I have no idea what you're trying to say Cartuna.

A cave is no dark room. — Cartuna

Sure, caves are for things that crawl, dark rooms for philosophising. Both the big momma redback and I were surprised when I picked up the pot the other day, and she found that her cave was not as quite as she had supposed. She moved to the less surprising gap under the cupboard, I unsurprisingly put on some gloves. See how well both these behaviours are explained by surprise avoidance?

The problem for the surprise minimisation hypothesis is explaining why Kenosha Kid goes night diving as well as why the spider likes dark crevices.

And it ain't there yet.

Did you read this thread?

I was hungry, for real. I had 0 burgers, 0 hotdogs, 0 eggs, and 0 liters of milk. I'm, for some strange and unfathomable reason, still hungry. — TheMadFool

And I have no idea what you're saying here. Is this an example of absurdity?

That baseline is achieved indeed by maintaining a certain free energy that's needed to live a life. — Cartuna

You mean a certain entropy throughput that sustains its "far from equilibrium" structure. Work must be invested in keeping the structure in a continuous state of repair.

In a darkened room, the brain will still be expending just as much energy in its metabolism.

The status quo can be maintained. If unexpected things happen, the free energy has to increase. Excitement occurs. Information increases, depending on the new situation. — Cartuna

The science says that the energy budget of the brain is surprisingly constant.

That is because focused attention is about shutting stuff down as much as turning stuff up. If faced with a duck-rabbit stimulus, or the ambiguity of a Necker cube, neurons favouring one interpretation get cranked up, and those favouring the alternative are actively suppressed.

One can arrange life to meet as little surprises as possible, like seek sanctuary in a dark room, but surprises are needed in life. — Cartuna

Again, you just absolutely miss the whole point of the Bayesian argument. Life is the ability to surf nature's entropic gradients.

As an energy mechanism it is a Brownian ratchet. And as an information mechanism, it is a Bayesian ratchet.

And there you have it. A unified theory of ratchets. :lol:

:lol:

Sure, caves are for things that crawl, dark rooms for philosophising. — Banno

Dunno. Dark rooms for philosophizing? What's to be found in them? Philosophical surprises? How is a dark room defined for philosophical purposes?

I dunno why I wrote what I wrote. It felt like the most reasonable standpoint given how the dark room problem is set up.

I got no counter. — Kenosha Kid

You certainly failed to offer any counter.

You quoted Friston saying one thing, then you went off on some riff about your everyday definition of surprise as "novelty". I pointed out that your use of novelty is ambiguous. It covers surprises that are both good and bad. Yet you wanted to stick with the idea that all novelty has positive valence ... because dopamine.

Fair summary?

Dark rooms for philosophizing? — Cartuna

I had in mind Descartes' "oven", in which he did his meditating. Perhaps an armchair in a comfy room with a laptop is the modern equivalent, the least surprising thing hereabouts.

In a darkened room, the brain will still be expending just as much energy in its metabolism. — apokrisis

I don't think so. Though it depends on the room. If it's dark and empty, no surprises will be there. You will do nothing at some moment. The free energy minimizes, to sustain the basic needs for life. But the urge for surprises will drive your brain to get in form. Free energy will increase.

Again, you miss the point. That was a "bye". Don't be clingy.

What I said still stands: the notion described in the article is interesting, but not decided.
— Banno
This is a reasonable conclusion. — Caldwell

That's what I do. I sit in my armchair and reach reasonable conclusions. No surprises there.

But surprisingly the sun is shining, so I'm off to look for more spiders under pots.

Point being, despite some protestations to the contrary, it is still not clear how this fits in with thermodynamics and information theory.

I had in mind Descartes' "oven", in which he did his meditating. Perhaps an armchair in a comfy room with a laptop is the modern equivalent, the least surprising thing hereabouts. — Banno

Aah, I get you now. Literal dark rooms to withdraw in. The need for surprises in the mind is overt in there. All hail to the dark room. "Dark room honey, I will follow". I can remember being in love and on the first night we entered the dark space of a small cargo ship we encountered. Free energy was increased in large amounts.

Point being, despite some protestations to the contrary, it is still not clear how this fits in with thermodynamics and information theory. — Banno

You like surprises obviously. New stuff increases your free energy. You brain aquires new forms, accompanied by entropy loss. A rather technical term to say you long for the spider under the pot.

No, that's a red herring of my own making. I think the theory is incoherent for needing to give rise to opposite predictions at different (what the article calls) scales, and doesn't actually relate to the biological or evolutionary processes that must mediate the optimisation. I.e. it's not predictive or descriptive.

Yes, you can pick something (dark caves), apply it one way (our models wouldn't be fit for such an environment, maximising present surprise), and get the desired answer. Or pick something else (going night-diving), apply it in a completely different way (curiosity adds information to our model, minimising future surprise), and get a very different answer. That's not encouraging.

Its not that there's no relationship either between gap analysis of this kind and e.g. fear of the dark, wariness of tigers, etc. which is why it's obviously going to have touchpoints with how nature actually works. But these things are generally mediated in myriad ways antithetical to what the article is talking about. Fear of the dark, of rustling bushes, etc. aren't actually to do with differences in prediction versus environment, real or imagined, but have been selected for by another, quite different optimisation algorithm that isn't minimising that gap but maximising a predominance of features in a way that largely bypasses environmental modelling, preferring a better-safe-than-sorry bias toward interpreting mundanity as surprise.

The red herring being that I made a judgment about _why_ such a theory might be attractive (totalising) and why it wouldn't work (treating many things as one thing). But that was quite a peripheral point.

If biological systems, including ourselves, act so as to minimise surprise, then why don't we crawl into a dark room and stay there? — Banno

Because we'd starve or die of boredom?

Don't be clingy. — Kenosha Kid

Don’t be evasive.

Point being, despite some protestations to the contrary, it is still not clear how this fits in with thermodynamics and information theory. — Banno

Not clear to you or not clear to the neuroscientists that drive Friston’s stellar h-index ranking. Shome difference shurely?

The dark room is a red herring.

What about when surprise becomes confusion?

Yes, you can pick something (dark caves), apply it one way (our models wouldn't be fit for such an environment, maximising present surprise), and get the desired answer. Or pick something else (going night-diving), apply it in a completely different way (curiosity adds information to our model, minimising future surprise), and get a very different answer. That's not encouraging. — Kenosha Kid

Indeed, as I noted earlier, a theory that explains everything, explains nothing.

Not clear to you or not clear to the neuroscientist — apokrisis

Both. I can minimise surprises by picking the garlic that will rot if left much longer. But in stead I came back to this surprisingly entertaining thread. Both of my actions are apparently explained by thermodynamics, and so thermodynamics explains nothing.

That is, again, the theory is interesting but incomplete.

Interestingly this seems to me to be much the same complaint that @Cartuna made in the Consciousness, Mathematics, Fundamental laws and properties thread - the tendency to overreach.

The dark room is a red herring. — Banno

Well yes, that too haha

What about when surprise becomes confusion? — Banno

And there's another problem. We _like_ confusion. The history of science is predicated on the attraction of surprise and confusion. There's nothing sweeter than an observation that doesn't fit the model. Moths to a flame.

Indeed, as I noted earlier, a theory that explains everything, explains nothing. — Banno

Aye.

I don't think so. — Cartuna

Everyone has their opinions but not the receipts it seems.

The free energy minimizes, to sustain the basic needs for life. But the urge for surprises will drive your brain to get in form. Free energy will increase. — Cartuna

You are taking this free energy thing too literally.

The clever thing Friston does is exploit the bridge that exists between the mathematical formalisms of physical entropy and information entropy.

Others do that too. Stan Salthe has his infodynamical general theory of life and evolution. Fundamental physics makes the connection in terms of blackholes and event horizons.

So there is a dual description to be had. Life and mind can be viewed physically as Brownian ratchets, and informationally as Bayesian ratchets - both sides of the coin being able to swim against the general entropic tide.

But you can’t save physical energy for other kinds of metabolic work by going into a dark room or even sitting in an armchair and thinking less. The separation of the two forms of “work” is also a key part of the story.

That is where Pattee’s epistemic cut, or Friston’s Markov blanket, come in. It is also why the brain can seem somewhat like a computer - a device for which the cost of switching the state of its physical gates is a constant, and so drops out of the picture as a drag on the act of computing anything.

A computer can generate nonsense or do something useful. The energy cost in terms of electricity coming out the wall, heat radiated into the environment, is the same. Human brains seem to operate in a similar fashion at times.

The dark room is a red herring. — Banno

For sure. It appears to be one of those fake philosophical posers intended to stoke the careers for those on both sides of the debate. Andy Clark has form here. :wink:

Both. I can minimise surprises by picking the garlic that will rot if left much longer. But in stead I came back to this surprisingly entertaining thread. Both of my actions are apparently explained by thermodynamics, and so thermodynamics explains nothing. — Banno

Aren’t you just playing Buridan’s ass here?

Why continue to pretend you don’t understand that the free energy principle is the constraint that drives adaptive learning? Is it to substantiate the impression that you have as little new to learn about the world whether you are picking your garlic or “engaged” in biological theories of everything?

What a pose.

There's nothing sweeter than an observation that doesn't fit the model. — Kenosha Kid

That's why, in my humble opinion, (religious) miracles are a scientific obsession and yet if you look at what Hume says - a miracle should only be believed if its falsity is even more miraculous - it would seem that scientists are extremely reluctant, even openly hostile, to give due consideration to miracles (basically counterexamples to the laws of nature). I just don't get it.

There's nothing sweeter than an observation that doesn't fit the model. — Kenosha Kid

How many sigma before you accept such exceptions as signal rather than noise?

Let’s get real about the scientific method.

That's why, in my humble opinion, (religious) miracles are a scientific obsession and yet if you look at what Hume says - a miracle should only be believed if its falsity is even more miraculous - it would seem that scientists are extremely reluctant, even openly hostile, to give due consideration to miracles (basically counterexamples to the laws of nature). I just don't get it. — TheMadFool

I think it's a question of the reliability of the evidence. One of the key features of miracles is that they're not reproducible. The Stern-Gerlach experiment is. They were both mysterious at some time, but the latter was reproducible, and therefore credible.

That said, there is an obsession with miracles and spirituality in the sciences at a lower (pop-sci) level, so I guess some people are looking into the kinds of thing you mean. I recall a New Scientist article on a theory about how the parting of the Red Sea might actually have occurred.

https://www.newscientist.com/article/dn19489-how-wind-may-have-parted-the-sea-for-moses/

^ Not sure if the same one. I remember the article being much older. Could be a reprinting, or the date it went digital.

Aren’t you just playing Buridan’s ass here? — apokrisis

Did you mean arse?

That the free energy principle is the constraint that drives adaptive learning is what is in contention.

Let’s get real about the scientific method. — apokrisis

Sound advice. To start, let's not take as fact that which has yet to be demonstrated.

And there's another problem. We _like_ confusion. The history of science is predicated on the attraction of surprise and confusion. There's nothing sweeter than an observation that doesn't fit the model. Moths to a flame. — Kenosha Kid

But it is not beyond the realms of possibility that curiosity itself drops out of the odd and obtuse considerations of thermodynamics - indeed, somehow, I suppose it must be so, if we are physical entities.

The question then is, is this the right calculation, or is there another, or is further nuance needed?

The point being, the point is unsettled.

:sweat: This is better than watching a horror movie.

Half the garlic crop plucked so far. Some rot - about a quarter lost, I think. Set out to dry.

Should have enough to last until mid winter, baring any surprises.

But surprisingly there is a film crew about to arrive. I wonder what surprises they are avoiding. Best go have a wash.

But it is not beyond the realms of possibility that curiosity itself drops out of the odd and obtuse considerations of thermodynamics - indeed, somehow, I suppose it must be so, if we are physical entities. — Banno

Yeah, you can't learn or build jack without it.

That the free energy principle is the constraint that drives adaptive learning is what is in contention. — Banno

No, if put that way. I'm lazy to go back to that article to lift a passage, but there's the comment, I think, by the Theorist regarding the relativistic nature of surprises, and the subjective response to surprises by the agents -- as in, a "surprise" relative to what?