« New issue of Econ Journal Watch | Main | It is worse than you think »
Dogs can, monkeys and wolves can't
The researchers held two containers, one empty and the other containing food, in front of chimpanzees and dogs. Then they pointed to the correct container. The canines understood the gesture immediately, while the apes, genetically much more closely related to humans, were often perplexed by the pointing finger.
That's not all. Many dogs were even capable of interpreting the researcher's gaze. When the scientists looked at a container, the dogs would search inside for food, but when they looked in the direction of the container but focused on a point above it on the wall, the dogs were able to understand that this was not meant as a sign.
Puppies seem able to do this before they have been socialized with human beings.
Says one researcher: "The great advantage of dogs is that we can study them in their natural habitat without any great effort," explains Adám Miklósi.
Here is the full story, hat tip to Mark Thoma.
Posted by Tyler Cowen on September 10, 2007 at 06:46 PM in Science | Permalink
Comments
Might be interesting to contrast wolves with dogs on this test.
Posted by: Alex J. at Sep 10, 2007 7:29:11 PM
Dogs understand sarcasm too. Really.
Posted by: Brad Hutchings at Sep 10, 2007 7:32:04 PM
Interesting. These dogs are obviously a lot more clever than mine, who, when I point, think I am drawing attention to my finger rather than to whatever is at the end of my finger's trajectory.
But I wonder whether the experimenters were careful enough to blind themselves as to who is observing them. Did they, for example, have the observing animals stand behind one-way glass, so that the actors don't know what species the current subject is?
Posted by: zlguocius at Sep 10, 2007 7:36:42 PM
I'd like to see the experiment done again using cats. I don't think the cats would be fooled. Cats are awesome.
Posted by: Frank at Sep 10, 2007 8:10:25 PM
Sounds like conditioning instead of natural smarts to me. Pavlov anyone?
Posted by: Chewxy at Sep 10, 2007 8:26:47 PM
Dogs are much better at reading people's intentions, after thousands of years of history of living with people in symbiotic relationships.
Posted by: Matthew C. at Sep 10, 2007 8:40:14 PM
Um, I don't get the experiment, maybe because I'm more like the ape than the dog.
Posted by: Paul N at Sep 10, 2007 9:42:19 PM
How did the experimenters control for the dog's sense of smell?
Posted by: Alex at Sep 10, 2007 9:45:55 PM
How did the experimenters control for the dog's sense of smell?
If the dogs were identifying the food by smell they would have gone to it before the gesture.
Posted by: Bernard Yomtov at Sep 10, 2007 10:23:19 PM
Alex J,
Dogs and wolves have already been compared on similar tests. Dogs look to humans when dogs face some obstacle in the way of reaching something they want. Wolves do not. Natural selection at work. Dogs have evolved to live with humans.
I've seen dogs look to me for help and come to me for help many times. On one occasion my Australian Shepherd came into my bedroom looking at me because he was choking on something half swallowed. I pulled out a rawhide chew and probably saved his life.
Posted by: Randall Parker at Sep 10, 2007 10:37:25 PM
I think we need to run an experiment where we tell people that there is $20 in one of two containers and
that they can keep it if they pick the right one. Then the experimenter simply points to the container
with the money in it. It would be interesting to see how many people actually pick that one. My guess is
humans would do better on average than monkeys but worse than dogs and we would do worse because we expect
people to mislead us.
Posted by: Ronald Brak at Sep 10, 2007 11:03:32 PM
Frank,
It's unlikely cats would do as well as dogs. Dogs have been with us for tens of thousands of years, since hunter-gatherer days, serving as hunting companions and a poor man's perimeter alarm security system. Cats, on the other hand, have only been with us since the dawn of agriculture, when crops and granaries started attracting vermin that needed to be exterminated. Dogs have had an order of magnitude longer time to adapt to us and our nonverbal cues and body language, and evolve accordingly.
Posted by: at Sep 10, 2007 11:34:49 PM
Not to mention dogs are social animals unlike cats, so they "get" the idea there are other beings besides offspring to interact with on a long-term basis
Posted by: barjac at Sep 11, 2007 12:45:11 AM
The experiment is about animal behavior: what dogs, monkeys and wolves do and don't do. Why do you cast it in terms of "can" and "can't"?
Posted by: rkillings at Sep 11, 2007 3:36:08 AM
The experiment is about animal behavior: what dogs, monkeys and wolves do and don't do. Why do you cast it in terms of "can" and "can't"?
Posted by: rkillings at Sep 11, 2007 3:36:33 AM
The experiment is about animal behavior: what dogs, monkeys and wolves do and don't do. Why do you cast it in terms of "can" and "can't"?
Posted by: rkillings at Sep 11, 2007 3:36:38 AM
Why do you cast it in terms of "can" and "can't"?
Because the wolves and apes clearly wanted the food, so if they had understood the gesture, they certainly would have picked the correct container. But they didn't.
Posted by: Slocum at Sep 11, 2007 7:08:52 AM
Wolves, like dogs, are social animals, only more intelligent. I am confident that wolves which have interacted with humans would "get" the gesture.
Posted by: triticale at Sep 11, 2007 8:04:08 AM
I am confident that wolves which have interacted with humans would "get" the gesture.
Or, wolf eat puny human.
Posted by: Independent George at Sep 11, 2007 8:25:11 AM
Is this perhaps a good reason that dogs, in particular, are undeserving of humans inducing cruelty upon them (relative to other animals (cows, deer etc)?
Posted by: nelsonal at Sep 11, 2007 8:51:47 AM
Did anyone catch this story on human skull evolution? The researches claim that a lot has changed in only 650 years. If dogs have really learned a lot, is the Planet of the Apes scenario plausible?
Posted by: 8 at Sep 11, 2007 8:52:06 AM
I wonder if they really pointed from a distance, or nearly touched the objects with their finger tips. I can get my dogs to sniff at anything by touching it. In any case, dogs communicate even with each other through subtle changes in posture and eye vector. Picking up larger gestures should be no problem for them.
Posted by: Some Random Economist at Sep 11, 2007 9:32:28 AM
To triticale: Did you actually read the article? Wolves aren't quite as good as dogs at understanding humans. They raised wolf pups for several months, treating them as dog puppies. They then conducted the container test. When the container with the food was closed, the wolf pups tried to force their way into the container. The dog puppies immediately gave up and stared at the humans.
Posted by: DM at Sep 11, 2007 10:47:31 AM
When I came home the other day I knew someone had been using my machine but the damn logs had been erased so I couldn't see which sites had been accessed. I finally remembered to check my DNS lookup log. Rover's not as smart as he thinks he is. Bad dog, you should not be accessing www.hotbitches.com, or at least you need to know more about Linux and forensics.
Posted by: Loki on the run at Sep 11, 2007 1:19:47 PM
When I came home the other day I knew someone had been using my machine but the damn logs had been erased so I couldn't see which sites had been accessed. I finally remembered to check my DNS lookup log. Rover's not as smart as he thinks he is. Bad dog, you should not be accessing www.hotbitches.com, or at least you need to know more about Linux and forensics.
Posted by: Loki on the run at Sep 11, 2007 1:20:41 PM
