A Dog Can’t Teach a Dog New Tricks (But It Can Teach a Wolf)

By Elizabeth Preston | February 7, 2014 9:05 am

dog opening box

In a dirt-floored room in Austria, a puppy sniffed and pawed at a wooden box with a treat inside. It circled the box over and over, unable to find a way in. Finally it sat at the feet of a nearby human and looked up at her appealingly, swishing its tail. The woman stared at the ceiling, ignoring the puppy. The answer wouldn’t come from her—if the dog wanted the treat, it would have to imitate the way it had seen another dog open the box earlier.

The puppy ultimately failed, as did most other dogs in this experiment. But all of the wolves that tried it succeeded easily. The reason may be that in creating domestic dogs that pay attention to us, we robbed them of the ability to learn from each other.

Friederike Range and Zsófia Virányi carried out this study at the Wolf Science Center in Austria. The two animal behavior researchers wanted to know how well dogs and wolves can learn from their own kind. We’ve bred dogs to be excellent at learning from us. But their ancestors were wolves that lived in packs and relied on each other; have their skills from that time been lost?

The teachers for both groups of canines would be domestic dogs, which are easier to train. (The authors figured the wolves would see a dog as basically one of their own. And because wolves and dogs cross paths regularly at the research center, both animals were comfortable having each other around.) Range and Virányi used a simple puzzle box that opened with the push of a lever. They trained one dog to open the box by pushing the lever with its paw, and another dog to use its mouth instead.

Then it was time for class. The pupils were 12 wolves and 15 mutts, all just six months old and raised in the Wolf Science Center. On the day of the experiment, a dog or wolf subject was brought into a room with the puzzle box in the center. While one researcher held the subject to the side of the room, the other researcher held one of the teacher dogs. When she told it “Box!” the teacher dog trotted over to the box, opened it with its preferred method, and returned to her side. Then the first researcher walked the student dog to the box and let it eat the treat inside.

After the student had seen this demonstration (and eaten a treat) six times in a row, it was let off its leash. The time had come to apply what it had learned—or not, as it turned out.

Here’s a dog trying to open the puzzle box on its own (the lever is on the right):

dog 1 dog 3

You can watch the whole video here if you want to see a heartbreaking five minutes of sniffing and tail wagging. Only four of the 15 dogs managed to get the box open; just two of them could do it a second time. And none of the successful dogs used the technique (paw versus mouth) they’d been shown. Even if they’d learned from the teacher dog that the box was openable, they didn’t copy their teacher’s method.

Wolves, however, were star pupils.

wolf 1 wolf 2

Every one of the 12 wolves got the box open on its first try. Nine of them matched the method they had seen their teacher use. Most of the wolves could repeat the trick over and over again.

Range and Virányi checked for other explanations besides the unflattering one that domestic dogs are just not very clever at learning from each other. Were the dogs too young? Wolves develop more quickly than domestic dogs. So the authors repeated the experiment when the same dogs were a year and a half old—but they did no better. Were wolves naturally skilled at opening the box, without needing to be taught? A control group of wolves tried the puzzle box without watching the teacher first; they failed.

Nor was the answer that wolves were bolder or more curious about prodding the box. Dogs and wolves were equally quick to run up to the box and even to sniff around the lever. If anything, wolves were more cautious; one of the wolves was so scared when the box opened that he refused to try a second time.

The only remaining answer was that dogs didn’t learn from each other as easily as wolves learned from them. If the dogs at the Austrian research center are representative of domestic dogs worldwide, this may mean that dogs lost their power to pay attention to each other when we bred them to pay attention to us.

Part of what makes a dog a dog is its eagerness to cooperate with humans. That skill, the authors write, may have started out as eagerness to cooperate with the pack. As we domesticated dogs, we would have selected and bred the ones that could best turn their focus to us. We cared less about how well they learned from each other, so that skill fell away. As we made them into our best friends, then, we may  have forever turned them away from their old companions.

Images: Range and Zirányi.

Range F, & Virányi Z (2014). Wolves are better imitators of conspecifics than dogs. PloS one, 9 (1) PMID: 24489744

  • Joan_Savage

    Did the experimenters mention how the dogs and wolves had been raised from birth? That might clear up a possible imprinting difference. If the wolf mothers had been allowed to raise pups privately away from human contact, the wolf pups could have a preference for a canid role model. If the dog pups were handled by humans while they were still very young, they’d have a mixed history of trusting humans and dogs-that-look-like-mom. If both the wolves and the dogs were hand-raised from birth by humans, then it would look like a cleaner experiment about species, and not about primary trust bonds.

    • epreston8

      The dogs and wolves were both hand-raised at the center.

      • Joan_Savage

        Thanks very much, that takes care of it, I guess!

  • duelles

    Border collies will learn from the older more experienced BCs how to herd and all the herder needs to do is instill commands or signals to go along with it . . . For his/her use.

    • Sinamon

      might be hard to compare BC’s to other domestic dogs anyway as they are the brainiacs of the dog world :-)

      • duelles

        Yeah! I lucked into having one. A two year old male, and am amazed about every day at the focus. But I needed to step up my game to keep him active, occupied, learning and happy. He may keep me young for the next 10-12 years.

        • Sinamon

          kudos to you! i am well-aware of my inability to challenge a BC enough so i don’t dare own one. they are too smart for me but i admire them so much (and the owners who do their best to keep them engaged).

  • http://www.hooft.net/rob Rob Hooft

    Very interesting. Could it be that dogs became more domesticated because they lost the ability of learning “bad habits” from their peers?

    • duelles

      The silver fox experiment from Russia might explain the domestication of foxes was through the breeding of juvenile traits back into the line. The most playful, friendly were bred with each other and became floppy eared, with puppy coats not useful for fur!!!

  • jason

    very nice trick.

    satellite tv for pc

  • George Watson

    I will say that I have seen my Boxer exit our yard after observing and learning from our neighbor’s dog as to how to enter our yard. He watch several times and before we knew it he was traveling back and forth with ease until we sealed the hole in the gate. I don’t know if this is a comparable analogy but he did learn from another dog. What are your thoughts?

    • epreston8

      Interesting! There have been other studies that suggested dogs might be able to imitate each other, but maybe only in certain situations (or certain dogs). It’s not a closed question, anyway, and there are lots of variables that could be involved (such as the breed of dog, how it was raised, etc.).

  • http://blogs.discovermagazine.com Terry Simpson

    When you bring a new dog into a household ‘pack’ of 2 or more dogs, the new dog DOES learn from the other two. They learn about going outside to do their ‘business’, they learn about sitting and waiting for their meals to be delivered to them, they learn about playing ‘hide and seek’ both indoors and out, and they learn a number of other things. But they act as a pack, not as individuals.

  • natsera

    My 7-month-old miniature dachshund learned to jump into my hands after observing my 5-month-old dachshund/rat terrier mix do it. She had never exhibited any offer to do that behavior before observing him do it. So I think this calls for more experimentation in different environments before you can make a hard and fast statement.

  • Laura Spindler

    Total garbage! Just one more NOT scientific study calling itself a scientific study that results in a bunch of crack head theories! It says they used “mutts”. and wolves that were all raised at “The Wolf Science Center” (no bias going on there!) were they all raised the same way? Were the dogs kenneled their whole lives? Were they raised together as wolves in a pack are raised?…was there a control group? Can these reactions be duplicated in another study with other dogs and wolves? Holy crap, please stop printing this swill!

    • Elizabeth Preston

      The Wolf Science Center treats dogs and wolves exactly the same so that they can try to tease out actual biological differences in the animals. All the puppies are hand-raised, and the animals live in packs.



Like the wily and many-armed cephalopod, Inkfish reaches into the far corners of science news and brings you back surprises (and the occasional sea creature). The ink is virtual but the research is real.

About Elizabeth Preston

Elizabeth Preston is a science writer whose articles have appeared in publications including Slate, Nautilus, and National Geographic. She's also the former editor of the children's science magazine Muse, where she still writes in the voice of a know-it-all bovine. She lives in Massachusetts. Read more and see her other writing here.


See More

@Inkfish on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar