Wolf Pups That Play Fetch | Hacker News Reader

Wolf Pups That Play Fetch

nytimes.com

83 pointsHooke45 comments

I always figured that the “playing fetch” behavior was just the dog (or wolf, I suppose) figuring out that it gets to chase the thing again if it brings it back to the thrower. If you teach a dog to play fetch, it has no problem instinctively chasing the thing. It’s the bringing it back that has to be trained. But once the dog gives it back, it suddenly learns that it gets to chase the thing again, which is fun and rewarding. Eventually you might be minding your own business but as soon as the dog finds the ball, it will bring it to you. The dog plays fetch because you’re going to throw the ball, not because you already did throw it.

Yeah. My cat will actually bring the laser pointer to me so that I play that with her. She seems to understand that it is connected somehow to the magic red bug that she likes to chase.

Same here. One of my cats will do that if given the chance. All of them know where the laser is kept, and order breaks down whenever I touch that drawer.

Dogs are frequently trained to retrieve game, which is never thrown again for their benefit (though some might be given to them to eat). Your model would say that this is unrelated to playing fetch.

But people also train dogs to do all kinds of things. They'll do whatever you want for no other reason than that you want them to do it. (With limits based on the dog's personality and intelligence.)

It all ties in. A lot of dogs are specifically bred for retrieving behavior. (See: "golden retrievers", but newfoundlands are bred to retrieve _people_ from the water, too.) A lot of dogs like interacting with their humans. Both of those are sufficient for many dogs to play fetch. A single-factor model where the reason to play fetch is that you like chasing things misses a lot of reality.

I don't know why my Australian Shepherd doesn't really play fetch well. But he'd much rather play interactive games with me or my wife where he and one of us both actively participate in the fun, which usually involves me playing tricks, misdirections, and sleight-of-hand with his other toys (which requires him to think).

We were warned of this sort of demand / expectation of the breed and it hasn't bothered us. But it's funny that he knows so many commands (verbal and nonverbal) and we just failed to teach him how to fetch.

Shepards aren't retrievers.

But they do like keeping track of things / people and keeping them "herded" and collected.

Not a failure of teaching, the breed just isn't programmed for it.

Yeah. Most "utility" dog breeds have been created simply by selecting for aptitude and propensity for a specific aspect of ancestral pack hunting behavior – tracking, stalking, pursuing, herding, catching, killing, or bringing kill back.

I suspect my rescue mutt has a lot of shepherd breed in him and he’s the same way. He gets extremely agitated if either my wife or I walk away from him. It’s almost impossible to get him to focus on any games, work, or even food if one of us has walked away. (He’s okay with one of us leaving him at home, but it triggers if one of us breaks off from him outside.

I suspect the herding behavior is probably bred/trained based on a highly overdeveloped impulse for wanting to keep the “clan” together.

My dad hunted ducks so I grew up with retrievers (Labradors), I just didn’t want to get too deep into the weeds since actual trained retriever behavior is a lot more advanced than just playing fetch. Though like any sort of behavioral conditioning, the hook that gets the animal initially started with the behavior (getting to chase a thrown toy) can be gradually changed to the desired behavior (retrieving game). My dad actually had a training tool for retrievers, it took a blank .22LR cartridge and launched a foam dummy for the dog to fetch. That way the dog gets accustomed to the sound of the gun going off and starts learning that it can be related to the retriever behavior. Eventually you get the dog trained so when you shoot the duck, the dog will wait patiently, you signal the dog to fetch the duck, and the dog brings the duck back to you. Retrievers also have the incredible instinct of carrying game in their mouths very gently.

But yes, dog behavior is much more deep and interesting than that. If you’ve ever been at a barbecue with herding breeds, even if they aren’t really trained they still instinctively don’t like letting people wander off, for example. But I think the chasing game is probably the primary hook that gets most dogs into playing fetch.

I did search and rescue training with dogs in the army and we always had to start things off with a toy to get the dogs "into it", even in "sharp" situations. But once they were started everything was like a game for them, they'd run ("play") until they couldn't anymore. You'd have to stop them.

If you're ever in a SAR situation and a dog finds you, you should be prepared for a little rough treatment from it. It'll likely jump all over you, be up in your face and all that. It's just looking for the toy, and when it find out you don't have it, it'll move on, calm down or go back to the owner. And it'll bark a lot. Like a lot. It's not out of anger or anything, it's just what it has been told to do when it thinks it has found what it was supposed to find. I've had a dog tear my entire goretex uniform to pieces looking for the toy when we were both trapped deep under snow. Good times!

What would happen if you coincidentally had a dog toy on you?

Maybe that’d be good for avoiding accidental mangling, but if it means it skips the barking and alerting step then maybe it’s not the best idea

They're looking for a few very specific toys, so that won't really be a problem.

For the most part, SAR dogs (and similar working dogs) are trained to view specific toys as incredibly high-value. For whatever play they really enjoy - fetch, tug-of-war, etc - they'll have a particular toy which only comes out in connection with "work" and always heralds praise and a play session with their trainer. Since SAR breeds are smart, high energy, and a bit intense, their reaction to that toy/event can be startlingly strong.

The goal is essentially to avoid this issue: a dog toy is unlikely, but maybe the subject has a peanut butter bar in their pocket, and that shouldn't derail the rescue. So ideally the desire for the work reward will outweigh basically anything found during work (squirrels, food, other toys, or even bad stimuli like loud noises), even if those things would be awesome to the dog when they're not "on the job".

I have 2 dogs, Sissy the older lab mix, who loves playing fetch all the time, and is in general possessive of the balls. But she'll chew them up. My other dog Melkor, who's a lot younger. I have to get him wound up before he'll like to chase the ball. But he's not that good at fetch. One thing I found he likes if he's on the dog bed, and I throw the ball up towards him, he'll catch it, let it go and I'll have to get it. I should train him more.

Who's playing fetch with who, then...?

The dogs trained me quite well ha.

It’s interesting that many cats, too, play fetch completely intuitively, no training required.

Interesting, I had never seen that before until I read this comment and went googling!

I don't know if it requires training, but here's a red fox playing fetch: https://m.youtube.com/watch?v=EchdjzFO-7U

My current cats won't, but I've had cats in the past that enjoyed playing fetch.

Beluga whale playing fetch: https://www.youtube.com/watch?v=7uMSl28_8Rw

Burmese cats are famous for playing fetch and behaving a bit like cat shaped dogs.

My cat is weirdly smart and he will play fetch. He's really good at figuring out whatever new "game" I'm trying to play with him. Some more examples of his social intelligence:

- He will play tricks on me. He's learned to paw at the patio door to show that he wants to be let outside. If I'm standing outside and he's inside the patio door, he will paw at the door as if he wants to be let out, and then run away as soon as I open the door. He will do this over and over and turn it into a game.

- He's VERY vocal and has some understanding of how a conversation works - I talk, then you talk. I only realized this because he hates car rides, and this is the only time he interrupts me.

- He's really good at understanding tone and body language, so usually he can figure out what I'm trying to tell him to do, even if it's a new situation using new words. For example, I was outside while he was playing, and a neighbor walked up to say hello. I told him "it's ok, you can go say hi" and he walked up to her and sniffed her hand.

It's really weird, I knew cats were intelligent but I had no idea until I got this cat.

Interestingly, my cat started to bring pieces of her food to me so I can throw it and she can pretend it's live pray she is catching. This behavior emerged suddenly out of nowhere without prior training.

I bet many pet owners can share same anecdotal evidence of certain level of social intelligence which, I think, is not widely accepted in scientific community.

Maybe related to "Belyayev's Fox Experiment": https://news.ycombinator.com/item?id=22143748

Non-paywall: http://archive.ph/5OBeO

Interesting question here: I can resolve archive.ph just fine, but archive.today doesn't work due to the ongoing issus with Cloudflare DNS (1.1.1.1). Anyone know why?

https://news.ycombinator.com/item?id=19828317

Trim link if your employer like mine blocks some of the archive sites. https://beta.trimread.com/articles/3236

I'm going to guess that the "fetch" reflex may have its roots in wolves chasing after bones humans around a campfire may have tossed to sate the wolves' appetites and keep them at bay.

Maybe bringing the bones back was a way of "nicely" letting humans know that more bones would be appreciated/required to keep living.

You often see wolves circle around humans around a fire in movies and books.

Maybe it's a symbiotic relationship like a remora/shark, or even parasitic like an oxpicker/large mammal.

*edit - Some people are very liberal with downvotes, especially for lighthearted conjecture that is not obviously wrong or intellectually misleading. Got any better ideas?

You're probably being downvoted for telling a just so story: https://en.wikipedia.org/wiki/Just-so_story . It's a particularly pernicious trap when trying to understand behavior and evolution!

That makes sense.

No, it doesn’t. Your comment was fine. If your comment was a “just-so story” then the first two paragraphs of the linked article are also a “just-so story”. I see no evidence that the scientists were using guaranteed purebreed wolves, if there even is such a thing alive today. During the evolution of dogs, surely wolves in general were constantly having their gene pools contaminated by genes from stray dogs. That could easily explain why some wolf pups would exhibit traces of dog-specific behavior. Oops, and now I’ve just told a “just-so story”. Maybe it’s safer if we turn off our brains and never comment again.

I meant it makes sense as a reason for why people are downvoting.

I don't think it makes sense as a way of thinking about things. And I certainly won't stop commenting because some people find it disagreeable/distracting.

I’d never heard about this before, so I read the Wikipedia article.

> Academics such as David Barash say that the term just-so story [...] is simply a derogatory term for a hypothesis.

I’m inclined to agree with Barash. Conjecture should not be confused for fact. However, conjecture is also necessary for discovery.

A Just-So story cannot be a hypothesis since a good hypothesis must be testable for actual science to be done later.

In contrast, a Just-So story will never be followed up on and simply exists to supply an easy to digest explaination for a phenomena. An explanatory meme so to speak. Hence their comparison to fairy tales via titles like "How the Tiger got it's Stripes"

Exactly. When I see just-so story used as a criticism, it's almost always because something isn't falsifiable, and more specifically, it's usually an argument from ignorance. The explanation slots into a space we don't know much about and defends itself with "it's possible and it makes sense", but its strengths are almost always narrative rather than factual. Like Kipling's actual Just-So Stories, they're interesting and memorable, but very dissimilar to the histories we actually know.

(As far as why this specific example qualifies, it's basically impossible to falsify the story, yet it's badly out of sync with what we do know. "Wolves circling the campfire" show up in books and movies, not in real life; even absent modern tools, wolf attacks on humans are rare and reliably come from desperation attacks on isolated people, exactly the wrong circumstances for this story. Keeping predators at bay by offering food that won't sate them is the exact opposite of how things generally work; it's the equivalent of tossing out chum to stop sharks from attacking. And we have a pretty good model of how canine domestication in general began - it does involve bone and meat scraps, but scavenged from waste, and with wolves scaring away other predators, not spontaneous cooperation.)

I really hope “just-so story” isn’t going to be the new thought-terminating cliché we see everywhere.

Slippery slope fallacy!

The just-so story is itself thought-terminating, insofar it presents a conjecture as that's just what it is (it's just so), without asking the right questions to connect to.

The story itself is not the thought-terminating cliche, labelling it as such (especially with no attempt to engage or correct the poster of it) is.

This is essentially an important skill for hunting animals. Some trainers call it “prey drive” or “chase instinct”. It’s all about their need to hunt and catch prey. Bringing back is a way to repeat the game, which is rewarding. Pretty sure that is the “understood” reason for this behavior.

Oh, cool!

The article doesn't mention anything like a prey drive, and seems to assume fetching is just a way of playfully interacting with humans. Seems like an oversight.

I'd say it's much simpler than that. In nature things don't fly across the ground unless they're alive, prey.

So a ball or stick flying/rolling across the ground mostly immitates a small prey animal.

I believe that is what our training is nurturing.

I have a jack russel and she is genetically predisposed to hunting after small things.

That makes sense. Why would a wolf pup bring it back though?

I wouldn't be surprised if wolves that brought back food/bones thrown by humans were then given more food, making those wolves a bit better off, fostering that trait.

Reply on news.ycombinator.com