Taking inspiration from the cephalopods, our definition of intelligence also needs to be revised more frequently. Recently, we've started acknowledging more of the complexity in other animals, like dolphins for example.
I like their color changing behaviors as well[0]. It's is a form of communication we don't fully understand yet.
So in this way, perhaps it's like a morse code to them. They can "talk" by shifting colors and detecting changes in the visual wavelengths... similar to how we vibrate our vocal cords at different frequencies to form words, and detect the changes in pitch with our ear drum.
Now that I've written it out, it seems silly to put "talk" in quotes. They are definitely talking. We just don't know the language. This makes me want to watch Arrival again :)
It's the opposite. The retinas don't have separate receptors for multiple colors, but the iris has a funky "W" shape, which amplifies the (usually undesired) chromatic aberration which is then interpreted by the brain.
'can't process colors the way we do' somehow becomes 'can't process color' .. this I think is roughly what Ned Block called The Harder Problem re consciousness
As a color blind person (as are 20% of half the population at least), I wonder if we can create glasses that mimic this shape of lens and just gradually train our brains to recognise the added spectrum.
(I know about Enchroma, but their marketing disgusts me)
I've thought for a long time that before we focus on communicating with extraterrestrials, it makes more sense to find ways to communicate more fluently with existing beings on our planet, starting with ones we share a common intelligence with (dolphins, raccoons, cephalopods, etc).
We take the animals on this planet for granted, as "lesser than," but what's to say they aren't as intelligent and just express it in different ways? You could say "well dolphins don't build cars or xboxes!" and I would say "dolphins also aren't systematically destroying the planet they live on."
I know we're slowly bridging the gaps of communication with animals (for instance, dogs who can understand verbal language or dolphins who can associate objects with words), but it would be interesting if there were more fluent ways of communication besides verbal. I think it would do us well to master these.
Dolphins not systematically destroying the planet might be more of a lack of ability to instead of desire. Hard to say, really, they're woefully limited in their ability to affect their environment.
IMO it's a sort of "inverted snobbery", where people (rightly) reject the idea that humans are uniquely superior, but they still want to feel special, so they compromise with the idea that humans are uniquely villainous.
I believe the correct approach is to attack the "unique" part: We aren't.
I read a short essay by David Brin, wherein he laments that despite all the years spent on studying dolphin language and intelligence, it's likely that we've been fooling ourselves all along.
IIRC, part of that argument was that if dolphins had a language, surely the word about the dangers of getting caught in seine nets would have spread through the community.
I agree, learning to really communicate with some other species on earth seems like a worthwhile challenge that would push us in interesting directions.
On another thread (about dolphins finding ways to eat octopi without getting suffocated by still-alive-and-fighting disembodied tentacles), some similar thoughts came up.
I have to wonder if one reason for slow progress is that we've bored our subjects. Maybe we should send poets and musicians as ambassadors, not literal-minded scientists.
I read of a researcher who played recorded dolphin sounds for dolphins, and she wanted to teach dolphins to say names she made up for three objects. I don't think the results were very encouraging. But that might be because the dolphins found it pointless.
Dolphins are aware of one another's sonar. So, they would never need to vocalize "Look at that over there." They already know what pod-mates are paying attention to. Why would they verbalize about things that are already obvious on the pod's echolocation radar? They might not have a use for concrete nouns. Dolphins' vocal communications might deal with abstract things like social dynamics, emotions, intuition, and humor.
One other clue: I read of a recording of an upset, irate dolphin. When slowed down significantly, it was clear the dolphin was mimicking it's trainer's reprimands. So, we have to start by admitting that their processing speed is way higher than ours. They're also the only animal with a more highly folded cortex than ours.
These are creatures that can be taught a command that means "innovate" - do something different that you haven't shown me yet today. Maybe innovation is a key aspect of their language. Maybe we need to think way outside the box about what a language could be.
As for dolphins and building ... they don't need protection from sun, rain, and snow; vast swaths of ocean are just the right temperature for them; they can sleep with half their brains at a time while swimming; and they are not interested in stockpiling food for later. What do they stand to gain from building?
What dolphins lack in material culture, they might make up for in social life, poetry, philosophy, and music. Maybe they're looking at us with pity: "Their pod dynamics wouldn't be such a mess if only they didn't waste so much time and effort building stuff..."
I would assume dolphins still die from lack of food, disease, and predation. One(human) would assume they might want to build something to reduce those odds?
You're assuming another intelligence values the same things as humans.
> I would assume dolphins still die from lack of food, disease, and predation.
Hell, we can find humans that don't value such things. Take a look at Christian Scientists, they'll forgo treatment of disease out of fear of rejecting the "divine Principle".
Dolphins use tools to hunt and catch prey. They also engage is pretty complex behaviors to eat.
It took humans roughly a hundred thousand years to find a solution to death from a lack of food through agriculture and we have opposable thumbs.
In terms of food and disease, I would argue they're better off 'seafaring' than settling down in dwellings.
In terms of predation, dolphins are close to being apex predators. Some species and populations have no natural enemies, while in some areas dolphins have to worry about a few of the larger species of sharks and type-2 orcas.
It seems that establishing a built residence would be both an asset and a liability when trying to evade predators. By roaming the open ocean, they're essentially playing a game of "battleship" on a very large board. In other words, there's only a problem when predators and the pod happen to bump into each other.
Having walls may help security, but it also ensures that your enemies know where to find you. Land mammals nest to protect babies before they become mobile. Dolphin calves swim almost immediately, and by drafting in formation with their mothers can keep up with the pod.
I'm curious whether it ever makes sense for carnivores to take up agriculture... both in terms of feasibility and considering that hunting is a cooperative outlet for natural carnivorous aggression.
"... man had always assumed that he was more intelligent than dolphins because he had achieved so much — the wheel, New York, wars and so on — whilst all the dolphins had ever done was muck about in the water having a good time. But conversely, the dolphins had always believed that they were far more intelligent than man — for precisely the same reasons."
> "dolphins also aren't systematically destroying the planet they live on."
You're arguing that dolphins aren't destroying the planet because, what, they decided not to?
Dolphins may or may not be smarter than we give them credit for, but the fact that they don't exhibit one particular behavior is not evidence of anything at all.
I'm not sure that it's valid in any way to say that we share a common intelligence with cephalopods. We share the same planet, and the same fundamentals of cellular biology, but I wouldn't go much further...
Looks like cephalopods have useful, active, and conserved ADAR activity that changes an Adenosine to an Inosine in many RNA transcripts.
ADAR (Adenosine Deaminase acting on RNA) is a protein that recognizes specific RNA sequences (likely, specific structures) [1]. And upon recognition it switches an RNA's 'A' to an 'I' (a non-canonical base pair (RNA is usually made of A, C, U, & G only)). The ribosome compiles a non-canonical 'I' as if it were a 'G'. So this process effectively intercepts some DNA sequences prior to compilation and swaps some (very particular) A's to G's with respect to coding a protein. This process was found by noticing this very consistent 'mistake' between DNA and and expected sequences in high-throughput analysis.
So during the minification of the genetic code prior to compiling, under certain conditions some very specific characters are swapped for out-of-bounds characters that are read-through as an alternative character (if the surrounding code suggests it should). So you can have alternate builds without changing either the source code or the compiler, but instead by turning on/off a bit in the the minifier (expression of ADAR protein).
This process happens a bit in humans and flies, but seems to happen a lot in squid - specifically in the code around compiling neuronal proteins.
Curiously they show that this editing process slows down the evolution around those neural proteins that utilize this mechanism. Because the ADAR protein recognizes the structure of an entire sequence set in order to perform its A->I swap, if there is any mutation in the entire sequence set, no A->I swap occurs. And if no A->I swap occurs, the neural protein likely doesn't work as well, lowering the fitness in any individual that has sequence drift anywhere in that entire recognized region. In organisms that do not have this A-I shift mechanism there is more evolutionary sequence drift in similar proteins than in Cephalopods which utilize the A->I swap. Super cool!
Thanks for the summary. You should be happy to know that because of this I went to your company's website, and realized it was the second time I have done so after reading a good in-depth comment of yours (or perhaps another cofounder).
Does anyone know if RNA recoding happens in other species, or where it started from? The paper mentions that RNA recoding sites slow down genomic evolution, so what species out there have the most conserved genomes + recoding sites?
According to [1, probably paywalled], it's present in both vertebrates and invertebrates, but not plants or fungi, indicating that it's animal specific, but still quite ancient. What the authors have found is not that cephalopods can edit their RNA, but that they do it a lot more than has previously been reported in other animals.
I assiduously avoid studying the brain and behavioral complexity because that stuff is incredibly complicated, but I'm pretty skeptical about the authors' claims in the popular press that this is likely responsible for---or even particularly connected to---cephalopod intelligence or learning. My bet would be that in general, you get intelligence by complicated wiring, but that you don't necessarily need complicated neurons to do that wiring. By analogy, what makes the latest Intel processors so much better than an 8086 has a little bit to do with the 64-bit architecture, but a lot more to do with the radically smaller transistors.
> By analogy, what makes the latest Intel processors so much better than an 8086 has a little bit to do with the 64-bit architecture, but a lot more to do with the radically smaller transistors.
I'm not sure you can draw that distinction. I don't program the transistors, I program the architecture, and the 64-bit architecture is massively better than the 8086 architecture. But the 64-bit architecture would not be possible without the radically smaller transistors, because it requires so many more of them, and that wouldn't be possible (on a realistic die size) without the smaller transistors.
Alternatively, on on 3,000 nm a scaled up 64 bit CPU would actually be slower due to latency issues for many workloads than the 8086. Further a scaled down 8086 would be slower on modern transistors than a 64bit CPU on most but not all workloads so transistors are the most important difference.
That's a good guess, but I think what you're speculating about is the abstracted level on top of the gene design that creates the 'hardware'. Yeah, that'd be cool to figure out the actual blueprint you'd need for neural connections -> conscious, but knowing how that system was designed in the first place may reveal that you're just one on branch that is the tree for the 'architecture'.
> By analogy, what makes the latest Intel processors so much better than an 8086 has a little bit to do with the 64-bit architecture, but a lot more to do with the radically smaller transistors.
It has to do with the fact that there are more of them, and that in turn depends for power usage and speed on their size, but not in an absolute sense. So you could have built a 64 bit processor in the 80's, it's just that it would have been fairly large and a lot slower due to the signal set-up time, and it would use an enormous amount of power.
Pfft! Altering RNA. That's nothing. Once I was in a Los Angeles book store, named Bodhi Tree, where a woman was giving a lecture on altering human DNA using psychic power.
Isn't more that that their RNA is constantly being rewritten than the cephalopods themselves doing writing? What force is responsible for choosing the new RNA?
I would not say it is the animals themselves doing the writing as a force of will, sentience, or consciousness, which is what the headline implies.
The article itself says:
> Their RNA is extensively rewritten, particularly the codes for proteins found in the animals' neurons.
That's a valid question, for which you shouldn't be downvoted, but you should know that it's not anything about it being a force of will by the animals, more that it's a new behavior of how the animal cells are diverse. In other words, maybe the animal is in higher temperatures that trigger a gene to begin executing this particular 'circuit', which causes proteins to form that make slightly more responsive nerve cell ion channels. Where previously we assumed that variation was spread mainly through genes in DNA, this headline shows that there are other ways beyond just DNA-> RNA-> protein.
I may have totally borked this attempted analogy, so hope that helps.
I really don't like their cookbook metaphor. It does nothing to really illuminate the crux of what this allows the creatures to accomplish. As a lay person with a strong interest in such things, I very much wish this article were written with more of an eye towards "Because they can edit their own RNA, they can do thus and such in real terms." instead of hand-wavy metaphors that "it is like the scribe was changed to a master chef and now he can alter recipes."
51 comments
[ 3.4 ms ] story [ 113 ms ] threadI like their color changing behaviors as well[0]. It's is a form of communication we don't fully understand yet.
[0] http://ocean.si.edu/ocean-news/how-octopuses-and-squids-chan...
Their eyes can "see" the wavelengths, due to their shape, but not process colors the way we do.
Now that I've written it out, it seems silly to put "talk" in quotes. They are definitely talking. We just don't know the language. This makes me want to watch Arrival again :)
Are you saying that despite the fact that their retinas are sensitive to multiple wavelengths of light, their brains throw this information away?
https://en.wikipedia.org/wiki/David_Chalmers
As a color blind person (as are 20% of half the population at least), I wonder if we can create glasses that mimic this shape of lens and just gradually train our brains to recognise the added spectrum.
(I know about Enchroma, but their marketing disgusts me)
We take the animals on this planet for granted, as "lesser than," but what's to say they aren't as intelligent and just express it in different ways? You could say "well dolphins don't build cars or xboxes!" and I would say "dolphins also aren't systematically destroying the planet they live on."
I know we're slowly bridging the gaps of communication with animals (for instance, dogs who can understand verbal language or dolphins who can associate objects with words), but it would be interesting if there were more fluent ways of communication besides verbal. I think it would do us well to master these.
I believe the correct approach is to attack the "unique" part: We aren't.
A cephalopod might disagree with that assessment:
https://news.ycombinator.com/item?id=14041682
http://scribol.com/environment/animals-environment/bottlenos...
http://www.bbc.com/earth/story/20150716-dolphins-that-kill-t...
http://www.slate.com/blogs/xx_factor/2009/05/13/dolphins_are...
Can't find it now, though :/
On another thread (about dolphins finding ways to eat octopi without getting suffocated by still-alive-and-fighting disembodied tentacles), some similar thoughts came up.
I have to wonder if one reason for slow progress is that we've bored our subjects. Maybe we should send poets and musicians as ambassadors, not literal-minded scientists.
I read of a researcher who played recorded dolphin sounds for dolphins, and she wanted to teach dolphins to say names she made up for three objects. I don't think the results were very encouraging. But that might be because the dolphins found it pointless.
Dolphins are aware of one another's sonar. So, they would never need to vocalize "Look at that over there." They already know what pod-mates are paying attention to. Why would they verbalize about things that are already obvious on the pod's echolocation radar? They might not have a use for concrete nouns. Dolphins' vocal communications might deal with abstract things like social dynamics, emotions, intuition, and humor.
One other clue: I read of a recording of an upset, irate dolphin. When slowed down significantly, it was clear the dolphin was mimicking it's trainer's reprimands. So, we have to start by admitting that their processing speed is way higher than ours. They're also the only animal with a more highly folded cortex than ours.
These are creatures that can be taught a command that means "innovate" - do something different that you haven't shown me yet today. Maybe innovation is a key aspect of their language. Maybe we need to think way outside the box about what a language could be.
As for dolphins and building ... they don't need protection from sun, rain, and snow; vast swaths of ocean are just the right temperature for them; they can sleep with half their brains at a time while swimming; and they are not interested in stockpiling food for later. What do they stand to gain from building?
What dolphins lack in material culture, they might make up for in social life, poetry, philosophy, and music. Maybe they're looking at us with pity: "Their pod dynamics wouldn't be such a mess if only they didn't waste so much time and effort building stuff..."
I would assume dolphins still die from lack of food, disease, and predation. One(human) would assume they might want to build something to reduce those odds?
> I would assume dolphins still die from lack of food, disease, and predation.
Hell, we can find humans that don't value such things. Take a look at Christian Scientists, they'll forgo treatment of disease out of fear of rejecting the "divine Principle".
Dolphins use tools to hunt and catch prey. They also engage is pretty complex behaviors to eat.
It took humans roughly a hundred thousand years to find a solution to death from a lack of food through agriculture and we have opposable thumbs.
In terms of predation, dolphins are close to being apex predators. Some species and populations have no natural enemies, while in some areas dolphins have to worry about a few of the larger species of sharks and type-2 orcas.
It seems that establishing a built residence would be both an asset and a liability when trying to evade predators. By roaming the open ocean, they're essentially playing a game of "battleship" on a very large board. In other words, there's only a problem when predators and the pod happen to bump into each other.
Having walls may help security, but it also ensures that your enemies know where to find you. Land mammals nest to protect babies before they become mobile. Dolphin calves swim almost immediately, and by drafting in formation with their mothers can keep up with the pod.
I'm curious whether it ever makes sense for carnivores to take up agriculture... both in terms of feasibility and considering that hunting is a cooperative outlet for natural carnivorous aggression.
You're arguing that dolphins aren't destroying the planet because, what, they decided not to?
Dolphins may or may not be smarter than we give them credit for, but the fact that they don't exhibit one particular behavior is not evidence of anything at all.
http://www.cell.com/cell/fulltext/S0092-8674(17)30344-6
ADAR (Adenosine Deaminase acting on RNA) is a protein that recognizes specific RNA sequences (likely, specific structures) [1]. And upon recognition it switches an RNA's 'A' to an 'I' (a non-canonical base pair (RNA is usually made of A, C, U, & G only)). The ribosome compiles a non-canonical 'I' as if it were a 'G'. So this process effectively intercepts some DNA sequences prior to compilation and swaps some (very particular) A's to G's with respect to coding a protein. This process was found by noticing this very consistent 'mistake' between DNA and and expected sequences in high-throughput analysis.
So during the minification of the genetic code prior to compiling, under certain conditions some very specific characters are swapped for out-of-bounds characters that are read-through as an alternative character (if the surrounding code suggests it should). So you can have alternate builds without changing either the source code or the compiler, but instead by turning on/off a bit in the the minifier (expression of ADAR protein).
This process happens a bit in humans and flies, but seems to happen a lot in squid - specifically in the code around compiling neuronal proteins.
Curiously they show that this editing process slows down the evolution around those neural proteins that utilize this mechanism. Because the ADAR protein recognizes the structure of an entire sequence set in order to perform its A->I swap, if there is any mutation in the entire sequence set, no A->I swap occurs. And if no A->I swap occurs, the neural protein likely doesn't work as well, lowering the fitness in any individual that has sequence drift anywhere in that entire recognized region. In organisms that do not have this A-I shift mechanism there is more evolutionary sequence drift in similar proteins than in Cephalopods which utilize the A->I swap. Super cool!
[1] https://en.wikipedia.org/wiki/ADAR1
[1] https://en.wikipedia.org/wiki/RNA-Seq
Does anyone know if RNA recoding happens in other species, or where it started from? The paper mentions that RNA recoding sites slow down genomic evolution, so what species out there have the most conserved genomes + recoding sites?
This paper was freaking fascinating.
I assiduously avoid studying the brain and behavioral complexity because that stuff is incredibly complicated, but I'm pretty skeptical about the authors' claims in the popular press that this is likely responsible for---or even particularly connected to---cephalopod intelligence or learning. My bet would be that in general, you get intelligence by complicated wiring, but that you don't necessarily need complicated neurons to do that wiring. By analogy, what makes the latest Intel processors so much better than an 8086 has a little bit to do with the 64-bit architecture, but a lot more to do with the radically smaller transistors.
[1] http://www.annualreviews.org/doi/pdf/10.1146/annurev-biochem...
I'm not sure you can draw that distinction. I don't program the transistors, I program the architecture, and the 64-bit architecture is massively better than the 8086 architecture. But the 64-bit architecture would not be possible without the radically smaller transistors, because it requires so many more of them, and that wouldn't be possible (on a realistic die size) without the smaller transistors.
It has to do with the fact that there are more of them, and that in turn depends for power usage and speed on their size, but not in an absolute sense. So you could have built a 64 bit processor in the 80's, it's just that it would have been fairly large and a lot slower due to the signal set-up time, and it would use an enormous amount of power.
I would not say it is the animals themselves doing the writing as a force of will, sentience, or consciousness, which is what the headline implies.
The article itself says:
> Their RNA is extensively rewritten, particularly the codes for proteins found in the animals' neurons.
Tune in next week for more headline pet peeves
I may have totally borked this attempted analogy, so hope that helps.