How the octopus got its smarts | Hacker News Reader

How the octopus got its smarts

cosmosmagazine.com

Octopus intelligence is essentially one of the multipliera for the drake equation for estimating the probability/ubiquity of intelligent ET life.

We have (as far as we know) one instance of life "seeding," many instances of complex life evolving. Several examples of quasi-intelligent species like dogs, apes, raven and octopi. Only one example of human-level intelligence.

The further they are from us/eachother in the evolutionary tree, the more likely they are to evolve independently regardless of early complex life's evolutionary choices. Octopi are way farther from us than birds are.

Fun stuff!

I have been thinking for a while that if you want to practice communicating with aliens, try octopi. If you can actually have a reasonable conversation with some of them, like you can with a house trained dog, then you are making some kind of progress.

This is almost literally the plot of Arrival :-)

Which is particularly relevant given their similarities to an octopus, albeit a 4th dimensional version of one.

I’d be happy with dolphins or whales.

That makes me wonder - do they have what we would consider reasonable conversations with each other?

Chimp communication is complicated enough that it can probably be called conservation, but I think they're one of the only species for which that is true.

I think most animals lack a wide enough range of responses to have a conversation.

In most cases humans have not yet figured out anywhere near the full range of particular animals’ communication, so this is hard to judge.

For example, prarie dogs https://www.nytimes.com/2017/05/12/magazine/can-prairie-dogs...

I don't think dogs are a great example of an intelligent species just because we can communicate with them reasonably well. They have evolved to be able to participate in human society. With the exception of some breeds of herding dogs, they're probably stupider than (to pick one example) raccoons. But dogs possess a key trait that raccoons lack: they can understand human emotion reasonably well.

My parent's dog makes them indirectly do what she wants => direct communication is potentially overvalued.

Sorry to be “that guy” that points out that the correct pluralisation of octopus is octopuses not octopi, or octopodes if you want to surpass me in pedantry!

https://en.wikipedia.org/w/index.php?title=Octopus#Etymology...

Sorry to be that guy, but I do believe the actual correct etymological term is "octopizzle".

Haha! I'll steal that one!

Fun Fact; ALL pluralizations are 'correct' (via Merriam Webster's: Ask the Editor series) : https://www.youtube.com/watch?v=n4PWP8uL-1o

Ignorami is probably right though!

Fo-shizzle!

it's interesting that you left off one of the most intelligent species: orcas. they are arguably more intelligent than any of the other species you named.

> Only one example of human-level intelligence

this made me think about the relationship between intelligence and technological development. i think they are related but does technological advancement mean a species is really more intelligent than another species with less technological developments? i am not for sure. for example, i have thought about orca intelligence quite a bit. they are severely limited in their technological development simply due to their environment. the aquatic environment naturally prevents the recording of information and development of tools.

I believe that the lack of prehensile appendages is a bigger problem than aquatic environment.

Aquatic is not anti-tool; octopuses can use tools. https://twitter.com/i/moments/935038333654372354

There are other intelligence tests beyond just technology, such as the ability to recognize a mirror for what it is.

orcas and dolphins do use tools. so do fish. and of course octupuses use tools. and i didn't mean that tools are impossible in the ocean. it's just that their use is made difficult.

orcas also easily recognize mirrors. i have watched many animals look into mirrors, and orcas and dolphins are by far the most interested and intrigued and show off many behaviors that easily show they recognize the object in the mirror as themselves.

of course there is more to intelligence than technology, which was what I was getting at. orcas use smooth pebbles as a massage center. they coordinate attacks with special roles, including behavior such as signaling to prey that they are leaving while leaving a silent orca behind to catch the prey as they relax. orcas are one of five known species, including humans, to go through menopause. they have complex social structures and dialects and have even been shown to communicate with dolphins.

> Only one example of human-level intelligence.

It's better to think of this as "human-like intelligence."

Talking about "levels" implies a hierarchy of which there's no proof, and what evidence has been trotted out for it is backed by highly questionable religious and species-centric reasoning over a long time.

There might be an octopus on Proxima Centauri b that is incredibly good at being an octopus, but if it can't build a radio, we can't talk to it. In this context, intelligence is defined in practical terms and we're interested in civilizations that can communicate with us over distance.

Ethics aside, octopuses also seem like a great candidate for "domestication" a la the Russian domestication of the silver fox. Except rather than select for the most docile offspring, selecting for the cleverest. With their relatively short lifespans, one could probably achieve some success in a relatively short amount of time.

That's not domestication, that's uplift :D

Regarding intelligent life in the cosmos:

If [most] life begins at the microscopic scale, wouldn't there be a higher chance of intelligence developing at much smaller sizes than the average human body?

Have we even learned if there's a minimum body size required for sapient intelligence to develop?

We can already see different types of intelligence in comparatively-small creatures like octopuses, but even overlooked organisms as tiny as the ant show forms of intelligence! Social communication, tool usage, agriculture, even domestication of other species (e.g. aphids.)

Well, our massive brain needs a pretty massive support system. There's probably a lower threshold to how efficient you can be at achieving the same level of intelligence.

Intelligence could be distributed across multiple, smaller bodies. Again, taking ants as an example.

At the scope of the Drake equation octopi are still pretty close to us since we both have neurons, so it's a few steps past life "seeding" (good way to phrase it btw). If we saw intelligence in something descended from a pre-neuron fork in the short time that we have existed, that would be pretty good odds.

> While the octopus has a large central brain in its head, it also has a unique network of smaller ‘brains’ within each of its arms. It’s just what these creatures need to coordinate the mind-boggling complexity of eight prehensile arms and hundreds of sensitive suckers, which provide the octopus with the equivalent of opposable thumbs (roboticists have been taking note).

A neurological network of loosely-coupled semi-autonomous subsystems...

Humans actually have a much more simplified system like this for coordinating fine motor movement. Bunches of neural tissue at each vertebrae and a few other strategic sites around the body can fire independently as reflexes. Also stomach has a large amount of intelligence going on. Second largest concentration after the cranium.

The enteric nervous system is interesting, but I'm not sure I'd call it intelligent. For one thing, you can chop big bits out of it with no adverse effects.

You can mostly do the same for parts of the central nervous system also. [0][1] While you are probably right I'm not sure if the argument actually justifies the conclusion. It could be argued that the more advanced the system, the more it can recover from losing parts by implementing redundancies and adaptation mechanisms.

[0] https://www.newstatesman.com/sci-tech/2014/09/how-it-possibl...

[1] https://motherboard.vice.com/en_us/article/9akj4z/what-its-l...

Well you can blow a freaking crowbar through your brain violently enough that it lands almost a hundred feet behind you and yet the lasting effect might not be much worse than a noticeable change in personality

EDIT

"I first noticed the wound upon the head before I alighted from my carriage, the pulsations of the brain being very distinct. The top of the head appeared somewhat like an inverted funnel, as if some wedge-shaped body had passed from below upward. Mr. Gage, during the time I was examining this wound, was relating the manner in which he was injured to the bystanders. I did not believe Mr. Gage's statement at that time, but thought he was deceived. Mr. Gage persisted in saying that the bar went through his head. Mr. G. got up and vomited; the effort of vomiting pressed out about half a teacupful of the brain [through the exit hole at the top of the skull], which fell upon the floor" [0]

[0] Harlow, John Martyn (December 13, 1848). Passage of an Iron Rod Through the Head. Boston Medical & Surgical Journal. 39 (20): 389–93 sourced from wikipedia.

He might have gotten away without any change in personality if he'd avoided the subsequent infection...

Watching someone vomit and a half cup of brain being pressured out of their head to drop on the floor. Very glad I didn't go into medicine.

https://en.wikipedia.org/wiki/Hemispherectomy#Results

Doesn't the eye do some object tracking itself?

Awesome, somewhat speculative and philosophical science book on this topic: Other Minds by Peter Godfrey

Wiki - https://en.m.wikipedia.org/wiki/Other_Minds:_The_Octopus,_th...

Review - https://www.theguardian.com/books/2017/mar/15/other-minds-pe...

Second. The end trails off a bit IMHO but it's an otherwise exceptional book.

"So far, the octopus has revealed three big clues as to how it generates brain complexity: it has multiplied its set of circuit-building protocadherin genes and its network-regulating zinc fingers. It has also unleashed RNA editing to generate more complexity on the fly."

Could someone elaborate on the last sentence please?

RNA editing involves modifying transcripts such that one gene can encode for many separate proteins. The different proteins may only differ at few amino acids but these modifications typically occur in functional domains eg ligand binding sites, pore domains etc. This allows an increase in functional diversity without an increase in genomic complexity.

I don't know why this was down voted so far it was killed so I vouched for it.

It is however a bit obscure. Perhaps this could be rephrased in more common HN terms as "the octopus's genetic mechanism quite aggressively uses epigenetic mechanisms" or "calls genetic 'subroutines' in ways unfamiliar in chordates (like us).

I found it weird that the article called the octopus' brain morphology "weird" given that we really only have one other model (again, that of vertebrates) to follow. Though there are more vertebrates by far, perhaps the octopus makes more sense and we are the centralised weirdos?

I don't know, I quite like the fact that I only need to protect my cranium in order to safeguard by intelligence.

It would be a bummer if a lot of neural processing was happening in my arm and my arm was lopped off.

"Weird" does not mean "bad" — just different [from what we commonly expect]. Anyway, there are also some aspects to the physiology that you might not expect, such as the esophagus running through the center of the executive brain. A bit of crab shell might pierce it, for example.

> "Weird" does not mean "bad"

I tend to agree, but it's interesting that depending on how much the norms of one's prevailing culture emphasize conformity, as they have in the recent past, "weird" can be interpreted as a pejorative.

What do you mean “there are more vertebrates by far”?

By any measure (number of species, number of organisms alive, biomass) invertebrates VASTLY outnumber vertebrates.

Sorry, I meant "vertebrates with brains"

meta: dead comments / shadow bans can be for other reasons than down votes. some new accounts get shadow banned immediately; I assume there's some anti spam stuff working on the backend. This was @jaspwn's first comment, so that's probably what happened.

It's a great use case for vouching.

    This allows an increase in functional diversity without an increase in genomic complexity.

is it not more like the genomic complexity is moved from one place in the DNA to another. Something has to code for the the RNA editing.

It's like compression - the unpacker plus compressed data takes less space than directly writing all modified copies.

It is actually higher complexity but shorter code. More vulnerable to mutations in one sense (damaging codon hits multiple genes), but less in another (shorter length means lower likelihood of damaging a codon).

The 'normal' process is that a cell starts with it's permanent DNA in the nucleus. Part of the DNA [1] is transcribed to a string of RNA, this RNA is translated to a protein. There is an encoding where the sequence of base pairs along the DNA strand unambiguously specifies the sequence of amino acids that form the protein. The proteins then do the work in the cell (eg. the protocadherins and zinc fingers article mentions).

This isn't the only way, and biologists seem to love studying the exceptions, but it is the main pattern found so far. This is the dogma the article refers to: https://en.wikipedia.org/wiki/Central_dogma_of_molecular_bio...

Cephalopods edit their RNA before translation to a protein on a huge scale. Apparently 60% of the RNA sequences in a squid brain study were modified (presumably before translation to a protein) https://phys.org/news/2015-03-squid-enrich-dna-blueprint-pro... So most of the proteins in that squid brain are not in it's genome (although similar ones are).

This is very simplified and I am not a biologist.

[1] Roughly speaking each bit of DNA that does this is a gene: https://en.wikipedia.org/wiki/Genes https://en.wikipedia.org/wiki/Genetic_code

I find this absolutely fascinating. For a long time we thought that humans had a monopoly on abstract thought and problem solving, then we realized apes have it, then pigs, and now octopi. It's weird to think that humans might not be the only truly intelligent creatures on the planet, but it does show to me that "intelligence" isn't this simple black and white system.

I have to say though, this article did kind of make me feel guilty for eating as much octopus as I have.

I think it's useful to think about evolutionary pressures forcing species to be "smart". Brain is an incredibly expensive organ, so it better be more useful than its energy consumption. We don't even quite know why primates have larger brains; we have some theories such as primates, being arboreal, might have evolved to process snakes and spiders visually (this also explains why they have poor olfaction compared to other mammals) [1]. Afaik, and please prove me wrong, we have fewer evidence why great apes have even larger brains and humans have even even larger brains. Might be better communication, group hunting, problem solving (using tools), caring for babies better, or something else, or a combination. It's interesting to think these questions with respect to pigs, octopi, dolphins, cats, dogs etc... Every mammal comes with their own supercomputer pattern matcher that is neocortex, and it's interesting to think that earlier mammals needed more computational power to survive than other animals.

[1]: http://www.sciencemag.org/news/2013/10/did-snakes-help-build...

What's especially interesting about octopus is that while they do communicate with each other, they aren't social in the way we are AFAIK. And, as discussed in the article, relatively short timespans (https://en.wikipedia.org/wiki/Giant_Pacific_octopus), and parents do not raise their young the way we do.

Octopi are apparently a lot more social while high on MDMA:

https://www.npr.org/sections/health-shots/2018/09/20/6487881...

I don't know why you're getting down voted. It's somewhat shocking to me that MDMA would have any kind of even slightly similar effect on such a wildly different animal.

Serotonin as a biological utility has been around since the beginning of multi-cellular life, maybe even sooner!

We organisms tend to share many of the same neurotransmitters. It's always pretty cool to observe similar pharmacological effects to those which humans experience as a result of a certain drug, and then watching a lobster react in a similar way!

their lifespans don't overlap with their offspring's very much

When natural selection doesn’t explain an expensive trait, chances are sexual selection is at work. Social creatures often have bigger brains, but my guess would be sexual selection, mixed with some other traits that made it feasible to fuel such a hungry organ, are behind our inordinately large heads. :)

> When natural selection doesn’t explain an expensive trait, chances are sexual selection is at work.

Sexual selection is an element of natural selection (which is about fitness measured by ability to have offspring that reproduce), so if it is at work, natural selection does, in fact, explain the trait.

Suppose you were born an octopus, with your current level of intelligence.

What would you do? How would you demonstrate it?

Note how difficult it would be to communicate.

I think many animals are a lot smarter than we give them credit for.

I am inclined to agree; we're only semi-recently seeing that animals other than humans and chimps use tools in some fashion, and I don't even know that that by itself is a great measure.

If anything it's evidence of our anthropocentric perspective. Use of tools used to be a proof of our intelligence, until people actually started studying tool use by animals and found some birds, chimps, and octopi all use tools naturally, and seals and dogs can be trained to use certain tools. Rather than declare those animals intelligent we seem to have collectively decided that tool use no longer indicates intelligence...

Octopuses are actually quite social - sending visual and tactile signals at each other. You can also spot them lying (less and more obviously) in those signals.

https://www.livescience.com/53514-octopuses-lead-social-live...

The main difference is that they aren't pack animals like dolpins or most primates, so you get a less condensed sample.

Interesting! I'd always heard that octopi lead a life of solitude. Thanks for sharing.

It's hard to distinguish your current level of intelligence with all the stuff you've learned. Knowing what I know now I'd probably go for tapping prime number sequences on the glass, but it took millenia of humans that were basically as intelligent as me to figure out that prime numbers are interesting.

That is true. I was thinking that morse code or writing would be effective, but my knowledge of these things is not really intelligence but rather things that I learned by participating in human society.

If you were magically transformed into an octopus, communicating with humans would be easy; you could still write.

If you were born an octopus, you wouldn't know how to write, but presumably you would be a lot more familiar with communicating with other octopuses.

Make no mistake, Humans are still far, FAR ahead of other animals. We even have the advantage of thousands of years of backed up knowledge and infrastructure. We still mostly have a monopoly on higher order intelligence.

I am not sure about Dolphins for example. Bats do 3d navigation and echolocation with tiny brains, so there is some reasons dolphins use that many resources to support a larger brain than ours.

We don’t have a great ability to communicate with them, but they also have a very different sensory world.

Size is not the only factor. What about density, specific areas, interconnects? I think edge count is a common measure in graph theory.

Dolphins have more neurons than us in their neocortex speaking to specific areas and the size vs density argument. Overall they have more interconnects than human brains.

The real question is what exactly they are spending all that energy maintaining a lot of processing power, it does not map into tool use but it maps into something.

and now octopi.

I don't think this is recent. I had a relative who raised octopi for sale in his garage in the 1970's.

His most time-consuming task was that he had to play with the octopi every day to keep their brains engaged or they'd get depressed and sad and die.

One of their favorite games was "jar." Where he would put various jars and other vessels in their tanks and they would figure out how to open and close them.

Eventually they would love the game enough that they would unscrew a jar, swim inside, and then screw it closed again with themselves inside. Then they'd look(?) at each other in the other jars and switch jars and basically frolic in the containers.

It was really odd.

That would have been a great video.

I found this which is pretty close to what is described:

https://www.youtube.com/watch?time_continue=209&v=ZU_2CzHnbd...

Thanks

Octopus species are frequently cannibalistic, even while well-fed. How did your uncle manage to keep more than one in a tank?

That's adorable.

I love octopuses and I hope they're given a fair chance to evolve to their full potential, perhaps as the next caretakers of Earth to succeed us if not as the second intelligent species to cohabit this planet with us.

It seems their biggest disadvantage is their very short lifespans? Apparently they get to live for barely 5 years! Worse, their reproductive system also seems to be rigged to make them die of starvation. [0]

[0] https://en.wikipedia.org/wiki/Octopus#Lifespan

In a stunning coincidence, this is from today's Washington Post: 'This is what happens to a shy octopus on ecstasy' https://www.washingtonpost.com/science/2018/09/20/this-is-wh...

Put enough neurons in a room with enough energy and you get intelligence!

Seems to be the recurring assumption. Does anyone in this field have any more insight into this, like is this just a journalistic mis-interpretation or is it really the accepted belief with X, Y and Z to back it up?

i recommend the book squid empire: the rise and fall of the cephalopods.

https://www.amazon.com/Squid-Empire-Rise-Fall-Cephalopods-eb...

https://motherboard.vice.com/en_us/article/nemzqz/researcher...

Reply on news.ycombinator.com