This line surprised me: “The dirty secret is that we don’t even understand the nematode C. Elegans, which only has 302 neurons [in contrast with the nearly 100 billion in the human brain]. We don’t have a complete model of this tiny organism.”
It's very interesting that there are creatures that have their entire behavior determined by just a few hundred neurons. Idid not know this.
Keep in mind that each of those neurons has over 10k genes and hundreds if not thousands are going to be expressed at any given time. Each cell is very complicated.
It's incredibly sad that they don't talk more in high school about DNA methylation and genetic expression in general. Or at least they didn't when I was a high-schooler. It changes the perspective on DNA - from being "the recipe of life, your entire source code" to "list of functions that may or may not be active, depending on other functions, environment and whatever is or is not active in the parent mechanism that does the cell replication (hidden state; c.f. Reflections on Trusting Trust)".
The analogy to a computer is a good one. A computer may have lots of software. Installed, but only some of it is used. Sometimes the software interacts with each other.
C. Elegans is the greatest model organism that has ever been studied. It was specifically selected (by Sydney Brenner) as a model for its properties:
it eats E. Coli, the most common model organism. So it's a model organism that lives on a model organism.
it's transparent so microscopy is really easy.
it always grows the exact same number of cells in the exact same place .
Among those properties:
first multicellular model organism to have its genome sequenced. C. Elegans continues to astound: 35% of its genes are homologous to human genes and the human forms can be used to replace the C. Elegans ones with minimal phenotypic difference.
each terminally differentiated cell (of which there is a fixed number) ends up in a specific location, which were identified in a heroic effort by John Sulston
Each of those cells can be targeted individually, both genetically and optically, allowing for very powerful experiments.
The Nobel Prize went to Brenner, Sulston, and Horvitz for their work.
the full neural connectome of the C. Elegans neural system has been mapped.
RNAi was discovered in C. Elegans.
OpenWorm is a project to build a virtual model of the organism. Having a complete model would require immense computation and storage, but OpenWorm is a reasonable approach.
Dollar for dollar, C. Elegans research pays off better than any massive science project I know of in biology.
first multicellular model organism to have its genome sequenced. C. Elegans continues to astound: 35% of its genes are homologous to human genes and the human forms can be used to replace the C. Elegans ones with minimal phenotypic difference.
Errr, I wonder why this is a surprise. If it respires, produces ATP, etc., why shouldn't is share common eukaryotic building blocks, that we should not be surprised are well conserved?
It does sound cool, but I think I'd rather work on E. Coli than an icky worm ^_^.
Fair criticism, although generally I think for organisms that are that far separated, you'd think the substitutability would be a bit lower given their evolutionary distance.
As for the worm being icky to be honest I hated C. Elegans when I first heard about it, mainly because the primary phenotypes were "dumpy" and "uncoordinated". That said, I feel like the value of the organism exceeds the ick factor for myself and TBH I find lumps of E. Coli pretty unappealing.
Surprising, probably not, but still amazing in my opinion! Our last common ancestor was in the Pre-Cambrian and we still share genes. That information has been preserved for billions of years!
I have a theory that the processing that happens in the brain does not all happen by physical means confined to the brain. I think that neurons either have quantum ability of using neurons in other dimensions or somehow have access to some of the machinery that runs the universe itself.
That's how we get these ideas about the universe and everything in it, because the brain taps into the computer that runs the universe. If you look at really simple organisms with a few neurons doing really complicated things, it makes you think there is something more going on there than just a few neurons firing in a neural network.
Brains must be at least orchestrating those computations, or you couldn't explain all the various ways cognitive functions can be disabled with almost pinpoint accuracy by damaging the brain in specific regions.
I think we know enough about brains of living things to be pretty sure all required computation is being done within the body. Last time I checked, we could pull something resembling a map of the environment out of rat's brain via fMRI.
I think there is a valid counter argument to what you say here. Just because we hear sound come out of the radio does not mean that this sound is created by the radio. Disabling various regions inside the radio will also cause various malfunctions.
> I think we know enough about brains of living things to be pretty sure...
Did you read the article? Seems like we have only a very feeble understanding of the brain.
That's why I wrote that it must be at least orchestrated, in a way that a radio orchestrates the sound by tuning in on correct frequency, amplifying and decoding (digital) or doing FM->AM conversion. I'm not saying that the original comment is for sure 100% wrong.
> Seems like we have only a very feeble understanding of the brain.
Yes and no. Our understanding is very selective; we definitely lack the details on everything, but we know the general picture and details of some parts - which lets us have some confidence about how the thing works. We know for instance that if brain is in fact just an antenna, it's not for anything that goes over EM spectrum, and most likely neither by any particle known to us. Otherwise we'd have figured it out already by accident - people would drop dead underground or close to (shielded) nuclear reactors. On the other hand, we can pull things that look like intermediate computation results out of brains via fMRI.
This does not discard original comment's hypothesis, but lets us classify it as "extremely, extremely unlikely".
> which lets us have some confidence about how the thing works
I don't buy this at all. Without understanding root causes, this is just stamp collecting, and we have no clue how far we are from those root causes. At the very bottom (we think) is quantum field theory, which we only really understand in simple situations (eg. single particle interactions). At the other bottom (and this is perhaps a more orthodox concern) is "chaoplexology", emergent phenomena and all of that.
If a four billion old intelligent alien race handed you one of their computers, do you think you would be able to work it out any time soon? (btw i'm not suggesting this is literally how we evolved, but just to give a different perspective on evolution as we currently conceive it.)
> I don't buy this at all. Without understanding root causes, this is just stamp collecting, and we have no clue how far we are from those root causes. At the very bottom (we think) is quantum field theory, which we only really understand in simple situations (eg. single particle interactions).
No matter how many more levels are down there, everything has to add up to normality. When we discovered relativity, apples didn't suddenly start to fall up. We may discover new ways of looking at the world, or new things hiding behind that n-th fractional digit, but the world will work in the same way the day after such discovery as it worked the day before.
Unless the fundamental laws of the universe have conditionals that branch on cognitive state of human beings, at which point all bets are off and you can assume anything can happen. We have hovewer no evidence that would suggest that hypothesis over the usual one, that the world works the way it works whether we know how or not.
> If a four billion old intelligent alien race handed you one of their computers, do you think you would be able to work it out any time soon?
I would be able to work something out. For starters, that I'm dealing with a computational device. It's hard to imagine what would be the level of technology four billion years old aliens, nor what problems would they be trying to solve - but I'm pretty sure that if you travelled back in time and dropped Isaac Newton a PC with a nuclear battery (and infinite supply of spare parts) he'd figure out some basics of how it works and why it does the thing it does. The very process by which humans do that is called "science".
I guess my point is - we definitely don't know everything, but that doesn't allow us to assume "shit's magic" and every point of view is equally valid. There is observable evidence that shifts probability towards some theories and away from the others.
> or new things hiding behind that n-th fractional digit
Indeed, this is where we disagree: you think brain science is at the n-th fractional digit (for some decent value of n) and I see little evidence of this.
> There is observable evidence that shifts probability towards some theories and away from the others.
This relies on having good priors and is therefore at the subjective side of science. Perhaps I have a less optimistic view of science than you: as a quite fallible human activity, often deceived, lacking in humility, corrupted by money, and basically chasing after whatever is in fashion at the moment. All of this gets us back to the issues discussed in the original article.
> Indeed, this is where we disagree: you think brain science is at the n-th fractional digit (for some decent value of n) and I see little evidence of this.
I didn't mean that in a way physicists at the turn of XIX century meant, i.e. "we've discovered everything that is to be discovered, what remains is just refining results to more fractional digits".
My point is that whether we stumble upon a new scientific revolution behind that nth digit or whether it hits us suddenly out of a blue, it all has to add to normality. The world didn't start to behave any different when we discovered quantum physics.
> Perhaps I have a less optimistic view of science than you: as a quite fallible human activity, often deceived, lacking in humility, corrupted by money, and basically chasing after whatever is in fashion at the moment.
Oh I have a pretty strong view of science, namely that 90% of "soft sciences" developed in the last few decades is utter crap, and so is most of the recent medicine. Physics has it little better, because things are easier to empirically verify; you can't bullshit your way through it so easily because someone will notice, at the latest when they try to build a device based on your research and realize electrons don't give a damn about p-values.
I just think - and that again is my determination based on the things I've read and learned - that we have enough direct, physics-rooted evidence to prefer "brain computes everything" over "brain is just an antenna" as a theory. Probably not enough to reject the latter, but enough to prefer the former.
>but the world will work in the same way the day after such discovery as it worked the day before... Unless the fundamental laws of the universe have conditionals that branch on cognitive state of human beings, at which point all bets are off and you can assume anything can happen
You might enjoy the web serial "Fine Structure" by qntm (Sam Hughes).
I actually downloaded a .mobi version few days ago, it's waiting for me having some free time. I've recently finished Ra, and I am in love with this guy's fiction.
It seems really interesting to consider connections between "computation" (or thought/creativity) and the physics of the universe. This is also related to the question of Platonism, ie. does mathematics exist outside of our own contemplation? For a long time I thought this was ridiculous, but recently I've been seriously wondering if the mere _structure_ of calculation (mathematical discourse as we understand it) could in fact be also driving the shape (what we see) of the universe. To some extent this is obviously true: mathematics does seem to underlie the physics we observe, unreasonably so. Does this suggest that the universe is a giant thought?
> I have a theory that the processing that happens in the brain does not all happen by physical means confined to the brain. I think that neurons either have quantum ability of using neurons in other dimensions or somehow have access to some of the machinery that runs the universe itself.
Bunk.
>That's how we get these ideas about the universe and everything in it, because the brain taps into the computer that runs the universe.
No, we get ideas about the universe by looking at it.
>In fact, there’s actually no such thing as big science; we should really be calling it big engineering.
This is so true. Engineering is undervalued in comparison to science. It's easier to sell multi-billion project to the general public if you label it as science.
Scientists are much more publicly visible than engineers; they are Nobel winners, geniuses, brilliant men single-handedly unraveling mysteries of the universe (or so it seems to the layman).
Engineers generally work in larger teams, don't seek individual fame, don't receive their Nobel prize.
And yet our lives depend upon the quality of engineers' work on a daily basis. Our largest global problems (climate, energy, pollution) are engineering problems. Shouldn't we praise engineers more?
I agree that the engineering involved in these big projects is astonishing and worthy of great praise, but I disagree with calling it "big engineering". Building a bridge is big engineering. Where would the engineers on these big science projects be without the physicists/scientists exact specifications?
Has there ever been an engineering project where the physicists/scientists give exact specifications? They are approximations at best, just enough insight that engineers can find a way to make them useful.
Bottom-up vs. Top-down. Empiricism vs rationalism. There is unfortunately a dramatic lack of flexibility between the two. It's like Liverpool vs Manchester United.
There is also the fact that money doesn't flow as easily towards ideas, immaterial entities, (rationalism) as towards empiric results, tangible entities (empiricism). Data are also much easier to gather and store than ideas to find, which fosters the nsaization of science. An idea is worth nothing, a petabyte of data is worth something.
32 comments
[ 19.2 ms ] story [ 1477 ms ] threadIt's very interesting that there are creatures that have their entire behavior determined by just a few hundred neurons. Idid not know this.
https://en.wikipedia.org/wiki/Caenorhabditis_elegans
it eats E. Coli, the most common model organism. So it's a model organism that lives on a model organism.
it's transparent so microscopy is really easy.
it always grows the exact same number of cells in the exact same place .
Among those properties:
first multicellular model organism to have its genome sequenced. C. Elegans continues to astound: 35% of its genes are homologous to human genes and the human forms can be used to replace the C. Elegans ones with minimal phenotypic difference.
each terminally differentiated cell (of which there is a fixed number) ends up in a specific location, which were identified in a heroic effort by John Sulston
Each of those cells can be targeted individually, both genetically and optically, allowing for very powerful experiments.
The Nobel Prize went to Brenner, Sulston, and Horvitz for their work.
the full neural connectome of the C. Elegans neural system has been mapped.
RNAi was discovered in C. Elegans.
OpenWorm is a project to build a virtual model of the organism. Having a complete model would require immense computation and storage, but OpenWorm is a reasonable approach.
Dollar for dollar, C. Elegans research pays off better than any massive science project I know of in biology.
https://en.wikipedia.org/wiki/Caenorhabditis_elegans
Errr, I wonder why this is a surprise. If it respires, produces ATP, etc., why shouldn't is share common eukaryotic building blocks, that we should not be surprised are well conserved?
It does sound cool, but I think I'd rather work on E. Coli than an icky worm ^_^.
As for the worm being icky to be honest I hated C. Elegans when I first heard about it, mainly because the primary phenotypes were "dumpy" and "uncoordinated". That said, I feel like the value of the organism exceeds the ick factor for myself and TBH I find lumps of E. Coli pretty unappealing.
I say this as somebody who has studied phylogenomics for a decade.
That's how we get these ideas about the universe and everything in it, because the brain taps into the computer that runs the universe. If you look at really simple organisms with a few neurons doing really complicated things, it makes you think there is something more going on there than just a few neurons firing in a neural network.
I think we know enough about brains of living things to be pretty sure all required computation is being done within the body. Last time I checked, we could pull something resembling a map of the environment out of rat's brain via fMRI.
> I think we know enough about brains of living things to be pretty sure...
Did you read the article? Seems like we have only a very feeble understanding of the brain.
> Seems like we have only a very feeble understanding of the brain.
Yes and no. Our understanding is very selective; we definitely lack the details on everything, but we know the general picture and details of some parts - which lets us have some confidence about how the thing works. We know for instance that if brain is in fact just an antenna, it's not for anything that goes over EM spectrum, and most likely neither by any particle known to us. Otherwise we'd have figured it out already by accident - people would drop dead underground or close to (shielded) nuclear reactors. On the other hand, we can pull things that look like intermediate computation results out of brains via fMRI.
This does not discard original comment's hypothesis, but lets us classify it as "extremely, extremely unlikely".
I don't buy this at all. Without understanding root causes, this is just stamp collecting, and we have no clue how far we are from those root causes. At the very bottom (we think) is quantum field theory, which we only really understand in simple situations (eg. single particle interactions). At the other bottom (and this is perhaps a more orthodox concern) is "chaoplexology", emergent phenomena and all of that.
If a four billion old intelligent alien race handed you one of their computers, do you think you would be able to work it out any time soon? (btw i'm not suggesting this is literally how we evolved, but just to give a different perspective on evolution as we currently conceive it.)
No matter how many more levels are down there, everything has to add up to normality. When we discovered relativity, apples didn't suddenly start to fall up. We may discover new ways of looking at the world, or new things hiding behind that n-th fractional digit, but the world will work in the same way the day after such discovery as it worked the day before.
Unless the fundamental laws of the universe have conditionals that branch on cognitive state of human beings, at which point all bets are off and you can assume anything can happen. We have hovewer no evidence that would suggest that hypothesis over the usual one, that the world works the way it works whether we know how or not.
> If a four billion old intelligent alien race handed you one of their computers, do you think you would be able to work it out any time soon?
I would be able to work something out. For starters, that I'm dealing with a computational device. It's hard to imagine what would be the level of technology four billion years old aliens, nor what problems would they be trying to solve - but I'm pretty sure that if you travelled back in time and dropped Isaac Newton a PC with a nuclear battery (and infinite supply of spare parts) he'd figure out some basics of how it works and why it does the thing it does. The very process by which humans do that is called "science".
I guess my point is - we definitely don't know everything, but that doesn't allow us to assume "shit's magic" and every point of view is equally valid. There is observable evidence that shifts probability towards some theories and away from the others.
Indeed, this is where we disagree: you think brain science is at the n-th fractional digit (for some decent value of n) and I see little evidence of this.
> There is observable evidence that shifts probability towards some theories and away from the others.
This relies on having good priors and is therefore at the subjective side of science. Perhaps I have a less optimistic view of science than you: as a quite fallible human activity, often deceived, lacking in humility, corrupted by money, and basically chasing after whatever is in fashion at the moment. All of this gets us back to the issues discussed in the original article.
I didn't mean that in a way physicists at the turn of XIX century meant, i.e. "we've discovered everything that is to be discovered, what remains is just refining results to more fractional digits".
My point is that whether we stumble upon a new scientific revolution behind that nth digit or whether it hits us suddenly out of a blue, it all has to add to normality. The world didn't start to behave any different when we discovered quantum physics.
> Perhaps I have a less optimistic view of science than you: as a quite fallible human activity, often deceived, lacking in humility, corrupted by money, and basically chasing after whatever is in fashion at the moment.
Oh I have a pretty strong view of science, namely that 90% of "soft sciences" developed in the last few decades is utter crap, and so is most of the recent medicine. Physics has it little better, because things are easier to empirically verify; you can't bullshit your way through it so easily because someone will notice, at the latest when they try to build a device based on your research and realize electrons don't give a damn about p-values.
I just think - and that again is my determination based on the things I've read and learned - that we have enough direct, physics-rooted evidence to prefer "brain computes everything" over "brain is just an antenna" as a theory. Probably not enough to reject the latter, but enough to prefer the former.
You might enjoy the web serial "Fine Structure" by qntm (Sam Hughes).
Bunk.
>That's how we get these ideas about the universe and everything in it, because the brain taps into the computer that runs the universe.
No, we get ideas about the universe by looking at it.
This is so true. Engineering is undervalued in comparison to science. It's easier to sell multi-billion project to the general public if you label it as science. Scientists are much more publicly visible than engineers; they are Nobel winners, geniuses, brilliant men single-handedly unraveling mysteries of the universe (or so it seems to the layman). Engineers generally work in larger teams, don't seek individual fame, don't receive their Nobel prize.
And yet our lives depend upon the quality of engineers' work on a daily basis. Our largest global problems (climate, energy, pollution) are engineering problems. Shouldn't we praise engineers more?