Aside from humans.
Are there animals which do this with no training? Are there animals which do this between themselves or have analogous behavior?
Dogs do it instinctively. Other great apes can be trained to do it, but only with great difficulty. No other animal gets it at all.
Wolves or recently domesticated wolves do it? Wolfdogs?
I’ve always heard this about dogs, and none of mine have ever demonstrated this capability…
I don’t think my dogs were ever good at it, they’ll stare at my finger. But there are a lot of breeds with different behavior and a lot of individualized behavior in dogs. There aren’t many generalizations to make about complex dog behavior.
Dogs definitely have this capability and it’s really unique, possibly unique among all species, though I did come across this article claiming that elephants can do it, too. Not sure that I really believe that, though.
Doubtful about wolves as dogs’ characteristics in that regard, and in many others, comes about from long and close association with humans.
I can’t remember where I read this, but I believe ravens can be trained to follow human gestures as well. The cites I did find say they point for each other using their beaks, and can be trained to follow the gazes of human surrogate parents.
Better than that, though, apparently elephants can understand your pointing.
No cite, but what I’ve read in the past said it’s unique to dogs and wolves just ignore you, even if reared by a human. Not sure about wolf-dogs and other canids.
Also, see A comparative analysis of animals’ understanding of the human pointing gesture (PDF).
It says that dogs, wolves (despite my earlier uncited belief), cats, monkeys, dolphins, and seals all have some success in understanding human pointing gestures. Their success rates are modified by things like how long the arm is held out, whether eye contact is first established, whether humans gaze at the target, and whether humans are physically close to the target. But in most of their trials, animals did better than chance at understanding the point.
Edit: As for dogs versus wolves, this paper says it comes more naturally to dogs, but wolves can be (with effort) trained to follow too.
I think this is a key issue. A lot of animals, especially those with strong social behavior, will look in a direction if they see other animals looking in that direction. But following just a pointing finger is a much more complex action.
Yes, but it’s not the ONLY factor. In the study I linked to, if I’m reading it right, seals and dolphins don’t rely on your gaze very much and can still follow your finger. Primates actually seem to have more trouble with it, doing worse when they can’t see your eyes.
I think all dogs have the ability to do this but hide it from us. Just like hiding their thumbs so they can’t be made to open doors or use screwdrivers or pretending that they don’t understand what we are saying to them.
The are very much smarter that they appear.
Bob
My dog does it, but he has to witness the act of me extending my arm and pointing in a direction. If he had his head turned and then looked at me with my arm already extended and finger pointing, the thought wouldn’t occur to him to look in that direction.
Actually, what I think is really happening is when the dog sees the motion of the arm, whcih pretty much simulates the throwing of a ball (or whatever) and the dog instinctively looks for the non existing ball in said direction.
Something like the above, I think.
It’s fascinating to see human babies when they first discover and explore the communicative power of finger pointing. They’ll point at anything, or nothing at all in mid air, just to groove on it.
Dolphins can do this.
Karen Pryor, one of the founding principals and head trainer of Sea Life Park, Hawaii, tells of dolphins spontaneously learning to follow a pointing finger in her book Lads Before the Wind: Adventures in Dolphin Training (1975).
A group of scuba-diving researchers were doing something-or-other out in the open ocean, and took one of the dolphins with them to act as messenger and general go-fer. They trained the dolphin to work for tokens instead of fish, with the tokens being exchangeable for fish later on.
Out on the job, they used a grease pencil and greaseboard to write messages to one another underwater. To send the greaseboard or miscellaneous tools from one person to another, they gave it to the dolphin and pointed to the person who should get it. The dolphin took it from one person to another until someone would accept it and give a fish-token. By and by, the dolphin made the connection that the person pointed to was the person who would accept the item.
Have any scientists trained animals to follow a pointed finger in one context and then tried pointing at something in a different context? I’d be interested in the results. It would show if the animals had actually learned the concept of pointing or were just being trained to associate a certain hand gesture with a reward in one context.
I rather easily trained my black lab to go to where I was pointing (where there would be either a treat, or a toy)…
I sometimes feed both cats & the dog from the same hotdog or whatever they will both eat. If one of the cats decides he is full and leaves a chunk behind my foot or something, when the food is all gone, I just point, no need to look and the dog will go right to it.
So we are doing food and he knows that if I point, it means food for him. He does not even need to see me make the gesture. He just looks at me and if I am already pointing, he goes right to where I am pointing. He has always done this. But he is a quick study I must admit.
He has me well trained now.
The cats just made me their servant, did not even have to train me. They are telepathic.
Last summer (2013) I had made friends with a local Western Scrub Jay. There is a tree in my yard with multiple branches just above head level. I had got the jay used to me putting a peanut on a branch and pointing at close range to show where it was when it flew in for a treat. Gradually I backed up so I was pointing from six to eight feet away. The jay seemed to grasp my pointing just fine, as it would fly to the correct spot most every time. Gaze may have been a factor, but it did not fly until I pointed.
Yes, in one of the dolphin studies, they were first trained by pointing at food, but later trained to go fetch tokens (which could be exchanged for food).
So the reward mechanism is still there, but the dolphins had to learn that the reward would come later.
What I don’t understand is how the dolphins stashed their tokens. Did they invent an underwater dolphin bank of some sort? The study wasn’t clear.
I wouldn’t be surprised. All the corvids (jays, crows, ravens, etc.) are very intelligent – the smartest birds we know of, and among the most intelligent animals overall, as measured by their abilities to figure out puzzles, their ability to deceive one another, their rudimentary but culturally transmitted grasp of language, their social behavior, and also their brain-to-body-size ratio. And as mentioned above they do point for each other with their beaks.
IMO: What we fail to appreciate in animals is more often a measure of our arrogance, not their inadequacy.
Although above I suggested that the movement of the arm may be a factor, I’d just like to add that I trained my pup with a particular command associated with the “follow my hand and don’t just stare at it” (“this way” is the command). The reinforcement that this he is obeying a command–A great pleasure for any suitable dog–makes his understanding more repeatable.
For a while I wondered if his association was that “this way” means “go to that patch of on the right, that grass, and poop,” because that’s how I always used it. But indeed he had generalized the specific of the hand response, not the other (last) part of the command.
Retrievers, I would think, are supremely adapted for this, perhaps even more than other hunting dogs.