The book is interesting and goes into how humans need to set up experiments properly to actually test non-human animals in ways that make sense (rather than just in some biased human way).
One quick example was testing tool use, the original experimenters left branches on the ground for the monkeys to use, but the monkeys can’t pick stuff up that’s flat on the ground since they’re normally in trees (their hands don’t have thumbs that move that way). When he redid the experiment with the tool raised they were able to grab and use it.
A dog I live with, not technically mine, can count to at least three. She hates waiting in her harness before we go on walks, and has learned that I am ready to go after taking three poop bags. If I only pull two bags, she won’t come downstairs and wait by the door. If I pull the third bag, she immediately walks down the stairs. I noticed it the other day because I had two bags on me; so, she only heard me pull one bag and refused to walk downstairs. Then I pulled a second one, still nothing. Finally, I pulled the third one and she immediately trotted on down. Now that I’m aware of it, I’ve been paying attention and yep... she only ever comes down after hearing a third bag. I’ve been pretty shocked by it to be honest (in a good way)...
Surely it implies an ability to count incrementally, which is all OP was implying? The dog does not have to abstract the concept of '3' to be able to count- it's still very interesting.
> Surely it implies an ability to count incrementally
Not necessarily. It could be just ability observe scale not individual numbers. i.e. 3 bags look big enough amount of bags for a walk not that there are 3 bags. Would the dog not come when 4 bags are taken? What if 2 bigger bags are taken? It's kinda hard to deduce anything from this anecdote other than the dog seems to understand a pattern.
Perhaps I was reading too far into OP's anecdote then, but it seemed there was a delay between bag pulls, and that only after the third bag the dog responded. I also was under the impression that this was occuring out of direct line of sight (as the dog was coming down the stairs following the third bag) and so suggested the dog was interpreting the pulls incrementally rather than picking up on the handful of three bags through direct observation.
There was this horse named Clever Hans whose trick was that you could give him a math problem and he would tap out the answer with his hoof. Of course, his powers were fake, but the way he actually did it is brilliant.
When the number of taps approached the correct answer, his trainer became excited. Hans picked up on this and stopped right when the trainer's excitement reached its peak. The trainer had no idea this was happening. He was fooled by a horse.
And there’s that famous computer vision story of a model that was supposed to detect tanks in images but actually just detected whether or not it was sunny (all the tank pictures in the training set were sunny, non tanks overcast).
Children will make similar classification errors when learning too.
None of this means that it’s not possible they can also learn counting, but just that scientists need to be clever about experimental design. The book goes into that.
Thanks - good to know the truth around that specific story.
The example and class of failure (even if it didn't actually occur in this specific case) is still the point I was trying to make. Those types of failures can happen - even if the specifics of the tank example are urban legend.
We can't know for certain (at least not from that article), but there is some evidence that at least a few exceptional birds can pass the mirror test, and Alex certainly appears to have been an exceptional bird. It's plausible that he understood it was his own image.
This is the new bit, if you've already read about the basics of animal counting:
> The crows mixed up a blank screen more often with images of a single dot than they did with images of two, three or four dots. Recordings of the crows’ brain activity during these tasks revealed that neurons in a region of their brain called the pallium represent zero as a quantity alongside other numerosities, just as is found in the primate prefrontal cortex.
(I'm a little disgruntled about skimming a longish article to find this 2/3 of the way down.)
> Some researchers, for example, propose that while humans have a “true” understanding of numerical concepts, animals only appear to be discriminating between groups of objects based on quantity when they’re instead relying on less abstract characteristics, like size or color.
It's a cognitive bias. People have this because their model of the world is ego centric with themselves as hero and Number 1, and animals far below.
There are some interesting ideas in Sapiens (which I've already been exposed to elsewhere) but I couldn't tolerate the breathless tone of the book. It was written in the same rhetorical style as one of those History Channel documentaries about how Aliens built the Pyramids.
I don't know if I'd call that objectively incorrect, maybe slightly pessimistic, but do you live in the US? A large percentage of people here were very ok with a smaller percentage of people dying so that their own lives wouldn't be inconvenienced, people have that world view even with their own species.
How do we know this isn't another dead fish result?[1] I'm not trying to be dismissive, it's a real question that needs to be asked for any brain scan.
I highly doubt this is the result of a scan, fMRI doesn't have high enough resolution to detect neurons that prefer absence versus a particular number. It's also rare to see fMRI used at all on animal models, maybe a little on monkeys but very uncommon on anything smaller. I would be absolutely shocked if they were trying to scan crows.
That said, I don't find the interpretation of the results very convincing. I'm going only on the Quanta article, perhaps the original research tells a more compelling story. But with the details provided so far I don't see how the physiology can rule out the idea that crows have a concept of "none" but don't consider "none" as a potential number like "two".
fMRI machines are expensive as hell, most institutions share a few between multiple labs and even if they do get a new one don't decommission the old.
IIRC the dead fish result had more to do with software than hardware though. My understanding is that the best of the best labs are careful with the analysis techniques applied, and they wouldn't have even made the dead fish mistake back then.
The commonly used analysis tools are continually improved, so improvements may have propagated to the median fMRI group, but to be honest I still start out skeptical of most fMRI papers unless they're from a group I have prior reason to trust (or they include other data types besides just scans). Too many groups just apply the popular pipelines without an understanding of what they are doing, leading to misuse of tools.
I'm tangential to the fMRI field though, so take with a grain of salt.
TBH I'm not sure exactly what the hardware entails, I just work on the same floor as a couple labs that use the fMRI facilities so I know how much people stress over booking time on the machines. They somewhat recently added a new one that has stronger magnets but the older ones are still frequently booked solid too.
From a neuroscience perspective I do think the good fMRI labs have improved their analyses over time. One thing increasing in popularity that has been helpful IMO is scanning the same participant across multiple days. It's a pain in the ass to get people to come back so it has generally been the norm to scan each participant in a study only once. But more powerful analysis techniques can be performed with these multi-scan datasets, as heterogeneity between and within individuals can be better accounted for.
This review focuses on the default network, but it mentions how the discovery that the default network is actually multiple networks that were being blurred together was enabled by the multi-scan approach. It also gives a decent general gist of some of the things that have been reliably reported in fMRI research versus some things that are lacking, but mainly just from the perspective of questions about the default network: https://pubmed.ncbi.nlm.nih.gov/31492945/
I also like the following paper as a deep dive into the scans of a single person, because he splits apart some seemingly small things like having coffee/breakfast in the morning and shows the effect on scan results. I also think a multi-modal approach is really helpful, so I appreciate the other datatypes included. Plus it's kind of funny the PI wrote a paper entirely about himself: https://www.nature.com/articles/ncomms9885
So yeah I think there is some self reflection and progress being made in the field in a meta sense, it just unfortunately can be drowned out (especially for people in other fields or even subfields of neuro) by the sheer popularity of shitting out fMRI papers.
Thanks that’s really interesting stuff. Since machine availability is a big pain point it would be interesting if anyone’s run the numbers on when it becomes worth it to organize a few groups to buy a machine. A solely research dedicated one.
I don't see how mixing up a blank screen more often with one dot than with multiple dots indicates they think of "none" as a number. The more things there are the easier it is to detect at least one, so it seems completely plausible that something with a concept of "none" but without a concept of "zero" would also mix up no items with one more often than it would with two or three.
Hopefully the original research addressed this, but it felt like the Quanta article was reporting the results in a way completely disconnected with the material it had just introduced.
If we know the animals arrange quantities like a number line, why is there no report on where the animals place an empty set in that experiment? If we know that some young children don't understand the empty set can be a number, then why aren't results from young children on these tasks compared?
There are some interesting results described, but I wasn't convinced at all by the argument about zero specifically, which they make a big deal of in the intro and discussion.
To be honest I think Quanta is a bit overrated on HN in general, but that's another issue.
I agree. I have read it all to really learn not that much (animal cognition is a pet peeve of mine) until this paragraph.
I don't know anymore if we should salute the efforts of the author to contextualize the news, to the detriment of the news itself, or I'd we should blame them for lengthening artificially the article...
Probably a bit of both, as always.
That being said, thanks for the summary of the article, great tl;dr; ;)
Saying "thank you" and upvoting is encouraging, and makes communities stronger.
Blaming people is discouraging, and divides communities. Criticise the article, not the author, and thank the author for what they got right (a good topic for discussion) so they'll create more in future.
We can also increase trust in the community using links. Mentioning names in a good way is particularly helpful, like academic citations.
Thank you for finishing with the "great tl;dr", xcambar, you're asking the right questions :)
The more I learn about animal intelligence, the more I suspect we grossly underestimate it.
I recently saw a video of a seagull that entered a grocery store through the automatic door, picked up a (non-transparent) bag of potato chips (or something similar), walked out again, opened the bag and ate the chips. Think about it for a moment, the bird has to understand (to a degree) how an automatic door works, and that there is food inside these shiny bags. When let that sink in, I was thoroughly impressed. I knew seagulls are very opportunistic eaters, but I did not know they were capable of this degree of intelligence and planning.
I wonder how much more there is discover, and how that might affect how we view and more importantly treat animals. We might have to rethink our relationship with them.
Seagulls eat clams. They break the shells by dropping them onto rocks from a height. They're able to learn to steal food from other birds when they see a chance, and conversely, to regulate from which height to drop clams of different sizes.
All of these require planning and risk assessment abilities. Being able to take a bag of potato chips away from human-infested buildings and opening it is not really that far outside the scope of their adaptations.
A few years back I was walking along the SF bay down in the marina with a friend when suddenly I heard a "clunk" and felt a splash of water on my legs. Initially I thought some drunk idiot threw a beer bottle out the window of their car, but a few seconds later a Seagull swooped down, grabbed a clam it had dropped from who knows how high up, and fly off with lunch. I had no idea they did that, I just stood there dumbfounded for a few seconds.
I once watched several seagulls peck at a closed clamshell container containing pizza crusts for over half an hour. Clearly there is an upper bound to their intelligence.
I watched a crack head at the park look for something on the ground for an hour once. Clearly there is an upper bound for human intelligence as well /s
It's been reported that younger gulls initially try to open clams (the animal clams, not plastic containers) by pecking. I believe they learn the dropping-from-a-height tactic from observation, and suspect that the pecking at a plastic container is merely due to them not having seen it being open by a gull before. Also, worth considering that in their natural habitat, the only transparent things around (jellyfish) are squishy and stingy.
I once spent ten minutes frantically looking for my wristwatch. It was very stressful, because I had to go to school, and I did not have much time left to catch the bus. So after a while (like I said, about ten minutes), I looked at my wristwatch to see how much time I had left. It took another 30 seconds for me to realize that I had been wearing my wristwatch the whole time.
Being smart does not save you (and not me, either!) from being dumb. And there have been times in my life when I was able to substitute inventiveness with stubbornness.
So there are upper bounds to everyone's intelligence, human or animal[0], genius or idiot. Given the variety we can observe in our fellow humans, I think it's fair to assume birds have their "village idiots", too.
[0] Except for barnacles, of course, but they are very discreet about it.
While this appears to be made in both jest and political commentary, I agree that there are bounds to human intelligence.
While we as a species have accomplished a lot, those accomplishments seem to be sparked/possible by the few of us who stop and figure things out or notice something interesting. It seems that most of us take most of the knowledge we have for granted; Even the most essential pieces of our knowledge are passed down, not figured out, by the vast majority of us.
Examples: fire, sanitary practices, what to eat (and what not to eat), language (and writing)
How I justify these examples:
It took a long time (many generations) to harness fire and still more to understand it well enough to use it as we do today. We are still pushing the limits of our understanding of fire (combustion) in microgravity to better understand how it works on Earth.
Sanitary practices are different in different cultures despite all of us having the same basic plumbing. Sure, we can all "poo in a hole" but even in modern times, we've had to re-learn to not make that hole too close to our water supply.
Food: There is no way of knowing what mushrooms will kill you without someone previously taking a hit for the tribe.
Language seems pretty self-explanatory but... there are languages we no longer know and are largely unable to decipher because we have stopped passing them down. Knowledge encoded in those languages may be lost until rediscovered in a new language.
I think the biggest bound to our perceived intelligence is the ability to pass it down. When viewed in the scope of animal intelligence, we often think animals that have learned from other animals are acting more intelligently than those that haven't. The problem is, both groups may be equally intelligent, one group is just more highly trained.
> The more I learn about animal intelligence, the more I suspect we grossly underestimate it.
It goes both ways, in the past we had Clever Hans[1]. People believed Hans could perform simple math, when his cleverness was actually watching his owners expression for small changes. So any observed intelligence ends up under suspicion of them acting on information gained from a much simpler side channel.
The thing about animal intelligence is that it is probably significantly different from ours. In studying it, we must avoid antromorphization of animals.
Mammals are somewhat close to us, but we have diverged from birds over a hundred million years ago and we still struggle to understand how a tiny and smooth brain such as corvids have can produce such an observably intelligent behavior.
Mammals are close to us in the dimensions of reproduction and biological ancestry; also, there are countless mammals which are (way) less intelligent than corvids.
Corvids may have small brains in terms of volume, but not measured by the number of neurons in the areas that count[1].
Maybe animal intelligence is not so different to ours, but our measure is.
"The more I learn about animal intelligence, the more I suspect we grossly underestimate it."
An example of something that might surprise many people: animals can use medicine.
Here is a fascinating podcast that discuss the subject with researchers and experts.
Podcast description:
> Listener Andrew Chen got in touch to ask whether animals use any kind of medicine themselves. After all, our own drugs largely come from the plants and minerals found in wild habitats. So perhaps animals themselves are using medicines they find in nature.
> We think of medicine as a human invention - but it turns out that we’ve learnt a lot of what we know from copying the birds, bugs and beasts.
This podcast is highly recommended - I guarantee you'll come away amazed and humbled by the intelligence of animals.
Blue Planet II depicts a group of Dolphins visiting a certain kind of coral that emits a substance that allegedly helps them fight off parasites.
Earlier BBC documentaries show animals (birds and fish) seek the "service" of certain types of fish to clean them from parasites.
So the idea of animals using medicine is kind of presenting itself. It's not totally in your face, but it's no more out there than prehistoric humans discovering the medicinal properties of certain plants.
I never fully understood why some cultures assume humans must be fundamentally different - it’s not a universal notion.
Some hypothesizes and argues that our ancestors extinguished every other species that showed intelligence by eating them, which I think is a bit too wild of a theory. But maybe the fact that it’s often small birds that shows intelligence is also interesting. They’re among hardest to hunt.
Descartes considered animals to be little better than automata and given his high social position and intellectual achievements, and the convenience of this argument for his co-religionists, it stuck. Even today (and right here on HN), any sort of conjecture on the inner life of animals tends to bring out a few finger-waggers condemning 'athropmorphism.'
> some cultures assume humans must be fundamentally different - it’s not a universal notion.
Are you referring to hunter gatherers doing a mystical bear or eagle dance? Reincarnation as an ant? in those cases, it is belief in a non-corporeal spirit-being that underpins the other belief. Do you believe in non-corporeal spirit-beings, and if you do, aren't you more mystified by people who don't believe in them; or vice versa?
Otherwise, I'm skeptical and would love citations?
Like, judeo-christianity? Western societies viewpoints on human intelligence and how it relates to animal intelligence for centuries? The notion, typically, is that while humans are animals our intelligence is not just different in degree but different in kind. I think it comes from that bit in the bible where their god places humans as above all other life forms, just a guess though.
It's not quite the same thing as "intelligence", but in "Braiding Sweetgrass" Robin Wall Kimmerer describes how (at least) several native American cultures ascribe great wisdom to the trees, since they've been here longer, and regard human wisdom as rather primitive.
That's a bit of hyperbole, isn't it? You never understood why some [most] cultures assume humans must be fundamentally different [than other animals]?? Well, smelting iron ore, and a long, long list of other unique accomplishments never replicated by any other species on the planet [or known universe for that matter] tends to crystalize the observation that there is something fundamentally different about this singular species. That animals are smarter than human cultures have ever generally given them credit for doesn't bridge this enormous gulf. Humans generally think less of (underestimate) anything not relevant to their immediate existence, including other humans (from other cultures, time periods, political parties, etc). Same reason my dog is uniquely intelligent, and can even understand English, but dogs in general are not intelligent. Maybe this tendency to denigrate is an evolved survival skill. Nonetheless, call me when the seagulls start splitting atoms.
> I never fully understood why some cultures assume humans must be fundamentally different - it’s not a universal notion.
Well, riddle me this. In this thread people are marveling at animals doing things like using automatic doors and opening bags of chips. Who would be impressed if a human did these things? I don't exactly find myself persuaded by these examples that actually animal intelligence is just like human intelligence.
> Think about it for a moment, the bird has to understand (to a degree) how an automatic door works
I'm not sure I see that, that is, the "understanding" part. I mean, this is not some new kind of behavior, even if everyone here reacts with surprise. Anyone with a pet dog or cat knows the various mischief they can get into. Many of us have been at the beach and watched seagulls do all sorts of interesting stuff. Animals aren't "dumb" and I don't know anyone who would say that they are. What people do object to is the unwarranted inference that because they display what looks like clever behavior then this must entail capacity for abstract thought with concepts. I don't see how abstraction is necessary in these examples at all. Language in the fullest sense demonstrates the existence of abstract thought. Non-human language does not appear to show any sign of descriptive and argumentative function, only signalling and expression (in the Popperian sense).
I have been training the crows in my neighbourhood to solve puzzles. They visit me every day now.
One goal I'm really excited about is to see if they can solve a puzzle that requires two crows to work together, because I've only seen videos of individual crows solving puzzles online. Can you tell me more about the teamwork that you're referring to? Are there videos or papers I can look at? I'm super curious!
Yes, adding a +1 to sibling comment: do tell about your puzzles. We've got some crows that just get bitchy if their bird bath gets dry, I'd like a more positive interaction if possible. :-)
with a peanut machine to attract them and regularly make the setup more intricate to see if they manage to find the metaphorical key to get more seeds. more like crowd strategy search ?
It all began just before the pandemic, when I saw some YouTube videos about crow intelligence that blew me away. There were crows all over my neighbourhood, so I left out peanuts on my balcony, and eventually a crow started coming by to get them. I would replenish them at night and the crow would come the next day.
After I did this consistently for a while, the crow came every morning. So then I decided to hide the peanuts in a container. I had a retainer case that springs open when you press on a button; I taped it to the railing and left it open with a peanut in it. The crow was very suspicious of it at first, giving it a wide berth while examining it from all sides. It took about a week for it to work up the courage to steal the peanut from the open container, and then another week for it to get used to the idea that the container wasn't going to cause it harm. Then I closed the container.
The crow learned to peck the button to pop open the retainer case, and that's when I got hooked. The crow brought over its partner, and later in the spring as I was stuck at home in lockdown, its fledgling came along too, and they would show the fledgling how to get the peanuts. I started putting out a series of different containers, and then eventually got them a dog intelligence toy, which they aced. I think these crows aren't as smart as the New Caledonian crows in the videos; it takes them a few days or weeks of experimenting to figure out a puzzle. It got to the point where I had four or five crows basically chilling at my house on most days, playing with the toys and taking baths and discussing the puzzles. I have way too much webcam footage of all this, which I've been meaning to post but haven't had the time to sort through and edit.
I have since moved to a new house and had to say goodbye to those crows. :( I have new crow friends now, though! I toss them peanuts whenever I see them on my street, and now they will follow me for a couple blocks if I walk by. This time I've made them a nice platform to land on, and they're learning to open containers, and are starting to play with a dog intelligence toy now.
Not only do crows learn to recognize PEOPLE by their FACES, they teach their offspring to avoid/mistrust those same people. Those offspring teach their offspring.
So the "grandchild" of the original bird learns to avoid a specific person. *(Note the documentary I watched with this revelation didn't follow any further generations. We don't know how many generations are taught this lesson.)
Think about that for a minute. I don't recognize the people that live 2 houses away from me, these birds teach their young that SPECIFIC humans are dangerous.
YES! I am pretty sure I have watched the same documentary! It was the one that a town hired a falconer to try to get rid of a murder of crows that had decided to move in?
That documentary was so interesting(in my opinion at least). I think it might be on Youtube, but I don't remember the name.
They are smart but TBH that crow already did all the sub-puzzles independently, he knew what the possible moves are just had to execute them in some order.
To be fair, that's how most of us approach puzzle games. "Okay I've learnt the individual tricks, now let me just try piecing them together in the order that they become available."
Yes but usually you have 100s or millions of possible combinations, in that crow puzzle he had to do 5 things and the order between 2,3,4 didn't matter. Effectively permutation of 3 so 6 possible orderings :)
I'm pretty sure some dogs could do this (if the tasks were changed to suit their natural range of motions and interactions with environment). When you teach a dog 4 tricks and try to teach him 5th he will randomly repeat the previous 4 in different combinations trying to get the treat. At least some dogs, others don't care much and if the first thing isn't working they give up.
My parents had a dog that was very smart and pretty aggressive to our cat. He had a dog kennel in a small section of backyard enclosed with high fence. Near the fence there was a small cherry tree where our cat liked to climb.
Cat wouldn't climb the tree when the dog was outside of his part of backyard, so the dog not only learnt to open the gate to the rest of the backyard (lifting a wire loop over the fence and pulling it just right), but also to shut it slightly to fool the cat. He would lie there pretending the gate is still closed, not reacting to the cat. Cat would climb the tree and then the dog would stop pretending, open the gate, go under the tree and bark at the cat till we come and free him. If we weren't home that could take hours - neighbors weren't too pleased.
Other example is dogs burying and re-burying treats when they see each other watching them bury it. That requires at least rudimentary theory of mind.
As for crows - I've personally seen them doing this trick (on a parking lot not on a pedestrian crossing, but same thing - they understood where the traffic is just right to be used for nut-cracking).
They are also huge assholes to other animals, bullying them in very creative ways.
I don't doubt crows are smart, just don't think that experiment was that impressive.
One time when my daughter was 2 years old, riding on my wife's back in an Ergo baby carrier while we were walking up the Brighton Pier eating an apple, a seagull came stealthily hovering in from behind on the strong winds and took a bite out of the apple from my daughter's dangling arm before any of us could react.
The idea that zero was "invented" at a certain time by recent humans and not in use until then has always seemed absolutely absurd to me. I will admit I haven't read deeply into it, but there's just no way people didn't have a concept for "none of a thing" until a couple millennia ago.
One of the justifications used for this reasoning is "You don't go to the market and buy 0 fish". But all that tells us is: There's no reason to record buying 0 fish or owning 0 acres of land. Another justification given is the difficulty children experience with 0 - but we don't teach children to start counting with 0, so it makes sense that they would get tripped up there.
I guess the argument sometimes seems to be that we didn't have a symbol for 0, and that this was somehow more confusing to adopt than other symbols? But if that's the case, then isn't claiming any culture "invented" zero the same as claiming a culture "invented" any concept they came up with a word for?
I'd love to hear all the reasons I am wrong and stupid.
I think the thing which came surprisingly late in history wasn't having a concept of 'none' but rather of allowing it to be considered a number with the same status as 1, 2, 3 and the rest.
Indeed some of the ancient Greeks were of the opinion that 1 was not a number (since numbers were for counting a plurality of things).
It's quite the realisation that you can use the usual rules of arithmetic for 1 and 0, and not have anything go wrong. (Except of course division does go wrong!)
You see this in programming languages too. Many languages distinguish between one of something and a collection of them, e.g. int x vs vector<int> x. In other languages, like Matlab, everything's a vector, and a scalar is just a vector of length one.
Well, its not that they distinguish between one of something and a collection, its that one of something is different from a collection of things.
So, int x is a handle to an integer value, whereas an array handle (your collection) is a handle to a pointer, which points to the start of your data structure and you can get the other elements by using an offset etc. etc.
My point being that they are differentiated because they are different things.
Having a concept of "none of a thing" is one thing. Abstracting that to a number is a different thing.
Even today, when you ask someone "how many kids to you have," and they don't have any, they say "I don't have any," not "zero." In other words, they respond with a phrase, often one with a negation (not), instead of a number.
Even the question whether one can 'think' of something without having a linguistic representation is a topic of active debate. My personal take is that its possible, predators plan ambush, but do they have a linguistic representation ? Its not clear that they do.
> Even the question whether one can 'think' of something without having a linguistic representation is a topic of active debate.
This seems fairly trivially true just based on how common it is for someone to have 100% full understanding of a concept and then be completely incapable of remembering the word describing it, or even a phrase that conveys a decent approximation.
Linguistic representation can deal with what's-the-word-again place holders though. So, not being able to recall the name of an object or concept does not prevent linguistic thinking.
Hellen Keller's writings may interest you. Her recollections from a time when she did not have an internal language is very interesting.
Many believe that animals do not have an "I" the self reflective "I", that they are not aware of themselves etc etc. This runs contrary to my beliefs, I have had several conversations/arguments on HN along those lines, but lets not dwell on mere beliefs.
What I find interesting is what test/experiment can one perform that can demonstrate that a human, who is not allowed to communicate linguistically, has the attributes mentioned above. If we cannot design such a compelling experiment that shows are inability to detect those attributes in animal even if those attributes maybe present.
We need the restriction of no-linguistic-communication so that animals and humans are on the same playing field. Hellen Keller, when she did not have language she would have been on the same playing field.
My position is that if I cant even prove/demonstrate to others that you are sentient in the senses described above, how can we even claim that animals aren't sentient. We have no way of demonstrating it even if they were.
Seems contrived. 'None' is an equally valid answer, and easy to conceptualize as a number, eg 'start with three, take away two, take away one, now you have none.'
There's a book called "Zero: The Biography of a Dangerous Idea" that goes into the history of zero in detail, admittedly I haven't finished reading the book myself but got at least partway through and it was fascinating to read about it and I'll one day finish it :)
What you said about it not being in the numeric system is definitely part of it (ex: roman numerals not having it) but also all the problems that come up with zero have to be dealt with (e.g allowing dividing by it allows you to prove anything, and there's a great proof that winston churchill is a carrot in the book showing as such), and there's some overlap with religions in fearing "nothing" and what that might mean
I more or less had understood that what was meant by "zero was invented" is zero as used in a positional number system. In other words, the digit zero, not the quantity zero.
Zero is an overloaded concept. There are at least like 5 or so different things that we use the word "zero" for, some of which are very sophisticated.
On the one hand, it means there's none of a thing in there. Most animals understand this. There is zero food here. There are zero predators on that island. Pretty sure crows have this one down pat.
Taking it one step further, you have zero as a number. Zero which is on equal footing as numbers like one or seven-- it is a number that you can perform arithmetic on or with. You can multiple by zero, you can add by zero, you can divide by -- wait, you can't divide by zero! The Greeks understood this. But lots of people say the Greek's "didn't have/didn't invent zero" because they used the word "nothing" instead of its own word. But Aristotle did say "nothing and nothing, added together, make nothing" and "there is no ratio of nothing to a number".
Crows probably don't understand that you cannot divide by zero. Maybe you could teach a crow that if two crows got ten total nuts, each crow would get five nuts, and so on. If you told that crow to split ten nuts between no crows, and ask the crow how many nuts each crow would get, the crow might tell you ten nuts, but it's not going to tell you that that operation is undefined.
Then there's the use of zero as a digit in a place-value system, like the one we use, as opposed to Roman numerals or Chinese numerals or so forth. Before there was a symbol for zero, you couldn't write one hundred and four as 104, it would just be 14 which was ambiguous. So when the Indians "invented zero" in the 7th century or whatever, it wasn't zero that they were inventing, it was the place-value system we use today to conduct our daily lives with, which is a shitload better than Roman numerals or trying to do arithmetic with a straightedge and compass or whatever.
There are a lot of people who claim that, for instance, the Maya were more mathematically advanced than the ancient Greeks, because the Maya had a symbol for zero (which looks sorta like a French baguette and an American football had a baby) and the Greeks didn't. While the Mayan way of writing down a number was much better than the Greek method, we don't have any evidence that the Mayans had much in the way of math beyond counting and extrapolating. The Mayans primary method of calculating is by drawing out tables: they knew the cycle of Venus is (roughly) 585 days, they knew the cycle of Mercury was 117 days, so you'd get a table with thousands of entries counting up the cycles. The Dresden codex has a table with 2340 entries, with 585 entries for Venus repeated 4 times and 117 entries for Mercury repeated 20 times. With a place-value system that used zeroes for zero digits. Which... ok, the place-value system was more advanced than the Greeks, but the Greeks would have done that same 'calculation' with a lot less tedium.
Then you have zero as in the additive identity, which opens whole new worlds of mathematics to you. Now you can start talking about things like rings, fields, and groups, and can have all sorts of fun. Pretty sure crows haven't sat down and discusses abelian vs non-abelian groups, but I wouldn't put it past them.
So anyway, any time someone starts talking about inventing or understanding the concept of zero but do not, themselves, display an understanding the concept of zero, (by explaining what it means in that context) it's generally a good time to stop listening.
If I don't give my cats the usual number of treats, they look at me like they know they're being cheated. I never thought they could count, but I figured they have a sense of relative quantity ('enough', 'more', 'less' and so on).
My dog just does a protest sit-in when I do that. He wouldnt move till he gets his due number (two). I can fool him by splitting a treat into two, so seems he is counting rather than going by mass.
It depends on cats, I have had a few in the past and currently have one that always tries to convince me when I get home that for some reasons my wife forgot to feed her.
Clearly she has clear the negative concept of "not enough".
Do the birds and the bees actually count? Or is there some more innate/intuitive process involved. I doubt I could tell the difference between 9 or 10 dots on a page without actually counting them.
Many years ago there was a study involving crows, as I recall. N, starting at 2, researchers went into a blind, then N-1 came out. They had to get up to N = 13 before they fooled the birds into thinking the blind was actually empty.
For us humans it is a process of establishing a one-to-one relationship between objects and a long-since-memorized sequence of words that requires just one member of the sequence to be remembered in order to represent the number of objects.
I think that's part of it to be sure. However I can readily account nine items versus ten items depending on how they are organized. We are counting when we are enumerating, however that occurs in man, bird or bee it seems to me.
This article goes into it and mentions "numerical distance effect", which looks appropriate but not what I remember.
I did learn that people who play first person shooters can usually recognize more objects at once instantly; 6 or 7 instead of a more typical 4 or 5, if I recall correctly.
Article seems to resolutely believe that the entities 600MM+ years ago could not understand numbers and thus counting emerged recently many times over in parallel across a variety of species [1], but provides nothing to back up this claim ([2] is not supported beyond “an expert said).
This is an odd claim —- based on the neural network experiment (which showed that gradient descent applied to object detection problems naturally results in the development of “counting” signals), it seems much more likely that for as long as beings have been interacting with their surroundings they’ve been able to count, and any modern slight differences in counting infrastructure are simply a divergence of the original counting machinery rather than the same thing being invented over and over by evolution in many different ways, but only recently (for some reason..?)
> The fact that those three species are from diverse taxonomic groups — primates, insects and birds — suggests that certain numerical abilities have evolved over and over again throughout the animal kingdom.
> Their last common ancestor “was [barely] able to perceive anything,” Avarguès-Weber said, much less count.
> Moreover, he added, “When you look at the history of mathematics, it turns out that [zero] is an extreme latecomer in our culture as well.” Historical research finds that human societies didn’t begin to use zero as a number in their mathematical calculations until around the seventh century.
There are huge differences between zero as the concept of "none" (I had five cows and someone took them all away, how many do I have?), zero as a more formal integer value (e.g. solution to 2 + 7 - 4 + x = 5), zero as unit placeholder (e.g. 101 dalmations), and zero as a real number (on the continuum from positive to negative rationals and irrationals).
When people make claims about "zero" they're essentially meaningless unless they specify which meaning of zero they're talking about.
I really wonder how zero came to be, was it a need to homogenize the algebraic tools (we have nice operations and want to reify the 'nothing' dead end into something that can be kept chugging along) or something else /
The ancient Greeks understood division and reciprocals from the perspective of inverting multiplication.
I have to imagine that the multiplicative identity (1) and additive identity (0) were also known at that time. They really figured out everything else there is to integer-based math.
They failed at irrational numbers (like sqrt(2)). But their mastery of integers was outstanding. Perhaps 0 wasn't used on the number-representation yet, but the concept had to have been known.
There's a whole book on this history of zero you may enjoy called "Zero: The Biography of a Dangerous Idea" that goes into the whole story of how it came to be including problems and resistance along the way
(mods I promise I'm not here to shill this book but people keep wondering about the history and that's a great resource that I like XD)
I think most such research don't delve upon ancient India wherein use of zero with a proper modern like numerals have been extensively observed at least since thousands of years [0].
> Hindu units of time are described in Hindu texts ranging from microseconds to trillions of years, including cycles of cosmic time that repeat general events in Hindu cosmology.
You don't appreciate zero until you try to invent a positional number system. In the past, when they used positional system they would often just leave a blank spot in the place of zero. This inevitably leads to errors. Is that two numbers or one number with zero between? How do you distinguish between 1 and 10?
Source:
A popular science book
"The history of numbers or the history of a great discovery" (Georges Ifrah)
Contains some fun chapters like "How to count if you can't count". For example if natives in a stereotypical jungle want to count numbers to a holiday. They memorize a sequence of items, for example body parts: little finger of one hand, palm, wrist, forearm, arm, shoulder, neck.... then they measure once and conclude one needs to wait as many days as there are from the little finger on left hand to, say, navel.
"An understanding of numbers is often viewed as a distinctly human faculty ... sets us apart from all other animals." Then it goes into animals that count. My, oh my.
Mom hen knows if a chick is missing. Can she count? Of course she can, how else would she know it. But show me a chicken that will abstract number 5. As 5 chicks, 5 pieces of grain, 5 drops of rain or 5 whatever, so that five becomes an object by itself. I thought it is common knowledge that our ability to abstract is what separates us from animals, but seeing people interested in horses and crows able to count I am starting to doubt it really is so. Animals (mammals at least) feel like us...love, anger, pleasure...they might not hate as us though. Animals can build...spiders, beavers and so on. They can think, communicate, count, have social structures, dream... It is ability of our intellects for abstraction that differs us from them.
Animals can not abstract numbers, they miss faculty for it, whatever that faculty is.
None of the experiments in the article seem indicative of animals/crows understanding the concept of zero as a symbol for nothing. All animals understand the concept of "none" or "not present", and can possibly count, but they don't necessarily map a symbol to represent absence.
> The crows mixed up a blank screen more often with images of a single dot than they did with images of two, three or four dots.
This can very well happen even if there is no understanding of zero as a symbol. You're looking for nothing, and the screen with 1 dot is almost that - in terms of the existing symbol occupying the least space, closest to nothing. Not evidence of a mapping of nothing to a symbol at all.
Cheyennes divide birds into 3 groups: plain, great and sacred. They count magpie in the sacred ones :-). There's a magpie on my fence as I'm typing this.
They like magpies because they're smart, talkative, they announce guests, they live close to humans.
Why are we so surprised that animals understand the concept of zero? Being hungry, searching for food, and finding nothing seems like as universal and visceral experience of zero as anything I can imagine. I don’t think it’s possible to evaluate food sources without some idea of what zero means.
There's a difference between having the concept of number and reacting to quantities or having expectations that are unmet. There is no need for abstract concepts. All animals (in the phylogenetic sense) other than ourselves perceive reality utterly concretely. I am not sure how these brain scans are supposed to really show otherwise.
Show me an animal that can describe and argue and you will have my attention. For now, I only see signalling and expression, but no language showing signs of description and argumentation.
> This is an odd claim —- based on the neural network experiment (which showed that gradient descent applied to object detection problems naturally results in the development of “counting” signals),
"Animals cannot, in principle, use abstract concepts because it requires language and brain centers which never been evolved in these branches of evolution."
[citation needed]
Also, bird brains are very different from ours and knowledge of brain centers in mammals cannot be easily transferred to birds.
In another universe, a bird with your attitude could claim that flight requires feathers and thus non-birds cannot in principle fly, because they never evolved feathers.
My friend, the whole "study" is based on false assumptions and consist of interpretation of the data in the way which confirms these assumptions. This is not even science, this is some blog posting kind of crap.
Every bit of observable behaviour could be explained in terms of habitual behaviour and physical pattern matching, exactly what a trained NN will do - it will perfectly accurately separate patterns which corresponds to different quantities, while posses no notion of any abstract concepts whatsoever.
Animals pattern match, but do not have abstract concepts. Well, enzymes do pattern matching but not counting.
"Every bit of observable behaviour could be explained in terms of habitual behaviour and physical pattern matching, exactly what a trained NN will do - it will perfectly accurately separate patterns which corresponds to different quantities, while posses no notion of any abstract concepts whatsoever."
I think this is interesting -- what's stopping us from making that exact assumption? For all we know, higher consciousness and "humanity" is simply an emergent property of lots and lots of rote pattern recognition stacked on top of itself. I think there's an argument to be made that there's no line between "pattern matching" and "abstract concepts" -- it's more of a smooth, emergent gradient.
Could you please stop breaking the site guidelines so we don't have to keep banning you? If you'd make your substantive points thoughtfully, we'd appreciate it.
The problem with the way you're posting is that it pushes the forum further in the crap direction which is already the internet default, which we're trying to stave off here—not for moral reasons but simply because it's uninteresting.
How does one show a flaw in a proof or a logical argument? A single contradiction or a plain error is enough. This is the core of science - a single flaw is enough to throw away the whole argument. Not just that, having the lack of basic understanding been exposed, a "scientist" should be marked as incompetent.
Yes, you systematically ban people like me and have narcissistic imposters and bullshitters instead. Millions of them.
This is the consequence of eradication of anonymity and heavy policing to comply with current idiotic pseudo-intellectual narrative of safe spaces or how do you justify it.
The next step would be enforced gender-neutral pronouns or something equally nonsensical.
An internet forum isn't a math journal. You're leaving out the possibility, and the necessity, of relating to other people. Perhaps this thread would help: https://news.ycombinator.com/item?id=27162386.
> Crows recently demonstrated an understanding of the concept of zero.
No they didn't.
All human cultures I'm aware of have, at least, the concepts of none, one, two, three, and many. They all understand that none is exactly one less than one, and most of them (but not all!) understand that there's no natural limit to the counting numbers, and have a way of expressing some amount of those numbers.
But zero is a human invention. It comes from India (probably, might be China) and is probably related to the abacus: but the abacus came to the West without the zero, which followed it later.
Zero is both a symbol for null and a placeholder used for writing numbers modulus the radix. This is a conceptually powerful force multiplier, making all arithmetic operations dramatically easier.
It also lets us make up a "name" for any integer we want. We can just add more zeroes; when this gets awkward we can start counting the zeroes using exponential notation, and when that becomes inconvenient we can develop ↑ notation, which gets us further than we really need for any practical purpose.
It should be obvious that crows neither know about this nor care. It's interesting and cool that they understand none, but so did the Ancient Greeks, who had no idea about zero and could have gotten much further in mathematics if they did.
I watched a thing about how Koko the guerrilla didnt really understand sign language. They showed extended, unedited clips of koko fumbling and just brute forcing a whole series of signs until she got a treat. A lot of her "understanding", along with other apes undergoing the same treatment, were all heavy interpretations of the "researchers". She never actually learned a real version of sign language either. It was made up by "researchers" who never learned sign language either. Honestly, when you watch the videos, it feels like watching the unedited videos of those tv psychics who fumble at cold readings. Cherry picking editing does wonders as "scientific proof".
Moral of the story, after reading this article, I agree with you. This isnt as groundbreaking or interesting as it seems. I'm pretty sure they're finding patterns that dont exist and are over rationalizing. Of course an animal would have the concept of none, a little, a bit more and a lot, especially when applied to the idea of scavenging for food. Appling the idea of mathematical concepts to them, with little support just feels like a cheap attempt for grant money. Which, surprise surprise, was the same sham pulled with Koko.
That and some of these tests they talk about, either they're shit at describing them or even I would fail to prove I could count or do arithmetic.
I would be a little kinder to the researchers than that: crows and other animals understanding the concept of none is interesting and plausibly true.
My quibble is conflating this with zero, which is a much more powerful and advanced concept. Greeks, Romans, early medieval Europeans, none of them had the concept of zero, and they suffered for it.
And what you've said about Koko the gorilla is true as far as I understand it. It was a media stunt, more Clever Hans than Mister Ed. It would be pretty cool if it wasn't, but I'd have to see some replication and I don't expect to.
> They showed extended, unedited clips of koko fumbling and just brute forcing a whole series of signs until she got a treat.
...what do you think human children are doing for the first ~4 years of their life?
Also, different learning methods are used by the subject depending on their mood -- most importantly frustration -- and other context. Brute force learning tends to be the last stage before they're so frustrated they give up or become aggressive. It's also a common exploratory tactic.
Teaching human sign language over a variant organically developed by the subject and team would have what advantage, exactly? It may be nominally more interesting to know a gorilla could adopt a purely human communication system -- very interesting. But it's also such a corner case of studying communication and particularly interspecies communication that to call it out as a problem is nothing more useful than a nitpick.
We're coming from an age where all animals were by default assumed to be a sort of squishy -- sometimes tasty -- automaton. We have a long way to go in unlearning that particular arrogance. Assuming that intelligence can only be demonstrated in the purely human terms you seem to be looking for is just another vestige of that same arrogance.
How does that strengthen his point? Koko stumbled around frequently, well after the learning period was advertised as done. Kids stumble around less than was claimed, as you mention.
You're asking an animal to internalize a system of communication quite literally designed from the ground up for different hardware. It's like asking why does my Oculus headset take so long to communicate over ZigBee. What would be magical is that it did it at all. Lamenting how long it took to bend the Oculus hardware into some contorted SDR would be stupid.
My point is you shouldn't fudge auxiliary data in making a point. I don't know if I can believe your actual argument due to that.
Consider, my dog can fumble around trying to make me happy with really good emotional understanding of my reactions and basic mimicry. I wouldn't claim that is a usage of language, though. From the counters here, there is an argument that if my dog was dextrous with his arms and hands, he could pass off similar results.
I didn't say it makes it a strong point. But deceitful numbers greatly weaken it.
That is, with how off the idea that 4 year old can't take language is, I didn't make it to the argument. And I am sympathetic to the idea that animals can pick up on language.
Why do you believe these sources, and your off-the-cuff analysis, and not the original research? Think of it from other readers perspectives: On one hand is the author of the article and the researchers, on the other is a commenter on HN. I'm not saying the parent comment is wrong, but we need some evidence.
Koko appeared much, much more in the media and pop sci publications than the actual scientific literature. There are also a number of researchers who questioned Koko in the literature, including people that tried to replicate with their own animals. There is really no way to make an argument from authority here, because Koko is a controversial topic among the "authorities".
We probably are underrating animal intelligence, and whether Koko understood language is largely dependent on what one considers to qualify as "language" - there definitely were some impressive displays of intelligence at times. But through an unbiased lens I think it's also pretty clear the researchers were constantly making the most charitable possible interpretation of Koko's signs.
As far as the GP comment, it is well documented that zero as a mathematical concept was developed surprisingly late, while the concept of "nothing" existed prior to any formal notion of integers at all. The Quanta article is conflating "nothing" with "zero", although perhaps the original research makes the distinction more clear.
They were training a younger parrot and trying to get the younger parrot to count to two by tapping twice.
Alex overheard the training and got impatient with the other bird. He yelled out “two” and then after two more taps “four” and then “six”.
The trainers were just expecting “two” each time.
It was this book: https://www.amazon.com/Are-Smart-Enough-Know-Animals/dp/0393...
The book is interesting and goes into how humans need to set up experiments properly to actually test non-human animals in ways that make sense (rather than just in some biased human way).
One quick example was testing tool use, the original experimenters left branches on the ground for the monkeys to use, but the monkeys can’t pick stuff up that’s flat on the ground since they’re normally in trees (their hands don’t have thumbs that move that way). When he redid the experiment with the tool raised they were able to grab and use it.
Same author also wrote Chimpanzee Politics and did this great video experiment: https://youtu.be/meiU6TxysCg