Ravens score just as high as big-brained chimps on cognitive tests | Hacker News Reader

Ravens score just as high as big-brained chimps on cognitive tests

arstechnica.com

160 pointsshawndumas60 comments

My fascination for corvus started when I saw Joshua Klein's Ted presentation in 2008[1]. Inspired by these recent findings I two weeks ago decided to get to know some crows. I bike commute every weekday and pass a place were I've noticed there's almost always a flock of hooded crows (corvus cornix).

I decided to feed them twice a day (when going to work and when leaving work). The first time they were very reluctant to pick up the nuts and seeds I threw to them. Me throwing things in their direction understandably made them hesitant. After I left however, I saw from a distance how they inspected the nuts and picked them up.

I did this two times a day for two more days. The day after that I saw a crow sitting on the ground a few hundred meters away. Soon it started flying straight towards me on the bike path. I slowed down and it landed straight in front of my bike and started cawing. I picked up the plastic bag of nuts, and it came right to me.

Fast forward to now. When they are at the place they come to meet me. A few times they have also showed up from nowhere (almost eerily) when I stop with my bike. One second I swear I can't see a bird anywhere, and a second later they land next to me. I've also had a fantastic experience were 10-20 crows flew with me and right next to me (when riding my bike) for a few hundred meters.

It has become a nice part of the day that lightens up my commute. I will keep doing this and see how things evolve. All in all, I would recommend building some interspecies relationships with crows. :)

[1] https://www.ted.com/talks/joshua_klein_on_the_intelligence_o...

Your fun story reminds me of the BBC story "The girl who gets gifts from birds":

http://www.bbc.com/news/magazine-31604026

Yeah, that article also served as inspiration before my first feeding day. :)

Feeding wildlife is generally not a good idea. You can easily end up hurting or even killing the animals that you're trying to help. I hope you did your research on their diet and behaviours.

I don't think he's trying to help them. He was experimenting to see if he could train/domesticate them.

Crows get at a lot of human rubbish to pick tasty stuff out of it. In Japan, garbage cans in many places have to be locked-down and covered with nets to deter crows. If that hasn't killed them off already I doubt some intentional feeding will. It might make them a pest if you don't want them around, though.

or in the worst case, train them to attack people for change

http://www.npr.org/templates/story/story.php?storyId=8787802...

We've had previous discussions on here about crows etc and I vaguely remember a story about someone who managed to start a gang fight between rival crow gangs. I wish I could find the link.

You may be thinking of the famous World War Crow greentext from 4chan: http://i.imgur.com/f50IZJS.png

Very cool! But sounds a little poetical - makes a good story anyhow.

This is brilliant. I so wish I enrolled for an ethology PhD now...

It's from an old 4chan thread:

https://imgur.com/gallery/rgfZm2d

Thanks both for posting the original.

When I was a teenager I found a baby crow that had been attacked by something and was in a bad state. I took it home looked after it for a month or so feeding it on cat food and keeping it in a garage. Maybe not the best move but I was a teenager. It grew up and lived in the wild in the end but would sometime fly down and land on my shoulder as I walked to the shops etc. I would stick it on a post box buy a few things and it would hop back on my shoulder when I came out. Made me feel quite badass ;)

Crows are extremely intelligent and they live quite a long time. When I was a kid I went on a vacation with my family to a cabin by a lake. A man living in the neighborhood had a pet crow. The crow would get cosy around our dining table that was outside trying to grab some food, so my mother chased it away once. After that for the next 2 weeks we stayed there the crow would fly over every day and take a shit on my mother's laundry. There were 5 people in the family, but the crow would always pick my mother's clothes. This is a true story.

That is hilarious, I wonder if crows have a sense of humour, sounds like they like toilet humour?

I wanna say 'we think'* parrots have a sense of humor and I'm pretty sure we attribute 'play' as a motive for various crow behaviors so it seems pretty reasonable to assume they might have something we'd recognize as a sense of humor.

* 'we think' as in various pop-sci articles, so obviously apply your necessary skepticism filters.

Whenever this topic comes up I always find myself thinking "why not?"

We know humans are animals, and as far as we can tell, toilet humor is universal across cultures. Animals aren't as intelligent as humans, sure--but does anyone think that poop jokes require a lot of intelligence to get?

I feel like the history of studying animal cognition can be thought of as long process of us just getting over how special and unique we think we are. First it was problem-solving vs. pre-programmed instincts, then it was "man, the tool user", then it was language, and now sense of humor? Well why not?

Not sure if humor or legitimate vengeance. In a group it's important to punish individuals infringing on your sense of entitlement ;)

Anyway, it's impressive that the bird managed to connect so many dots correctly and creatively play an inter-species conflict.

It's interesting that this story sounds far-fetched but if you told it about a chimp no one would bat an eye.

I think we can accept how intelligent chimps are more easily because we can understand their body language and facial expressions. Even dolphins and dogs are more relatable.

There is a phenomenon where individuals with hydrocephalus, a condition where the brain cavity is filled with fluid, can still exhibit normal intelligence. I suppose the two broad explanations are that brain matter is either denser in these individuals, or that a brain can still compute well with less brain matter at least in some cases (possibly by sacrificing on some other not-easily-measured mental faculties)

"There's a young student at this university," says Lorber, "who has an IQ of 126, has gained a first-class honors degree in mathematics, and is socially completely normal. And yet the boy has virtually no brain."

http://www.psych.ufl.edu/~steh/PSB6099/brainnecessary.pdf

One guess is that our brains are "lazily" allocating neurons to tasks, since we have so much space available compared to most animals. The signals are simply allowed to branch out / propagate much further (and redundancy is probably involved too). The paths are simply built where the signals goes.

But when it is constrained, and there's still enough of all the important brain structures, the density (interference?) of different signals probably causes them to propagate much less such that the processing of each signal will get processed by fewer neurons, and the same functions are still achieved.

Kind of like a space optimized FPGA design vs a lazy one that uses all the logical elements because there's no need to be compact.

In our neighborhood we see quite a few corvids, crows, scrub-jays and Steller's jays. The jays in particular are entertaining. They are very fond of raw peanuts in the shell. When the birds are visiting, tossing a few peanuts leads to observing bird behavior not unlike that of human primates.

They get into "arguments" with each other over possession of the nuts. Sometimes they seem more interested in "fighting" than eating. So they are sort of competitive with each other, yet they also cooperate, since most of the jays wind up with the opportunity to grab a peanut or two.

A few individuals have physical deficits: one's missing part of its lower mandible, another one has only a right foot. Yet they adapt to these handicaps figuring out ways to get a share of the peanuts and other goodies we supply.

Reading about avian neuroanatomy shows very differently organized brains. Of course their brains are smaller in absolute size vs. primates. I've wondered if the nuclei-organized avian brain is a more efficient "processor" vs. mammalian layered brain. Maybe that accounts for the surprisingly excellent corvid performance.

Glad you posted and I like how you describe the adaptations and determination of that particular group. Good notes! Regarding your last point and line of thinking, I've been studying a lot of flight processing resources and that might also be a characteristic of their brain structure. Being able to use the motor skills and air conditions for flight - to me - seems like it's more complex than walking around on two legs. Fun to think about!

You should talk to an experienced skydiver about what it's like to operate in 3 dimensions. I skydived long ago, and while I never got really great at it, I remember doing things "instinctively" to make or avoid something happening. For example, dipping a shoulder, especially without thinking about it, to move somewhere or out of the way of someone.

I guess bicycle riding is another example of making things happen in unnatural or differently constrained environments.

Also playing first person games. It is a weird feeling when the buttons "disappear".

Musicians sometimes talk about freeing themselves from their instruments.

I'm not really familiar with the intricacies of flight processing, I think you mean the neuromuscular mechanisms used by birds to accomplish flight. It is amazing to watch, though it's interesting how birds vary in coordination ability. Our jay friends aren't as graceful as other species, like the northern flicker, when landing on the suet cage. It can take 2 or 3 tries for jays to find the right balance point.

My thought is that the birds with superior problem-solving skills are accomplishing it with a much less massive brain that primates, perhaps their neural arrangements are more space efficient, packing more processing power into the available volume by having less redundancy of "wiring" or more direct interconnections among "subprocessors" vs. the layered organization of mammalian brain.

It's sort of like the discussions around computer/processor architectures. The standard von Neumann organization may not be the most efficient and alternatives are possible. Not that I know a lot about it, but seems to be an idea that parallels avian/mammalian comparisons.

Ravens are just better programmed for obtaining difficult to access food (makes sense, see insects).

Compare human and goldfish intelligence according to whom better understands tiny currents and you'll find the goldfish wins every time, even against oceanographers.

I too welcome the coming of our supercavitating goldfish overlords.

Chimpanzees are known to 'fish' for termites, which is a fairly complex behavior involving a hidden food source, so I'm not sure the test is obviously biased toward ravens (unless they also administered the Raven's Matrices). It would be interesting to see if there were any sex-related differences in performance, as it is female chimps who do most of the insect hunting, and teach it to their daughters).

> Chimpanzees are known to 'fish' for termites, which is a fairly complex behavior involving a hidden food source, so I'm not sure the test is obviously biased toward ravens (unless they also administered the Raven's Matrices)

Caledonian crows fish for/spear grubs (hidden in wood) using small twigs, fashioning them into hooks when the sticks aren't naturally barbed.

Chimpanzees are also known to cooperate when hunting monkeys.

Experiments have been done that control for this. Here's an impressive video of a Crow solving an 8-step puzzle to get a reward.

http://fb-192.dailymegabyte.com/wild-crow-shows-incredible-i...

It's more than just digging and pecking.

I didn't say it's just digging and pecking.

What I'm contending is that the conflation of complex behavior with intelligence is anthropomorphization of the worst kind.

1. We are aware of very complex systems that are definitely not conscious, or even animal. (See natural selection, robots)

2. We're also aware of complex behavior which is probably not conscious, given the lack of a proper brain (see Bees, Ants, and other insects).

3. We are aware of very complex behaviors that humans may do without being conscious of them (unconsciously answering questions, sleep walking/talking, complex muscle-memory activities like driving, doodling).

So, given the wealth of data showing that complex behavior can be completely unconscious, in the absence of proof we ought to remain skeptical. When claiming that a complex behavior is conscious or intelligent behavior, even among apes, the burden of proof lands on the person making the claim.

I've yelled at a squawking crow every time it cawed for a few minutes. After that I stopped and hid from the crow under various rooftops and the crow would fly around to get a visual vantage point on me (still cawing of course), I would move out of its visual range, it would fly to a new position to spot me. It was eerily fun.

"Study suggests we don't yet know how to design meaningful cognitive tests. Ravens score just as high as big-brained chimps."

intelligent birds use a structure called the 'visual wulst' for reasoning tasks, a highly-evolved brain 'bulge' used for making decisions about visual stimuli.

It's in some ways different from our cortex but similar in that it's a recent structural addition, under selection pressure and impairs learning when damaged.

There may be some truth to the author's claim that 'absolute brain size is less important than relative'. Spider brains are so important that some small jumping spiders have neural tissue spreading into their legs for lack of space. (Not the same phenomenon as octopi with distributed brains in their arms; spiders have one large central brain). And jumping spiders have a reputation for learning / reasoning smarts.

At first I thought it meant the Baltimore Ravens and I was like, "No way..." but now I see they mean those birds and that makes more sense.

Corvids have a higher average Wonderlic score than the Baltimore Ravens and fewer DUIs.

Anticipating such tests to be on the horizon the Baltimore Ravens drafted this guy: https://en.wikipedia.org/wiki/John_Urschel

I'm a bit confused "forgo a quick reward in favor of a bigger one that comes later" but as I read it, it was smacking head against glass or not?

Yeah, that was a bit awkwardly written. I tried looking up more info on it but got worried the article itself was some kind of meta test to measure human intelligence and decided to just move on...

> [Animal a] could do [task]. [Animal b] could do [task].

Ok, but it doesn't follow that they are similarly intelligent. You may have just picked a too easy task to distinguish.

They did not tested if the animals were intelligent or not, that was a known fact. They seem to have tested the intelligence relative to different species (instead of testing the intelligence absolute limit) and it wasn't an easy task at all because even primates like gorillas failed more at it!

If you click through one of the links in the article, you'll see a couple crows instigate a fight between two cats. I was wondering if they were trying to get one cat to kill the other, so there'd be something to eat. That'd be some long term thinking.

Also, honeyguides are birds that guide humans to beehives, so the humans will open the hive for their own purposes. The birds then eat the leftovers. https://en.wikipedia.org/wiki/Greater_honeyguide

I never knew how smart they were until I saw them at a small zoo and the keeper gave a talk.

Apparently they will setup bait in their cage for other smaller free birds so they can capture and eat them.

They also sometimes trade their treats with birds on the outside. Thug life.

> The Corvus species performed on a similar level to the great apes, despite vastly smaller absolute brain sizes. A chimpanzee brain is roughly 26 times larger than a raven's; nevertheless, both species achieve 100% success.

I don't think that you can conclude from that that ravens and chimps have similar intelligence. Rather, you test was too easy. You need some level of failure to be able to measure a difference in performance.

For some reason, not being a native English speaker I thought this referred to the Baltimore Ravens, and not to the animal of the same name...

Even worse, Raven's is an intelligence test.

https://en.wikipedia.org/wiki/Raven%27s_Progressive_Matrices

Conveniently next up on my Kindle reading list is a book about why we should look at the number of neurons in a brain rather than the brain volume when thinking about how intelligent a creature is.

https://mitpress.mit.edu/books/human-advantage

AI has reached peak hype. I think it's time to start an uplift breeding program to create intelligent animals. I say we start with ravens or octopuses.

http://www.thecrowbox.com/

> The Crow Box is a device designed to autonomously train crows.

   Except, as the researchers on this new study point out, 
   the original study did not test corvids, known to be among 
   the most intelligent birds.

There has to be more to it than this. How could supposedly-intelligent researchers make an error like that?

They are just big apes in the end...

Prior to clicking the link I really thought this would be about concussions in the NFL Baltimore Ravens. I chalk that up to a working "raise awareness" campaign.

Nevermore.

It is best to give wildlife some money, take them to the shop and let them pick what they like.

I would say some form of universal income should do.

We detached this subthread from https://news.ycombinator.com/item?id=11540734 and marked it off-topic.

Reply on news.ycombinator.com