|
| > There are no simple ideas at the top of the stack.
|
| I don't know what a "simple idea" is here, or what an
| abstraction is in this context. The latter has a
| technical meaning in computer science which is related to
| formalism, but in the context of physical phenomena, I
| don't know. It smells of reductionism, which is
| incoherent [0].
|
| [0] https://firstthings.com/aristotle-call-your-office/
| ffwd wrote:
| > To untutored common sense, the natural world is filled
| with irreducibly different kinds of objects and
| qualities: people; dogs and cats; trees and flowers;
| rocks, dirt, and water; colors, odors, sounds; heat and
| cold; meanings and purposes.
|
| It's too early to declare that there are irreducible
| things in the universe. All of those things mentioned are
| created in the brain and we don't know how the brain
| works, or consciousness. We can't declare victory on a
| topic we don't fully understand. It's also a dubious
| notion to say things are irreducible when it's quite
| clear all of those things come from a single place (the
| brain), of which we don't have a clear understanding.
|
| We know some things like the brain and the nervous system
| operate at a certain macro level in the universe, and so
| all it observes are ensembles of macro states, it doesn't
| observe the universe at the micro level, it's then quite
| natural that all the knowledge and theories it develops
| are on this macro scopic / ensemble level imo. The
| mystery of this is still unsolved.
|
| Also regarding the physics itself, we know that due to
| the laws of physics, the universe tends to cluster
| physical matter together into bigger objects, like
| planets, birds, whatever. But those objects could be
| described as repeating patterns in the physical matter,
| and that this repeating nature causes them to behave as
| if they do have a purpose. The purpose is in the
| repetition. This is totally inline with reductionism.
| ajross wrote:
| > I don't know what a "simple idea" is here
|
| To be blunt: it's whatever was in your head when you
| decided to handwave-away science in your upthread comment
| in favor of whatever nonsense you wanted to say about
| "Cartesian dualism".
|
| No, that doesn't work. If you want to discount what
| science has to say you need to meet it on its own turf
| and treat with the specifics. Color is a theory, and it's
| real, and fairly complicated, and Descartes frankly
| brought nothing to the table.
| growlNark wrote:
| > I think this too easily skips over the fact that the
| abstractions are based on a knowledge of how things actually
| work - known with certainty.
|
| models [?] knowledge, and a high degree of certainty is not
| certainty. This is tiring.
| achierius wrote:
| But doesn't this argument defeat itself? We cannot, a priori,
| know very much _at all_ about the world. There is very, very
| little we can "know" with certainty -- that's the whole
| reason Descarte resorted to the whole cogito argument in the
| first place. You and GP just choose different lines to draw.
| growlNark wrote:
| Yes, I agree completely. I think the apriori/aposteriori
| distinction is always worth making though.
|
| This really does matter a lot more when floating signifiers
| get involved; I'm not actually contesting that our models
| of electrical engineering model reality quite well.
| AIPedant wrote:
| This seems like a misreading of the comment. The models and
| knowledge of arrays, classes, etc, are known with
| "arbitrarily high" certainty because they were designed by
| humans, using native instruction sets which were also
| designed by humans. Even if this knowledge is specialized, it
| is readily available. OTOH nobody has a clue how neurons
| actually work, nobody has a working model of the simplest
| animal brains, and any supposed model of the human mind is at
| best unfalsifiable. There's a categorical epistemic
| difference.
| jemmyw wrote:
| It doesn't skip over it. First this is an example and not the
| primary thing it's talking about. But secondly, just above, the
| article states that some lower level knowledge is necessary in
| the transit example. If you map those things, as written by
| someone who, as they say, isn't they knowledgeable about
| programming, then they make sense without diving into the
| specific.
| pimlottc wrote:
| > Nobody in CS is approaching the computer as an entirely black
| box and making up how they think or hope it works.
|
| Haven't you heard about vibe coding?
| ImHereToVote wrote:
| "Nobody in CS is approaching the computer as an entirely black
| box and making up how they think or hope it works."
|
| That is literally how we approach transformers.
| danielmarkbruce wrote:
| Who is "we"? Lot's of people (including me) know how how
| transformers work. Just because we can't do all the math in
| our head quickly enough to train a model or run inference
| mentally, doesn't mean we don't know mechanically how they
| work.
| ImHereToVote wrote:
| We know how they are trained. We just don't know how the
| trained model works, since the program is emergent.
| danielmarkbruce wrote:
| Lol, we also know how inference works. The fact that LLMs
| turned about to be surprisingly effective doesn't mean we
| don't know how it works. There are many fields where we
| know the underlying physics and it's just difficult to
| actually predict real world results because there are so
| many calculations. What's next, you are going to tell an
| aerospace engineer that flight is "emergent" because we
| need to run simulations or experiments?
| ImHereToVote wrote:
| The actual program is a black box. We have been able to
| dissect some details but the whole system is hard to
| understand. The program is grown more than developed.
| Understanding the concept of inference doesn't help you
| much.
| danielmarkbruce wrote:
| "the program" is some silly abstraction you've made up.
| If you don't understand the underlying mathematical
| operations that's fine, but many of use do. And they
| aren't that complicated in the grand scheme.
|
| Every complex system is hard to understand due to the
| number of variables v human working memory.
| lo_zamoyski wrote:
| Computer science has nothing to do with physical computing
| devices. Or rather, it has to do as much with computers as
| astronomy has to do with telescopes. You can do it all on
| paper. The computing device doesn't afford you anything new,
| but scale and speed for simulating the mechanical work doing it
| on paper would. Electrons are irrelevant. They are as relevant
| to computer science as the species of tree from which the wood
| in your pencil comes from is relevant to math.
|
| Obviously, being able to use a computer is useful, just as
| using a telescope is useful or being able to use a pencil is
| useful, but it's not what CS or software engineering are about.
| Software is not a phenomenon of the physical device. The device
| merely simulates the software.
|
| This "centering" of the computing device is a disease that
| plagues many people.
| kelseyfrog wrote:
| The cybernetic psychology idea of comparing sensed vs. desired
| states maps cleanly onto a PID controller: P reacts to the
| present error, I accumulates past errors, and D anticipates
| future ones.
|
| It's not just a comparator, it's how long I've been off (I), how
| fast I'm drifting (D), and how far I am right now (P).
|
| In this framework, emotional regulation looks like control
| theory. Anxiety isn't just a feeling--it's high D-gain ie: a
| system overreacting to projected errors.
|
| Depression? Low P (blunted response), high I (burden of
| unresolved past errors), and broken D (no expected future
| improvement).
|
| Mania? Cranked P and D, and I disabled.
|
| In addition to personality being setpoints, our perceptions of
| the past, present, and future might just be PID parameters. What
| we call "disorders" are oscillations, deadzones, or gain
| mismatch. But like the article pointed out, it's not really a
| scientific theory unless it's falsifiable.
| p_v_doom wrote:
| If we want to be pedantic both go back to cybernetics and the
| idea of the cybernetic loop.
| rini17 wrote:
| But as recent AI advancement hints at, these states are highly
| manydimensional. And in such spaces our intuitions fall apart.
| Even simple gradient descent works quite differently when you
| have millions of almost-cardinal directions to pick at any
| point, and PID regulation is even more complex.
|
| And even in 3D the plain PID just can't do when the space is
| discrete or when there are large signal delays relative to
| response time of the system. We don't say "oh it's got anxiety"
| lol but we replace it with updated algorithm.
| kelseyfrog wrote:
| That sounds like a great research question, "Given this
| model[1], what's the dimensionality of the space?"
|
| 1. Or more accurately, a model in this family
| hnuser123456 wrote:
| This is tempting, because we have a tight grip of control
| theory. However, there are thousands of things you could
| measure, and picking one at a time to improve could be
| catastrophic. Give a severely depressed person a big bump in
| motivation and nothing else, and it might be their end. We'll
| need to pick a collection of probably at least 100 measurements
| of various personality facets, try different therapies and
| medicines, and see which ones reduce the average error the most
| for different starting conditions. And do this consistently.
| Then, maybe we can say things like "given your scenario,
| treatment X will improve your overall condition the most but
| may leave these facets suboptimal, and treatment Y will provide
| a slightly lower overall improvement but does better for these
| facets you were most concerned about" with actual mathematical
| confidence.
| floxy wrote:
| Someone need to get the author a copy of GEB.
| hybrid_study wrote:
| From a layman's perspective I've yet to find a better
| understanding of psychology than what Steven Pinker posits in
| hist https://en.wikipedia.org/wiki/How_the_Mind_Works (1997), and
| even today it doesn't seem like the basic paradigm (the
| evolutionary constraints) has changed much.
| RetroTechie wrote:
| Interesting concepts there, as applied to psychology. And kudos
| for making it available freely. Definitely worth a read imho!
|
| Skimming through its chapter on AI, made me think of Dave from
| EEVblog fame. In some of his videos he wears a T-shirt saying
| "always give negative feedback!". Which is correct - for those
| who understand electronics (specifically: opamps).
|
| In short: design circuit such that when output is above target,
| circuit works to lower it (voltage, in this context). When below,
| circuit works to raise it. Output stability requires a feedback
| loop constructed to that effect.
|
| There's analogies in many fields of technology (logistics, data
| buffering, manufacturing, etc etc, and yes, thermostats).
|
| I'll leave it there, other sites like Wikipedia (or EEVblog!)
| better explain opamp-related design principles.
|
| From what I've read, current AI systems _appear_ like opamp
| circuitry with no (or poor) feedback loop: a minor swing in input
| causes output to go crazy. Or even positive feedback: same thing,
| but self-reinforcing. Guardrails are not the fix: they just clip
| the output to ehm.. 'politically correct' or whatever. Proper
| fix = better designed feedback loops. So yes, authors of this
| book may definitely be onto something.
| jotux wrote:
| An emotional analogy I've often made, related to EE, is
| automatic gain control. When you experience long periods of
| emotional stability without significant highs or lows, your
| brain applies a form of gain control. This makes the threshold
| for something amazing very low and something awful very high.
| As a result, people can feel overwhelmed by relatively trivial
| issues. Many self-help books, religions, and philosophies
| emphasize appreciating past experiences or considering how
| situations could have been worse. I see this as a way to
| counteract the brain's natural tendency to adjust its gain
| control.
| temp0826 wrote:
| Been a long time since my ee courses (a degree I never
| used...but I digress...). How did that saying go?
|
| Want an oscillator? Design an amplifier. Want an amplifier?
| Design an oscillator.
| dogleash wrote:
| The control loop metaphor was an interesting idea until I started
| worrying halfway through that it would would be operating on
| something with all the same categorical problems as the
| impressionistic style derided the start.
|
| The sensations we tie to urination or breathing have quick cycle
| times making it easy to test the causal loop. Thus confounding
| factors such as a UTI causing a bladder full sensation, or
| nitrogen asphyxiation without feeling suffocation are things we
| understand well.
|
| The "Make Sure You Spend Time with Other People System" is a good
| example for a blog post but it's already a fair bit looser. But
| when you consider they want to investigate things without
| preexisting understanding as well as we understand loneliness it
| smells like sneaking back towards tautologically defined systems
| like "zest for life."
| perching_aix wrote:
| For those looking to skip the yap, an overview of the titular
| "new paradigm" starts a third of the way in, at the section "A
| GOOD WAY NOT TO DIE".
| procaryote wrote:
| Thank you
|
| It's still just fluff though. So the author thinks the mind is
| a control system... Sure, that's a model.
|
| Does it explain observations better? What predictions does this
| model let us make that differ from other models?
|
| The article was needlessly wordy that I might have missed if
| this was hiding somewhere
| perching_aix wrote:
| I felt the same way, so I just stopped reading shortly after
| - can't tell you, sorry.
| vl wrote:
| I put text of the article into ChatGPT 4.5 and used various
| questions to extract relevant and interesting bits and pieces.
| Great use case for LLMs.
| jondlm wrote:
| The topic of this article felt familiar to me. It's similar to
| ideas in IFS: internal family systems. IFS also uses control
| systems to describe our internal landscape.
|
| If the concept of multiplicity (we humans being a system of
| smaller systems) resonates with you, consider reading No Bad
| Parts by Richard C. Schwartz. I've personally found it immensely
| helpful.
| Melatonic wrote:
| Ive heard about this and also some newer works based partially
| on IFS. Looks very interesting
| standardUser wrote:
| The monopoly analogy is brilliant and I appreciate the author's
| focus on the important of language and terms. Plus, the gentle
| but brutal takedown of academic orthodoxy, such as...
|
| > Those divisions are given by the dean, not by nature.
| ebolyen wrote:
| If anyone is interested in a more formal descriptions of these
| control-loops, with more testable mechanisms, check out the
| concept of reward-taxis. Here are two neat papers that I think
| are more closely related than might initially appear:
|
| "Is Human Behavior Just Running and Tumbling?":
| https://osf.io/preprints/psyarxiv/wzvn9_v1 (This used to be a
| blog post, but its down, so here's a essentially identical
| preprint.) A scale-invariant control-loop such as chemotaxis may
| still be the root algorithm we use, just adjusted for a dopamine
| gradient mediated by the prefrontal cortex.
|
| "Give-up-itis: Neuropathology of extremis":
| https://www.sciencedirect.com/science/article/abs/pii/S03069...
| What happens when that dopamine gradient shuts down?
| ikesau wrote:
| what's unclear to me is how you identify the Real Units.
|
| needing-to-breathe-ness is (probably) a gimme, but what are the
| units that will explain which route i take on my walk today? and
| how do you avoid defining units that aren't impressionistic once
| you need to rely on language and testimony to understand your
| research subject's mental state?
|
| my understanding of psychological constructs is that they're
| earnest attempts to try and resolve this problem, even if they've
| led us to the tautological confusion we're in now.
| agos wrote:
| > Imagine how hopeless we would be if we approached medicine this
| way, lumping together both Black Lung and the common cold as
| "coughing diseases", even though the treatment for one of them is
| "bed rest and fluids" and the treatment for the other one is "get
| a different job".
|
| Well, this is definitely happening for some parts of medicine,
| like IBS or many forms of chronic pain.
|
| > If you feel like you're drowning, your Oxygen Governor is like
| "I GIVE THIS A -1000!". When you can breathe again, though, maybe
| you only get the -1000 to go away, and you don't get any
| happiness on top of that. You feel much better than you did
| before, but you don't feel good.
|
| Anecdote but: you absolutely feel good. At least, I did.
| vanderZwan wrote:
| I was going to say the same thing: medicine does this all the
| time. The crucial thing to keep in mind though is that it
| usually has a decent awareness of the difference between when
| it has a decent working model of underlying causes, or if it is
| currently stuck at a _" least-terrible reverse-engineered
| behavioral rule of thumb"_ stage of understanding. The latter
| can still often be better than doing nothing, and what kind of
| doctor would choose to not help their patient just because they
| don't understand _why_ the cure works most of the time?
|
| > _Anecdote but: you absolutely feel good. At least, I did._
|
| I don't know you but glad you're still with us, first of all.
| Anyway, I think the author was trying to say that the feeling
| of relief doesn't typically make one seek out the experience of
| drowning again. Although benign masochism exists of course
| (which is the technical term for the way people can learn to
| enjoy negative emotions and experiences). Which probably would
| be an interesting thing to explore using this cybernetics
| approach.
| kgwxd wrote:
| Flagged? The article appears to have triggered some very
| interesting discussions here. Exactly the stated purpose. Should
| we all go back to talking about how the government and
| billionaires are ruining the world instead?
| vixen99 wrote:
| And line I really liked: 'But really, I like cybernetic
| psychology because it stands a chance of becoming
| overturnable.'
| marviel wrote:
| There are two good books that explore this concept from different
| angles:
|
| "The Emotion Machine", by Marvin Minksy (the AI view)
|
| &
|
| "The Art of Empathy", by Karla McLaren (the Internal / Emotional
| View)
| winddude wrote:
| I'm being pydantic, "So unlike the thermostat in your house,
| which doesn't have to contend with any other control systems, all
| of the governors of the mind have to fight with each other
| constantly.", but what about automated blinds, self tinting
| windows, automatic skylights, humidistats, and humans open and
| closing windows.
|
| It's also important to note that other control systems in the
| body that affect control systems in the mind, eg. endocrine.
| Animats wrote:
| That's a book review. Read the actual book.[1]
|
| Notes:
|
| - Prologue:
|
| _(Behaviorism) ended up being a terrible way to do psychology,
| but it was admirable for being an attempt at describing the whole
| business in terms of a few simple entities and rules. It was
| precise enough to be wrong, rather than vague to the point of
| being unassailable, which has been the rule in most of
| psychology._
|
| - Thermostat:
|
| An intro to control theory, but one which ignores stability.
| Maxwell's original paper, "On Governors", (1868) is still worth
| reading. He didn't just discover electromagnetics, he founded
| control theory. Has the usual problems with applying this to
| emotions, and the author realizes this.
|
| OK, so living things have a lot of feedback control systems. This
| is not a new observation. The biological term is "homeostasis", a
| concept apparently first described in 1849 and named in 1926.
| (There are claims that this concept dates from Aristotle, who
| wrote about "habit", but Aristotle didn't really get feedback
| control. Too early.)
|
| - Motivation:
|
| Pick goals with highest need level, but have some hysteresis to
| avoid toggling between behaviors too fast.
|
| - Conflict and oscillation:
|
| Author discovers oscillation and stability in feedback systems.
|
| - What is going on?
|
| Author tries to derive control theory.
|
| - Interlude
|
| Norbert Wiener and cybernetics, which was peak fascination with
| feedback in the 1950s.
|
| - Artificial intelligence
|
| _" But humans and all other biological intelligences are
| cybernetic minimizers, not reward maximizers. We track multiple
| error signals and try to reduce them to zero. If all our errors
| are at zero -- if you're on the beach in Tahiti, a drink in your
| hand, air and water both the perfect temperature -- we are mostly
| comfortable to lounge around on our chaise. As a result, it's not
| actually clear if it's possible to build a maximizing
| intelligence. The only intelligences that exist are minimizing.
| There has never been a truly intelligent reward maximizer (if
| there had, we would likely all be dead), so there is no proof of
| concept. The main reason to suspect AI is possible is that
| natural intelligence already exists -- us."_
|
| Hm. That's worth some thought. An argument against it is that
| there are clearly people driven by the desire for "more", with no
| visible upper bound.
|
| - Animal welfare
|
| Finally, "consciousness". It speaks well of the author that it
| took this long to bring that up. It's brought up in the context
| of whether animals are conscious, and, if so, which animals.
|
| - Dynamic methods
|
| Failure modes of multiple feedback systems, plus some pop
| psychology.
|
| - Other methods
|
| Much like the previous chapter
|
| - Help wanted
|
| _" If the proposal is more or less right, then this is the start
| of a scientific revolution."_
|
| Not seeing the revolution here. Most of the ideas here have been
| seen before. Did I miss something?
|
| Feedback is important, but the author doesn't seem to have done
| enough of it to have a good understanding.
|
| If you want an intuitive grasp of feedback, play with some op
| amps set up as an analog computer and watch the output on a
| scope. Or find a simulator. If The Analog Thing came with a scope
| (which, at its price point, it should) that would be ideal. Watch
| control loops with feedback and delay stabilize, oscillate, or
| limit. There are browser-based tools which do this, but they
| assume basic electrical engineering knowledge.
|
| [1] https://slimemoldtimemold.com/2025/02/06/the-mind-in-the-
| whe...
| amatic wrote:
| > Not seeing the revolution here. Most of the ideas here have
| been seen before. Did I miss something?
|
| The reviewer is a psychologist, with some interesting opinions
| and criticisms of psychology. My impression is that applying
| control theory to study human behavior should be the
| revolutionary thing, for psychology.
| MarkusQ wrote:
| <snark>It has been described as such every time it's
| tried.</snark>
|
| https://en.wikipedia.org/wiki/Psycho-Cybernetics
| Animats wrote:
| This is not new ground. See _Cybernetics: Or Control_and
| Communication in the Animal and the Machine_ (1948) by
| Norbert Wiener. Wiener wrote a popular version, "The Human
| Use of Human Beings".[2] There's a whole history of
| cybernetics as a field. This Wikipedia article has a good
| summary.[3] The beginnings of neural network work came from
| cybernetics. As with much of philosophy, areas in which
| someone got results split off to become fields of their own.
|
| [1] https://en.wikipedia.org/wiki/Cybernetics:_Or_Control_and
| _Co...
|
| [2]
| https://en.wikipedia.org/wiki/The_Human_Use_of_Human_Beings
|
| [3] https://en.wikipedia.org/wiki/Cybernetics
| amatic wrote:
| > This is not new ground. See Cybernetics
|
| Control theory and cybernetics were supposed to transform
| psychology in a much more dramatic and all-encompassing
| way, as argued by W.T. Powers, for example[1]. In modern
| psychology, the concept of negative feedback control is
| treated like a metaphore, a vague connection between
| machines and living things (with the possible exception of
| the field of motor control) . If psychology would take the
| concept seriously, then most research methods in the field
| would need to be changed. Less null-hypothesis testing,
| more experiments applying disturbances to selected
| variables to see if they are controlled by a participant or
| not. That is the meaning I'm getting from the call to
| revolution.
|
| [1] https://www.iapct.org/wp-
| content/uploads/2022/12/Powers1978....
| Animats wrote:
| Ah. The linked paper goes into that in more detail.
|
| This was a hot idea right after WWII because
| servomechanisms were finally working. In movies of early
| WWII naval gunnery, you see people turning cranks to get
| two arrows on a a dial to match. By late WWII, that's
| become automatic. Anti-aircraft guns are hitting the
| target more of the time. Early war air gunner
| training.[1] Late war air gunner training - the computer
| does the hard part.[2] Never before had that much
| mechanized feedback smarts been applied to tough real-
| world problems.
|
| This sort of thing generated early AI enthusiasm.
| Machines can think! AGI Real Soon Now! Hence the
| "cybernetics" movement. That lasted about a decade. They
| needed another nine orders of magnitude of compute power.
| Psychology picked up on this concept, but didn't do much
| with it.
|
| Looks like it's coming around again.
|
| [1] https://www.youtube.com/watch?v=DWYqu1Il9Ps
|
| [2] https://www.youtube.com/watch?v=mJExsIp4yO8
| pointlessone wrote:
| > An argument against it is that there are clearly people
| driven by the desire for "more", with no visible upper bound.
|
| At the moment it's unknown whether it's a normal function of
| the system or a result of miscalibration/malfunction (assuming
| the theory is correct). Existence of these people is only an
| evidence for the capacity of the system, not a description of
| the nominal function.
| MarkusQ wrote:
| > An argument against it is that there are clearly people
| driven by the desire for "more", with no visible upper bound.
|
| The real problem is that the distinction is meaningless. These
| people could just be described as "minimizing the risk of
| running out of money" (or paperclips, or whatever). Any
| "maximizing system" is isomorphic with a "minimizing system"
| using the inverse of the metric.
| p_v_doom wrote:
| Don't think this is a new paradigm. More of a revival of an old
| one - that of cybernetics. It was incredibly successful at the
| time, to the point where so many of its findings today are
| integral to our world, and we kind of forget where they come from
| and the original principles behind them. I am happy to see this
| revival though, I think it heralds good stuff...
| vanderZwan wrote:
| In short, the brain is a complex dynamic feedback system and
| cybernetics is a really good way to break down complex dynamic
| feedback systems? Seems like an obviously sensible take.
|
| One thing I'm missing here is the idea of _context_.
|
| To give a very concrete example from personal experience: I'm
| colorblind (protanomaly, meaning one of my color cones is
| miscalibrated, which can actually be partially "corrected" for by
| manipulating the _input_ spectrum a.k.a. wearing specifically
| shaded glasses). As a result I 've learned a lot about color
| theory over the years.
|
| One fun thing is that if you look at the history of color
| perception theories, once we realized what the eye is made of we
| started with coming up with ever more sophisticated models for
| each of our cones for a while, which was a great improvement...
| and then someone realized "hey, the way I perceive this shade of
| red _also depends on which color it is surrounded by_ ", and if
| you want to model _that_ accurately then things get really
| complicated (it 's also usually not needed for accurately storing
| and then later reproducing colors). But it explains things like a
| question of "what color is the dress?" taking the internet by
| storm.
|
| So if we look at the "control systems" of the brain, I bet those
| "target values" will be a combination of static baseline plus
| contextual correction from other control systems. And teasing
| that apart will be quite a task.
| AndrewKemendo wrote:
| I'm very excited to see that cybernetic thinking is finding its
| way into a more mainstream audience as people are understanding
| the generality of computing systems in the longest term I have
| not read the book in question yet so I will reserve judgment
| however based on this post it would seem that the majority of its
| claims have already been captured in much more specific detail in
| the 2018 Sutton and Barto reinforcement learning textbook (1).
|
| Specifically, if you look through section 3 chapters 14 and 15,
| they specifically address the psychology and neuroscience
| congruence with the TD(L) algorithms and Bellman reward updating
| to solve the HMM and markov decision process generally.
|
| The newest TD Lambda paper from this year goes even further link:
| https://arxiv.org/abs/2410.14606
|
| Exciting times ahead
|
| 1 -https://archive.org/details/rlbook2018
| YeGoblynQueenne wrote:
| >> Here's the meat of The Mind in the Wheel: the mind is made out
| of units called control systems.
|
| I see someone re-invented Rodney Brooks' Subsumption Architecture
| [1].
|
| It's a good idea but it doesn't work. _Enfin_ , it works, but up
| to a certain point, which is how well robotics work today. Yes
| roombas and industrial robots. No robot maids/butlers and no
| level-5 self-driving cars. Sure doesn't work as a model of a
| human mind. I don't think Brooks even had the temerity to suggest
| it should tbh.
|
| More importantly this just goes to show what happens when
| psychologists and AI scientists stop talking to each other, the
| former because they're lost up their own bums [2] and the latter
| because they've switched focus to creating the next automated
| shaker of the magick money tree.
|
| What happens is that the ... well, _wheel_ , keeps getting
| reinvented, over and over again, and nobody is really trying to
| make, you know, a vehicle. With wheels. That do something else
| than just sit there being wheel-y.
|
| _________________
|
| [1] https://en.wikipedia.org/wiki/Subsumption_architecture
|
| [2] I just mean the reproducibility crisis.
___________________________________________________________________
(page generated 2025-05-16 23:02 UTC)