The Future of The Brain (2014)

For the discussion of the sciences. Physics problems, chemistry equations, biology weirdness, it all goes here.

Moderators: gmalivuk, Moderators General, Prelates

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

The Future of The Brain (2014)

Postby aph » Tue Nov 01, 2016 2:36 pm UTC

The title refers to this book: https://www.amazon.com/Future-Brain-Ess ... 069116276X
Subtitle: Essays by the World's Leading Neuroscientists

I found it echoing a lot of the stuff I believe, and wrote about last time, but these people say it in a nicer way. Here is the first paragraph of the preface:
There’s never been a more exciting moment in neuroscience than now. Although the field has existed for two centuries, going back to the days of Phineas Gage and the tamping iron that exploded through his left frontal lobe, progress has in many ways been slow. At present, neuroscience is a collection of facts, still awaiting an overarching theory; if there has been plenty of progress, there is even more that we don’t know. But a confluence of new technologies, many described in this book, may soon change that.

Although many neuroscientists might take for granted that the principal process by which the brain does its work is some form of computation, almost all agree that the most foundational properties of neural computation have yet to be discovered.


There is even a guy using that Feynman's quote to illustrate how building an artificial brain would be a demonstration of understanding neural computation.

The general consensus seems to be that we're all sorta striving toward and waiting for the "big unifying theory", trying out (almost) random stuff and attacking the problem from many different sides. Fun fact: there are apparently about ten thousand independent neuroscience labs around the world (!?!).

----

I feel like neuroscience will be the foundational ground for other sciences - once we really understand how the brain works, we will understand the limitations of our theories and biases introduced by the brain. Math and physics should be influenced to begin with. Psychology will completely transform from a mostly "soft", "pop" science to something more precise and predictive. Then, maybe, economics and sociology will follow. And we'll have artificial brains around us to help or whatever.

Tub
Posts: 373
Joined: Wed Jul 27, 2011 3:13 pm UTC

Re: The Future of The Brain (2014)

Postby Tub » Wed Nov 02, 2016 12:00 am UTC

I keep being told that I'm a pessimistic party pooper, but let me offer a counter-point anyway. I'm as exited as you about any new results, I just don't share your optimism about sudden radical breakthroughs or the transformation of related sciences.

As far as we can tell, brains do not require magic or souls or unicorns or anything special to work. They're just synapses, and those are reasonably well understood.

But the brain as a whole is an incredibly complex network of these synapses, and the complexity appears to be both irreducible (a small brain does qualitatively different things than a large brain) and indivisible (you cannot pick a part and study it independently, because it doesn't make sense independently).

And while we can and will improve our understanding both of brains and of neural algorithms, I'm pessimistic about the idea that our brain will ever be able to fully understand itself. The interesting insights are not in the basics, but in the emergent properties that only happen when billions of neurons are connected in just the right way. Tackling the problem from higher levels (psychology) is just as valid as tackling the basics, and I doubt that both approaches will meet anytime soon. See also https://xkcd.com/1605/ - knowing the basic building blocks does not imply knowing all the consequences. I'd sooner expect quantum mechanics to "solve" chemistry than neuroscience to "solve" psychology, but even the former hasn't happened in almost a century.

And yeah, the artificial brain, for the last 50 years we keep hearing that it's going to be due in a few years. While we have made tremendous advances in AI over the last few years (due to a rise in processing power and massive collections of training data), we're nowhere near the complexity of an actual brain - the most complex AIs or brain simulations I know of are still off by at least 10 orders of magnitude.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Wed Nov 02, 2016 4:14 pm UTC

There are very small brains that do qualitatively similar things to large brains, produce very similar behaviors in possibly similar ways, so I wouldn't agree completely on irreducibly of complexity. Nervous systems of fruit fly larvae or adult fruit flies in many ways resemble mammalian systems, there are topological and functional similarities. But they are still much simpler than mammalian brains. And recently, someone published that they are at 50% of the connectome for the larvae (or something in that order).

Indivisibility is similar. Some parts could be more easily separated and studied then others. The retina is a good example of an accessible part for study.

I do agree with the psychology comment - it is just as valid to approach the problem from a higher level. But I think that low-level and high-level theories could meet pretty soon. Say, inside a decade. If we can understand general principles that explain worms and flies, that is the big start. But it needs to really encompass the long standing questions - perception and sensory integration, motor control and coordination of movement, development and learning...

AI comes as a free side effect of understanding natural nervous systems. Maybe the first one will be as dumb as a fly, but then it will improve with our understanding. Or maybe the two sciences will inform eachother.

User avatar
Copper Bezel
Posts: 2416
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: The Future of The Brain (2014)

Postby Copper Bezel » Thu Nov 03, 2016 1:35 am UTC

Tub wrote:And while we can and will improve our understanding both of brains and of neural algorithms, I'm pessimistic about the idea that our brain will ever be able to fully understand itself.

That's not a reasonable goal. No one brain can fully understand Windows 10. That doesn't make Windows 10 an impossible system for human beings to study.

For any system beyond a certain level of complexity, we have to work in abstractions, specialize in parts, record and organize the information, and accept a limited degree of familiarity with the whole. And automate.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

Tub
Posts: 373
Joined: Wed Jul 27, 2011 3:13 pm UTC

Re: The Future of The Brain (2014)

Postby Tub » Thu Nov 03, 2016 11:33 am UTC

aph wrote:There are very small brains that do qualitatively similar things to large brains

As far as I'm aware, we have small brains that do small brain stuff, like image recognition. And we have our large brains that can also do small brain stuff. But the interesting parts of psychology are those where our large brains do large brain stuff, like self-awareness, conciousness, holding a meaningful conversation etc. I haven't seen those in small brains yet; at least not to a level suitable to increase our psychological understanding of humans.

aph wrote:Indivisibility is similar. Some parts could be more easily separated and studied then others. The retina is a good example of an accessible part for study.

Sure, it's just that the parts that you can study independently are not the parts leading to the interesting phenomena. Human behaviour is not in the retina.

Copper Bezel wrote:That's not a reasonable goal. No one brain can fully understand Windows 10. That doesn't make Windows 10 an impossible system for human beings to study.

Well, for one, operating systems are still orders of magnitude simpler than the human mind. For two, they are created in a structured way out of small, independent building blocks, which makes piecewise study possible. Third, by design, there are no emergent properties in operating systems.

The concept of emergence is poorly understood. Let me give you an example.
* You want to output sounds. You write a sound driver. You have a new feature and new interactions, but no emergence.
* You don't care about sound. You write a video driver. You write a keyboard driver. Suddenly you notice that you have sound. You take the keyboard driver away, the sound is gone. You take the video driver away, the sound is gone. You have no idea where in your code the sound generation happens, but it's now there. You have an emergent property.

Most software is barely more than the sum of its parts. If emergent properties exist, they're usually qualified as bugs. For every behaviour, there's a specific module dedicated to implement that behaviour. (At least in operating systems; video games often embrace emergence.)

In a human brain, the interesting phenomena appear to be emergent.

We do know that systems with emergent properties must be examined at each level of emergence. Otherwise, physics would have a lot less branches. Predicting the high level phenomena from a low level understanding rarely works out. Maybe we can derive temperature and pressure from a simulation of very few gas molecules, and extend that understanding to gases with 10^30 molecules. But in many systems, including the human brain, the emergent properties only appear with sufficient complexity, to the point where you need to examine more of the basic stuff at the same time than we can handle. Predicting human behaviour directly from neurons (or even a full neural map) is infeasible. Finding meaningful levels of emergence between neurons and human behaviour seems difficult (and is the crux of finding that "overarching theory" talked about in the preface).

And that's why I'm much less optimistic about neuroscience radically transforming or even replacing psychology, economics and sociology.

User avatar
Copper Bezel
Posts: 2416
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: The Future of The Brain (2014)

Postby Copper Bezel » Thu Nov 03, 2016 5:31 pm UTC

Well, we're going to get into the "Purity" chart at this rate, but I really don't see how it could ever stop being useful to be able to step back a layer or two of abstraction for specific kinds of probems.

I'm well aware of what emergeence is, and granted that it's not an issue in understanding software. What I'm saying is that I don't think the human brain is a strict limit on what human scientific study can understand. In a sense, that's actually because of emergent properties of society.- like ants, we can sometimes act just as if we understand very complex things when no one individual does.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

Tub
Posts: 373
Joined: Wed Jul 27, 2011 3:13 pm UTC

Re: The Future of The Brain (2014)

Postby Tub » Thu Nov 03, 2016 7:08 pm UTC

I never said that neuroscience isn't useful (I greatly respect anyone who approaches an understanding of the human mind without invoking Freud). I just don't think it'll yield the kind of results that aph envisioned in the last paragraph of the first post.

But we'll see. I'd love to be proven wrong on this.

User avatar
Copper Bezel
Posts: 2416
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: The Future of The Brain (2014)

Postby Copper Bezel » Thu Nov 03, 2016 7:13 pm UTC

I'd be less than certain of the breadth of those predictions, too. But I do honestly believe that psychology is coming up on a revolution comparable to what molecular has done for biology, and that it's going to come from neuroscience.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Thu Nov 03, 2016 8:00 pm UTC

Tub wrote:
aph wrote:There are very small brains that do qualitatively similar things to large brains

As far as I'm aware, we have small brains that do small brain stuff, like image recognition. And we have our large brains that can also do small brain stuff. But the interesting parts of psychology are those where our large brains do large brain stuff, like self-awareness, conciousness, holding a meaningful conversation etc. I haven't seen those in small brains yet; at least not to a level suitable to increase our psychological understanding of humans.

All nervous systems are connected to receptors and muscles, so looks like they're fundamentally related to movement; and we don't yet know exactly how. That is one of the things present in all nervous systems, and if we solve it for tiny worms, we could more easily solve it for humans. Once worms are understood, you get a transfer of theories and hypothesis, of methods of research, computational models..

Sure, it's just that the parts that you can study independently are not the parts leading to the interesting phenomena. Human behaviour is not in the retina.

That just depends on what someone thinks is interesting? There are definitely interesting things in the retina, if you look at the structure of the network of neurons and all that.

Well, for one, operating systems are still orders of magnitude simpler than the human mind. For two, they are created in a structured way out of small, independent building blocks, which makes piecewise study possible. Third, by design, there are no emergent properties in operating systems.

The concept of emergence is poorly understood. Let me give you an example.
* You want to output sounds. You write a sound driver. You have a new feature and new interactions, but no emergence.
* You don't care about sound. You write a video driver. You write a keyboard driver. Suddenly you notice that you have sound. You take the keyboard driver away, the sound is gone. You take the video driver away, the sound is gone. You have no idea where in your code the sound generation happens, but it's now there. You have an emergent property.

I don't think it is the "concept of emergence" that is poorly understood. Particular examples of emergence might be poorly understood, but that is just a matter of proper counting of parts. Like, hardness of ice and liquidity of water are emergent properties of a collection of individual H2O atoms. They don't have 'hardness', but they interact differently on different temperatures. That is one of the examples of emergence that we do understand - because we counted everything we needed to count. And when we count everything relevant, we get just about what we predict.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: The Future of The Brain (2014)

Postby Izawwlgood » Thu Nov 03, 2016 8:27 pm UTC

Oh, this again.

aph wrote:and if we solve it for tiny worms, we could more easily solve it for humans.
Sigh...

aph wrote:Once worms are understood, you get a transfer of theories and hypothesis, of methods of research, computational models..
Sigh...

Why do all your posts on science smack of degrees of purity (as already pointed out) and like a physicist being introduced to a field for the first time?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
doogly
Dr. The Juggernaut of Touching Himself
Posts: 5381
Joined: Mon Oct 23, 2006 2:31 am UTC
Location: Somerville, MA
Contact:

Re: The Future of The Brain (2014)

Postby doogly » Thu Nov 03, 2016 8:31 pm UTC

whoa whoa, no true physicist. homeboy has a firm grasp of newtonian mechanics and then went on a big ol' trip.
LE4dGOLEM: What's a Doug?
Noc: A larval Doogly. They grow the tail and stinger upon reaching adulthood.

Keep waggling your butt brows Brothers.
Or; Is that your eye butthairs?

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Thu Nov 03, 2016 8:47 pm UTC

Izawwlgood wrote:Oh, this again.

Oh, piss off. It is probably because of papers like this one [url]http://www.cell.com/current-biology/pdf/S0960-9822(16)30744-8.pdf
[/url]
The ability of animals to flexibly navigate through complex environments depends on the integration of sensory information with motor commands. The sensory modality most tightly linked to motor control is mechanosensation. Adaptive motor control depends critically on an animal’s ability to respond to mechanical forces generated both within and outside the body. The compact neural circuits of insects provide appealing systems to investigate how mechanical cues guide locomotion in rugged environments. Here, we review our current understanding of mechanosensation in insects and its role in adaptive motor control. We first examine the detection and encoding of mechanical forces by primary mechanoreceptor neurons. We then discuss how central circuits integrate and transform mechanosensory information to guide locomotion. Because most studies in this field have been performed in locusts, cockroaches, crickets, and stick insects, the examples we cite here are drawn mainly from these ‘big insects’. However, we also pay particular attention to the tiny fruit fly, Drosophila, where new tools are creating new opportunities, particularly for understanding central circuits. Our aim is to show how studies of big insects have yielded fundamental insights relevant to mechanosensation in all animals, and also to point out how the Drosophila toolkit can contribute to future progress in understanding mechanosensory processing.


Or this one: http://journals.plos.org/plosone/articl ... ne.0071706
Understanding sensory-motor transformations at the level of genes, neurons and circuits has important implications for neurobiology and medicine. The somatosensory system of Drosophila larva is an especially tractable model system for tackling this problem, due to the small number of neurons (ca. 10,000) in its nervous system and excellent genetic tools for selective manipulation of single neuron types [1,2]. Furthermore, the organization
of somatosensory afferents and motor neuron dendrites in the larval ventral nerve cord not only resembles their organization in adult flies and other insects [3,4], but also the organization of the vertebrate spinal cord [5,6].

User avatar
Copper Bezel
Posts: 2416
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: The Future of The Brain (2014)

Postby Copper Bezel » Thu Nov 03, 2016 10:45 pm UTC

No one is saying those things aren't useful, but they don't quite mean what I think you mean. Particularly not after you assimilate them into your pet theory about how all nervous functions everywhere should be modeled as versions of this particular one you like.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Fri Nov 04, 2016 1:11 am UTC

Maybe? I'm not following your train of associations there.

There are some common principles in NS topology and function across animal species, so possibly also computationally similar processes are happening, particularly in (or to start with) low-level movement circuits. You make an arena, let the larvae loose, record the movements, make a model, compare, improve, repeat. Since movement is not solved in any species, solving it in one species will be a huge step forward. Whose ever pet theory ends up being the closest one.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: The Future of The Brain (2014)

Postby Izawwlgood » Fri Nov 04, 2016 2:46 pm UTC

aph wrote:There are some common principles in NS topology and function across animal species, so possibly also computationally similar processes are happening, particularly in (or to start with) low-level movement circuits. You make an arena, let the larvae loose, record the movements, make a model, compare, improve, repeat. Since movement is not solved in any species, solving it in one species will be a huge step forward. Whose ever pet theory ends up being the closest one.
Yeah, you're still doing it. Woefully misinterpreting and/or extrapolating to fit things into your pet theory.

But, hilariously -
Our aim is to show how studies of big insects have yielded fundamental insights relevant to mechanosensation in all animals, and also to point out how the Drosophila toolkit can contribute to future progress in understanding mechanosensory processing.


Yes! I concur that using model organisms have yielded enormously useful insight into 'the ways things work'! Indeed, my PhD used Drosophila to do just that, indeed, asking numerous questions, and here's the kicker - some even about the behavior of motor neurons.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Fri Nov 04, 2016 5:16 pm UTC

Izawwlgood wrote:But, hilariously - [..]

Yes! I concur that using model organisms have yielded enormously useful insight into 'the ways things work'! Indeed, my PhD used Drosophila to do just that, indeed, asking numerous questions, and here's the kicker - some even about the behavior of motor neurons.

So hilarious. You even asked questions about motor neurons. Rofl.

---

Though, I noticed how these so called "world's leading neuroscientists" hold views you called "disingenuous", "moronic", "narrow minded" and stuff like that when I said them. You noticed? I mean, one is right there in the first paragraph I quoted. Another in the second paragraph I quoted. Feynman is quoted all over, I'm not even going to bother. How is this one, from Antrony Zador, CSHL:
Each year more than thirty thousand neuroscientists gather to share what they have discovered, enough to fill thousands of scientific papers. The rate of progress is staggering. Yet we still don’t really understand how the brain works.
Why? I would argue that the reason we don’t understand how the brain works is that we are missing crucial information. Although we know a great deal about molecules and single neurons, and also about the gross organization of brain areas, our knowledge is scarce between these two extremes, at the level of neural circuits. For this a vital prerequisite is knowing the wiring diagram. The good news is that because of recent advances in technology, it may soon be possible to obtain the wiring diagram, or “connectome,” of the brain at single neuron resolution. Proof that we are not there yet— that we still haven’t “solved” the brain— comes from the fact that we are still apparently quite far from being able to build one. If we really understood the principles behind thought, we could build a machine capable of humanlike thinking.


To "woefully extrapolate", we still don't understand the worms and flies. Though, maybe you do, you did a PhD where you "asked questions".

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: The Future of The Brain (2014)

Postby Izawwlgood » Fri Nov 04, 2016 6:17 pm UTC

aph wrote:So hilarious. You even asked questions about motor neurons. Rofl.
Yes, rofl indeed.

aph wrote:Feynman
WHOA FEYNMAN STFU, FEYNMAN IS QUOTED HOTDAMN SON

aph wrote:To "woefully extrapolate", we still don't understand the worms and flies. Though, maybe you do, you did a PhD where you "asked questions".
I now wonder if you're even on the same page as the criticisms that have been brought up. I don't think you are. Carry on, you always do bring such a show.

Lets back up, because it's always hard to parse out exactly what you're failing to understand int he midst of what exactly you're trying to claim - you're suggesting that we should study insects as model organisms to better understand the behavior of neurons, so we can ultimately... what? Or because neuroscientists as a whole... aren't listening to... something? Are you under the impression that neuroscientists the world over are packing it in because they're convinced they've done it all? Or the opposite - that neuroscientists the world over are packing it in because they're convinced they know nothing at all?

Or are you just excited that people are still doing neuroscience, and think that your pet theory is somehow vindicated by something written here?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Fri Nov 04, 2016 6:29 pm UTC

No, I'm just saying you should go suck a bag of nuts. The topic is the book in the title, and "the future of the brain" theories. I don't care much about your opinion about my opinions about the nervous system or theories or feckin feynman.

User avatar
doogly
Dr. The Juggernaut of Touching Himself
Posts: 5381
Joined: Mon Oct 23, 2006 2:31 am UTC
Location: Somerville, MA
Contact:

Re: The Future of The Brain (2014)

Postby doogly » Fri Nov 04, 2016 6:35 pm UTC

You don't care about the nervous system? But you care about the brain?

I got some news for you...
LE4dGOLEM: What's a Doug?
Noc: A larval Doogly. They grow the tail and stinger upon reaching adulthood.

Keep waggling your butt brows Brothers.
Or; Is that your eye butthairs?

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: The Future of The Brain (2014)

Postby Izawwlgood » Fri Nov 04, 2016 6:41 pm UTC

Literally your first post in this thread is you fanboy squeeing over Feynman being quoted.

Again though, literally no one is saying that the entirety of the brain is understood, but you are again gushing about how everyone OBVIOUSLY agrees with your pet theory, despite having provided, yet again, literally zero evidence of it beyond a couple papers that sans context and sans you really saying anything other than 'Look look, quotes that have words I like!', and the typified inability to respond to any criticism or requests for clarification of what you're going on about.

But yes, we 'don't understand worms or flies', except we do, or something, and ergo, 'control theory'.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Fri Nov 04, 2016 6:51 pm UTC

Ok, maybe I do like Feynman a bit too much.

User avatar
doogly
Dr. The Juggernaut of Touching Himself
Posts: 5381
Joined: Mon Oct 23, 2006 2:31 am UTC
Location: Somerville, MA
Contact:

Re: The Future of The Brain (2014)

Postby doogly » Fri Nov 04, 2016 8:31 pm UTC

Well, that's unlikely.
LE4dGOLEM: What's a Doug?
Noc: A larval Doogly. They grow the tail and stinger upon reaching adulthood.

Keep waggling your butt brows Brothers.
Or; Is that your eye butthairs?

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Fri Nov 04, 2016 9:06 pm UTC

Well, either way, Feynman is being quoted because even in neuroscience, or in science in general, building computational / mathematical models of processes is the linchpin of understanding. Or at least, that is what a whole bunch of computational neuroscientist and AI people think.

User avatar
doogly
Dr. The Juggernaut of Touching Himself
Posts: 5381
Joined: Mon Oct 23, 2006 2:31 am UTC
Location: Somerville, MA
Contact:

Re: The Future of The Brain (2014)

Postby doogly » Sat Nov 05, 2016 3:17 am UTC

...yes. models are good.
is that the extent of the claim you are making?
LE4dGOLEM: What's a Doug?
Noc: A larval Doogly. They grow the tail and stinger upon reaching adulthood.

Keep waggling your butt brows Brothers.
Or; Is that your eye butthairs?

User avatar
Copper Bezel
Posts: 2416
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: The Future of The Brain (2014)

Postby Copper Bezel » Sat Nov 05, 2016 8:38 am UTC

aph, the idea from Feynman is a truism that's accurate and a useful reminder where it's applicable, and this is one of the places where it's applicable. I don't think anyone's taking any issue with that by itself. I also have not the slightest doubt that studies of simpler nervous systems will have applications to the study of more complex ones, because largely, that is why people study them, because that's what "model organism" means.

But in the past, your threads on the brain and neuroscience have tended to misrepresent the field as cluelessly waving their hands in the air for want of a unifying model, which of course you feel you have, and if they'd just talk to you, we'd solve everything in a month. But emergence is a real thing, complex systems are complex to model, and no system is beholden to following the dictates of a simple nesting system of logical chunks like you've wanted to make nervous systems work in the past.

Like, look at where we are with ontogeny. To my understanding - and this is all in my rough understanding of ontogeny research - people have identified damn near all of the chemical signals involved and the processes by which the signal gradients form. We understand invagination and protrusion, differentiation, apoptosis, all the tools the development process has to work with. We can identify the genes that trigger these things. For very simple organisms with fully determined development, we've tracked the determination and "fate" of every cell in the developing embryo, where you can count positions on the blastula, point to a cell, and say exactly how many times it's going to divide, what it's gong to grow up to be, and where. That's exciting and important work. But it doesn't allow us to predict the entire development process of a given organism by looking at a genome sequence. We'd still just have to grow the thing. And if we had a complete, predictive, mathematical model of the whole process and infinite computing resources, we could someday simulate it - by building the damn thing in a computer and growing that instead.

You didn't start this thread with your own choice of super-reductionist "unified theory", and you rarely do start there, but this

All nervous systems are connected to receptors and muscles, so looks like they're fundamentally related to movement; and we don't yet know exactly how.

strongly smacks of it, and I think we're going to be looking at yet another aph PCT thread very quickly if this one continues. They're a fairly frustrating experience that I'm not personally eager to repeat. And I can't imagine what they're like for Izawwlgood, since he actually knows what we're talking about.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sat Nov 05, 2016 12:36 pm UTC

The idea that principles of motor control are preserved across species is more general then specifying which theory of motor control explains it all. There is no smack in that. Of course, I believe it is this specific theory I might have mentioned once or twice, but not yet here, and not in the last thread either. No one likes my evangelical work. It is fine, I'll leave it out.

Copper Bezel wrote:in the past, your threads on the brain and neuroscience have tended to misrepresent the field as cluelessly waving their hands in the air for want of a unifying model

That is the main topic, it is discussed in the book in the title - what is the present state of neuroscience and what are the possible futures. I like your comparison with ontogeny - we do have a huge amount of data in molecular and cell level in neurobiology, which I acknowledged multiple times. Though, if you equate neurobiology with neuroscience I can understand the frustration with someone saying "it is all a mystery".

On the other hand, neuroscience is a wide field, and in the context of cognitive neuroscience, neurorobotics, behavioral neuroscience, neuroethology... or even neuro-law or neuro-marketing (which kinda smell of pseudoscience), it is not necessarily clueless, but it is definitely lacking unifying principles. A lot of researchers and theorists have their own candidates for unifying principles - network dynamics, emergence from complexity, prediction of temporal patterns, variations of cybernetic ideas.. It is lacking proven unifying principles, to be more accurate. There are plenty of proposed ones.

The biggest problem, it seems to me, is relating circuit function to actual behavior of the animal in the real world - and that has been hard to study, because you can't really record both at the same time. And there many new tools, they are getting smaller in size and improving in temporal and spatial resolution. Ergo, optimism about the future of brain theories.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: The Future of The Brain (2014)

Postby Izawwlgood » Sat Nov 05, 2016 12:58 pm UTC

Well, you tried CB.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sat Nov 05, 2016 2:12 pm UTC

Copper Bezel wrote: I can't imagine what they're like for Izawwlgood, since he actually knows what we're talking about.

I imagine he is suffering.
It is fine, my child. You tried.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26086
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: The Future of The Brain (2014)

Postby gmalivuk » Sat Nov 05, 2016 3:33 pm UTC

aph wrote:No, I'm just saying you should go suck a bag of nuts.

If this is how you're going to react when your ideas are criticized, perhaps this isn't the forum for you?
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sat Nov 05, 2016 3:56 pm UTC

gmalivuk wrote:
aph wrote:No, I'm just saying you should go suck a bag of nuts.

If this is how you're going to react when your ideas are criticized, perhaps this isn't the forum for you?

Oh, the drama... You could take that as a literal bag of nuts, though.

Also, you haven't noticed how this book talks about neuroscience not understanding fundamental properties of computation in the brain, something I expressed last time, and was informed that my opinion was 'moronic'? I was quite happy to find a whole book discussing how much we don't know about nervous systems.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: The Future of The Brain (2014)

Postby Izawwlgood » Sat Nov 05, 2016 7:48 pm UTC

And IIRC, we discussed how despite finding a WHOLE ENTIRE BOOK that didn't say we know nothing about the brain, and pointing you towards whole fields of study on brains and specific subfields dealing with brains, you somehow managed to simultaneously admit that you had a lot to learn, that you didn't trust what was known because you were sure what was known was basically nil, and that your theory was the correct one.

EDIT: Just took a few minutes to peruse the last three threads you started on this very topic. Enormously frustrating to see you're still doing the same thing of quibbling over levels of understanding and insisting that the answer is your pet theory.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sat Nov 05, 2016 8:50 pm UTC

Izawwlgood wrote:And IIRC, we discussed how despite finding a WHOLE ENTIRE BOOK that didn't say we know nothing about the brain, and pointing you towards whole fields of study on brains, you somehow managed to simultaneously admit that you had a lot to learn, that you didn't trust what was known because you were sure what was known was basically nil, and that your theory was the correct one.

EDIT: Just took a few minutes to peruse the last three threads you started on this very topic. Enormously frustrating to see you're still doing the same thing of quibbling over levels of understanding and insisting that the answer is your pet theory.

Well, that tone seems a bit less condescending than what you sported so far on this thread.

Of course I have a lot to learn. Who doesn't? There are a ton of theories about the brain, on various levels of study, tons of great published papers, books, etc. It is good to learn about all the approaches, but it is not necessary to believe that the models they use are correct. It is pretty obvious that some of the models are self-contradictory, or contradictory to models at other levels, but there are no better ones, so you just go with what is most useful. As in mental illnesses, for example.
Or you can learn about all the approaches to motor control, and just the sheer number of approaches tells you that the story is far from certain, but it is still very useful to get acquainted with many of them.

And while I do believe that my pet theory is correct, I also realize that that belief is an extrapolation from about a dozen experiments and definitely needs more confirmation. Also, I have no qualms about 'having a pet theory' and believing it is correct, that is the motivation for doing further experiments. I suppose everyone, or at least a lot of people in science, have their pet theory and do research to confirm it. They do things out of conviction that something they are doing will be important. And they also realize they could be wrong.

Don't believe it is possible to 'have no theory' and just collect data. That in itself is a theory and a philosophical position.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: The Future of The Brain (2014)

Postby Izawwlgood » Sun Nov 06, 2016 5:27 am UTC

I think this is the *5th* thread you've started on the same topic, and they all go the same way. You should maybe reread them for a hint at why I at least would be frustrated and not particularly interested in being patient and polite. Don't tone police when frankly at this point it's easy to see why I, and others, would be tired of repeating the same conversation with you wherein you exhibit the exact same refusal to listen or learn. I feel like there's a control theory joke to be had here regarding your baseline state, but frankly, I don't think you'd recognize it as the criticism that it would be.

Again, it is spectacularly arrogant of you to keep admitting you don't know much about the field whilst simultaneously claiming that the field probably doesn't know what it's doing and should listen to your brilliant theory that will fix everything and move it all forward. You have a pet theory. Great. Neato. Congrats. Welcome to where many scientists got their start. Time to do some work advancing it instead of insisting your theory is right and everyone who works in the field is wrong.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
The Great Hippo
Swans ARE SHARP
Posts: 6980
Joined: Fri Dec 14, 2007 4:43 am UTC
Location: behind you

Re: The Future of The Brain (2014)

Postby The Great Hippo » Sun Nov 06, 2016 8:44 am UTC

If we're going to squee over Feynman, we might as well squee over Feynman being relevant: Feynman on explaining how magnets work.

Basically, a physicist can't explain to you why magnets attract/repel other magnets unless you're willing to take the time to also become a physicist. And the functionality of the brain is extraordinarily more complex than the functionality of a magnet.

When a neuroscientist talks about things like discovering the "foundational properties of neural computation", they're doing what Feynman described with the rubber band example: Providing crude, flawed metaphors that fail to completely encapsulate what's actually going on. It's meant to whet your curiosity -- get you excited -- and maybe convince you to fund them (because they think they're onto something). But reading articles about neuroscience doesn't really teach you much about neuroscience; only neuroscience teaches you about neuroscience.

A meteorologist tells you that the greenhouse effect is like a magical blanket that lets heat in but doesn't let heat out. Do you now understand the greenhouse effect? Could you come up with a theory about this magical blanket that meteorologists would find interesting -- perhaps even revolutionary? Do you think you can see the future of meteorology without -- well, being a meteorologist?

There's this old urban legend about a truck that gets stuck underneath a low-hanging bridge; everybody is trying to figure out how to get it out. Some little girl comes up with the solution: Let the air out of the tires. It's a cute story, because it's fun to imagine dozens of adults -- police officers, firemen, government officials -- people trained to solve problems like these -- all scratching their chins... when a little girl walks in with a fresh set of eyes and points at the answer sitting there right in front of them.

But science is not a truck stuck under a bridge. To use some crude, flawed metaphors of my own: Science is a nuclear fusion reactor with a subtle engineering flaw that causes it to overheat. It's fifty thousand lines of code with an 'off-by-one' error hidden somewhere. It's figuring out a way to send a probe 4 billion miles through space and land it safely on a comet with a max speed of 84,000 miles per hour. It's hard, it's tedious, it's complicated -- and it's not accomplished by someone walking in and letting the air out of the tires.

What I'm saying is this: Yes, there is some non-zero chance you've struck upon the future regarding neuroscience. But unless you're a neuroscientist... it's far more likely you're just the guy saying we'll soon be building planes out of flight recorders.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sun Nov 06, 2016 11:02 am UTC

Izawwlgood wrote:I think this is the *5th* thread you've started on the same topic, and they all go the same way. You should maybe reread them for a hint at why I at least would be frustrated and not particularly interested in being patient and polite. Don't tone police when frankly at this point it's easy to see why I, and others, would be tired of repeating the same conversation with you wherein you exhibit the exact same refusal to listen or learn. I feel like there's a control theory joke to be had here regarding your baseline state, but frankly, I don't think you'd recognize it as the criticism that it would be.

Again, it is spectacularly arrogant of you to keep admitting you don't know much about the field whilst simultaneously claiming that the field probably doesn't know what it's doing and should listen to your brilliant theory that will fix everything and move it all forward. You have a pet theory. Great. Neato. Congrats. Welcome to where many scientists got their start. Time to do some work advancing it instead of insisting your theory is right and everyone who works in the field is wrong.


Frustration, goddess, sing the frustration of Izawwlgood, a man, Science incarnate!

I did reread a couple of the threads, and this one and the last one I wrote on (about laws of nature) are not about control theory. I noticed you enjoy a bit of a rockstar status up in these parts - when I said I wasn't interested in your opinion, two mods jumped in saying that I should leave if I wasn't interested in science. Pretty cool!

I didn't say I don't know much about the field, but that I still have a lot to learn. I did spend five years studying a related field, and my master thesis was in computational neuroscience. I found a position, I am continuing with the work and not insisting or trying to verbally prove anything about control theory anymore (as you could have noticed). It is cool, you're not into that stuff, apparently no one else on this forum is. I did notice a couple of labs publishing papers using the theory or closely related theories (on c. elegans and drosophila larvae; also rats; also psychotherapy).

So, back to the topic, what are your opinions about the future of the field? When will we have the grand unified theory? You have a pet theory? What is your stance on the link between psychology and neuroscience?

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sun Nov 06, 2016 11:19 am UTC

The Great Hippo wrote:Basically, a physicist can't explain to you why magnets attract/repel other magnets unless you're willing to take the time to also become a physicist. And the functionality of the brain is extraordinarily more complex than the functionality of a magnet.

When a neuroscientist talks about things like discovering the "foundational properties of neural computation", they're doing what Feynman described with the rubber band example: Providing crude, flawed metaphors that fail to completely encapsulate what's actually going on.

There is one crucial difference, though. Physicists have very precise and proven models of how magnets work. In neuroscience, there are many models of neural computation. Starting with neural code - there are simple on/off spikes, timing between spikes, population coding, rate coding...

I definitely agree that there are many crackpots advancing their silly theories (some may even have tenure :/ ). There is that famous dilemma from Hamming, paraphrased: ignore the crackpots outside of your field (where the new ideas come from) and miss the next big idea, or listen to the crackpots and waste all your time.

User avatar
The Great Hippo
Swans ARE SHARP
Posts: 6980
Joined: Fri Dec 14, 2007 4:43 am UTC
Location: behind you

Re: The Future of The Brain (2014)

Postby The Great Hippo » Sun Nov 06, 2016 11:45 am UTC

aph wrote:There is one crucial difference, though. Physicists have very precise and proven models of how magnets work. In neuroscience, there are many models of neural computation. Starting with neural code - there are simple on/off spikes, timing between spikes, population coding, rate coding...
Right; a physicist can model a magnet with math, because -- at their core -- magnets are functionally pretty simple. But a physicist still can't explain to a non-physicist what's up with magnets. Unless you're a physicist, you're not going to understand how magnets work -- and because of that, you're probably never going to tell us anything about magnets that we didn't already know.

And the human brain? That's more complex than a magnet by several orders of magnitude! The chances of you telling us something about human brains that neuroscientists don't already know is breathtakingly remote!

(Of course, as you learn more about the human brain, those chances get a little better -- because you're starting to become a neuroscientist! But right now, it sounds like those chances are so small that even talking about them gives them more relevance than they deserve)
aph wrote:There is that famous dilemma from Hamming, paraphrased: ignore the crackpots outside of your field (where the new ideas come from) and miss the next big idea, or listen to the crackpots and waste all your time.
New ideas do not come from crackpots outside the field. New ideas come from new people entering the field, learning it, and building on top of it.

Innovation within a field comes from within that field -- from the people who have put the work in.

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sun Nov 06, 2016 2:34 pm UTC

The Great Hippo wrote:And the human brain? That's more complex than a magnet by several orders of magnitude! The chances of you telling us something about human brains that neuroscientists don't already know is breathtakingly remote!

Like the chances of a monkey typing out the Odyssey. Sure. That is why the chances that someone has a new discovery isn't the relevant variable. And I'm not talking about me telling something new. The likelihood of a new discovery is very small, even for long-time neuroscientists. The relevant variable is just how well their prediction fits actual experimental data. I don't know if it would even be possible to calculate the likelihood that, say, a patent office clerk would discover something new about space, time and gravity.

aph wrote:New ideas do not come from crackpots outside the field. New ideas come from new people entering the field, learning it, and building on top of it.

Innovation within a field comes from within that field -- from the people who have put the work in.

This was from Hamming's The Art of Doing Science and Engineering. According to him, many innovations come from outside fields. Though, you're probably right, innovations usually come from within the field.

Neuroscience is still a young field. There are very few neuroscience-specific undergrad programs, and a lot of people come in for PhDs or postdocs from other fields. There are people from many different backgrounds. Jeff Hawkins was an electrical engineer before turning to brains, to name a famous outsider. Other than various biologists, there are mathematicians, physicists, computer programmers, psychologists, roboticists.. with significant contributions. I definitely wouldn't be surprised if some new machine learning algorithm devised by a computer programmer turns out to closely resemble a biological learning process. Programmers are just better trained in thinking about algorithms.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26086
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: The Future of The Brain (2014)

Postby gmalivuk » Sun Nov 06, 2016 2:39 pm UTC

aph wrote:I noticed you enjoy a bit of a rockstar status up in these parts - when I said I wasn't interested in your opinion, two mods jumped in saying that I should leave if I wasn't interested in science. Pretty cool!

1) No one thinks of Izawwlgood as any kind of rock star or rockstar analogue in "these parts". We simply recognize that, on this topic, he is more well-informed than the vast majority of other forumites. If he repeatedly started physics threads with the same attitude and arrogance you exhibit here, he'd face the same sort of disdain as what you're facing here.
2) Only one mod has ever posted in this thread.
3) My problem wasn't your disinterest in science, it was your petulant little response to having your ideas questioned.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

aph
Posts: 296
Joined: Tue Nov 12, 2013 7:48 am UTC

Re: The Future of The Brain (2014)

Postby aph » Sun Nov 06, 2016 2:53 pm UTC

gmalivuk wrote:No one thinks of Izawwlgood as any kind of rock star or rockstar analogue in "these parts"

Oh, I'm trolling a bit, it was on the previous topic, this post. See how if I don't care to find out what he thinks, I don't care about 'actual science'. I love that line.


Return to “Science”

Who is online

Users browsing this forum: Google [Bot] and 8 guests