Dogs can tell canine size through growls

Dec 17, 2010 by Lin Edwards report
Examples of the looking behavior. A, Subject looking at the middle. B, Subject looking at the left picture. C, Subject looking at the right picture. D, Projections of dog pictures. Image: PLoS ONE doi:10.1371/journal.pone.0015175

(PhysOrg.com) -- New research has demonstrated that dogs can tell the size of another dog by listening to its growls. They are able to do this so accurately they can match the growl to a photograph of a dog of the same size.

Lead researcher Péter Pongrácz of the Eötvös Loránd University in Budapest, Hungary had previously demonstrated that have different growls for different purposes, and have a specific "this is my bone" growl. The new study tried to determine if dogs could identify the size of another dog from its growl.

Arrangement of the experimental room. Image: PLoS ONE doi:10.1371/journal.pone.0015175

The researchers recruited 96 dogs of various breeds and ages from the database of the Clever Dog Lab in Vienna. They presented a test group of 24 of the dogs with two projected images of a large (over 60 cm) or small (under 52 cm) dog: one life-sized and one of the same dog scaled up or down by 30 percent. They then played a recording of a food-guarding growl from either a small or large dog, with the sound coming from midway between the two images. They filmed the behavior of the dogs to determine which of the images they looked at first and for longest.

This video is not supported by your browser at this time.
Video sample of the experiment. The subject heard a big dog's growl, the matching picture was the left one (from the view of the dog), while the modified picture was bigger than the actual growling dog. The subject looked first and markedly longer at the matching picture, while observed the modified one for just a couple of seconds. Video: PLoS ONE doi:10.1371/journal.pone.0015175

The control dogs were shown either images of the silhouettes of large and small cats or large and small images of shapes such as triangles while listening to the recorded growls. All 24 dogs in the cat group were familiar with cats and either lived with a cat or had done so in the past. A final control group of 24 was shown large and small dog images while listening to Brownian noise.

All the dogs were well-socialized animals and were tested while their owner was present to encourage them to be comfortable in the laboratory environment. To ensure owners did not unconsciously affect their dog’s behavior they were listening to music via headphones throughout the experiment and were unable to hear the growls.

The results, published online in the journal , were that 20 of the 24 test dogs looked at the image of the appropriate-sized dog first and looked at it longest. The controls presented with images of triangles showed little interest in the images, while those shown of cats spent more time looking at them.

The control dogs looking at pictures of cats often looked first at the image on the left, a finding that supports earlier studies that have found dogs tend to look to the left when presented with something unexpected. No such bias was found in the other groups.

Pongrácz said the dogs’ ability to match the growl to the photograph of a dog of the same size is a "complex cognitive talent previously seen only in primates." The research also shows that dogs do not lie about their size, and this is the first time research has shown animals can determine another’s by the sound it makes.

Explore further: Dutch barnacle geese have more active immune system than same species in the North

More information: Faragó T, Pongrácz P, Miklósi Á, Huber L, Virányi Z, et al. (2010) Dogs' Expectation about Signalers' Body Size by Virtue of Their Growls. PLoS ONE 5(12): e15175. doi:10.1371/journal.pone.0015175

Related Stories

Dog 'laugh' silences other dogs

Dec 05, 2005

Washington state researchers report discovering what might be the sound of dog laughter. The scientists say the long, loud pant they recorded has a calming or soothing effect on the behavior of other dogs, ABC News reported.

Computer savvy canines

Nov 28, 2007

Like us, our canine friends are able to form abstract concepts. Friederike Range and colleagues from the University of Vienna in Austria have shown for the first time that dogs can classify complex color photographs and ...

The truth about cats and dogs

Sep 08, 2008

Thinking about adopting a perky little puppy as a friend for your fluffy cat, but worried that they'll fight -- well, like cats and dogs?

Penn State studies storm-phobic canines

Dec 15, 2005

Penn State University researchers have determined pet owners can't resolve storm phobia in their dogs, but having a multi-dog home may reduce stress.

Recommended for you

Nature offers video of 10 cutest animals of 2014

10 hours ago

(Phys.org)—The journal Nature has released a video that ventures a bit from its traditional strictly-science approach to technical journalism—it's all about the cutest animal stories of the past year ( ...

Big data and the science of the Christmas tree

13 hours ago

Often called the "Cadillac of Christmas trees," the Fraser Fir has everything a good Christmas tree should have: an even triangular shape, a sweet piney fragrance, and soft needles that (mostly) stay attached ...

User comments : 19

Adjust slider to filter visible comments by rank

Display comments: newest first

rgwalther
3.7 / 5 (3) Dec 17, 2010
I don't know about the photographic ID, but I would think anyone (human) familiar with dogs could match growl types to dog type as well as type of growl or bark to a specific event stimulus/response.
epsi00
3.7 / 5 (3) Dec 17, 2010
This is a new result for people ( and researchers ) who do not have dogs and have no clues about what dogs can do.
Smellyhat
5 / 5 (1) Dec 17, 2010
I'm not saying that the conclusion of this research is incorrect; it seems highly probable, to me. But does anyone else here see a major, major flaw in the experimental protocol?
Skeptic_Heretic
5 / 5 (1) Dec 17, 2010
I'm not saying that the conclusion of this research is incorrect; it seems highly probable, to me. But does anyone else here see a major, major flaw in the experimental protocol?

Yep:
All the dogs were well-socialized animals and were tested while their owner was present to encourage them to be comfortable in the laboratory environment.


Just one of 5 I can see from the abstract alone.
gwrede
3 / 5 (2) Dec 17, 2010
These scientists haven't been around animals. If they had, they'd know that even the neighbor's cat can tell the dog's size from it's growl. And so do even non-dog people.

Probably why this hasn't been Officially Studied, is just that it is so patently obvious to all. Maybe next time they find out that the sky is blue.
Pongrácz said the dogs’ ability to match the growl to the photograph of a dog of the same size is a "complex cognitive talent previously seen only in primates."
I'd say this is utter crap, until proven otherwise. This is just like in the old days books said that dogs don't have acute vision, and certainly no color vision, and they rely on hearing and smell /only/ when recognizing people.

I'd rather say that it's about the scientists not designing tests that seem interesting and relevant to the animals. (Non-primates don't see the point in staring at slides on the wall, and certainly don't consider them relevant to some artificial sounding "growl" tapes.)
geokstr
2.3 / 5 (3) Dec 17, 2010
wow. Such great science is being done here.

I'll close my eyes and match a "yip-yip-yip" to a chihuahua and the "ROUWF-ROUWF" to a St Bernard too. So what?

And I've never even owned a dog, but I have seen them on TV.
rgwalther
5 / 5 (1) Dec 17, 2010
Since two open heart surgeries in 10 days, I've spent 5+years, 24/7 with my 8+ years, 65lb German Shepherd/Basinji mix. I can tell from how he (Stjohn)reacts physically and barks/growls: volume, intensity, duration, bass, who/what is coming. Dog(IFF), deer, raccoon, cat..., neighbor, child/adult, stranger, mailman. Variations in growl and bark are easily ID'd. A specific puppy whine (subsets for 8 Kids and 5 grandkids). The whimper for the children is amazing in a dog with a wolf look and perfect, symmetrical Shepherd markings.
He knows which cars are potential threats, neutral or friend/family.
I easily accepted all Stjohn's vocals when he had a 20' picture window & yard with a long, clear view. After the surgeries, foreclosure, total financial loss, now 8+ months living in one son's back room with no view. Dog nor I can see, but I know exactly who is outside.
I have tried for years to get dog to respond to
TV. Nothing. Today woke him with hi-res, wolf audio, a 1st, lasted 10sec.
epsi00
5 / 5 (1) Dec 17, 2010
research in this field ( animal cognition ) is very hard. We first need to get rid of our misconceptions about what animals can do and what they can't do. And there are many layers of misconceptions and the above is just one of them. A puppy would recognize its mother among many growling mother and conversely. So this is nothing new. The next finding will be that dogs can think! and sometimes logically. True but if we start by assuming they can't then we will look silly presenting common things that happen every day as a deep result.
geokstr
2.3 / 5 (3) Dec 17, 2010
A puppy would recognize its mother among many growling mother and conversely.

And I'll wager that would still be true if you blindfolded them and plugged up their ears. Unless you think the way they find drugs and their favorite fire hydrants and other like stuff has anything to do with their sight, hearing or limited brain power.

It's call smell.
Skepticus
1 / 5 (2) Dec 17, 2010
Just listen to the howls and brouhahas on the net over wikileaks and one can accurately tell who is the big dog. No canine needs to be involved.
karenbichon
1 / 5 (1) Dec 18, 2010
We had one dog who could assess pack sizes and acted accordingly to maintain his place of alpha dog he would eat exactly 1 inch from every chicken wing thrown around the area and in every food bowl before permitting the pack to eat. The larger the pack the more he took, when pack was reduced he would eat only the amount that dogs he had with him but always 1 inch from the tip end. He growled in grunts to each dog and none would step out of line, yet he never ever bit, by the way these are toy dogs
Mercury_01
3 / 5 (4) Dec 18, 2010
I think we should start designing animal cognitive experiments to test things we wouldn't normally consider possible. I read an article about voice mimicry in dolphins. The scientists involved had reason to believe that the dolphins were even teaching each other to mimic the human voice. The sounds they made diddnt match a human, but apparently they tried. Maybe animals can do anything that we expect of them if we really try to reach out and speak their language.
Raveon
5 / 5 (1) Dec 18, 2010
I don't know about the photographic ID, but I would think anyone (human) familiar with dogs could match growl types to dog type as well as type of growl or bark to a specific event stimulus/response.


Good post, you're mostly right. Some people aren't that smart even having been around dogs.
Skeptic_Heretic
not rated yet Dec 18, 2010
I have tried for years to get dog to respond to
TV. Nothing. Today woke him with hi-res, wolf audio, a 1st, lasted 10sec.
Mileage may vary on that one. My dogs respond to TV and recognize that it is TV as opposed to reality, but, my dogs have no capability to differentiate different people or noises to identify things. I also have bulldogs, so it's quite a bit different as we've tweaked these breeds quite a bit over the years compared to the Shephards, Collie, and asiatic large breeds.
Raveon
not rated yet Dec 18, 2010
I think the same is somewhat true with domestic cats, only it isn't size it's toughness. I would bet that cats size each other up by their feline version of a growl but since they are all about the same size they are measuring how tough the other cat sounds.

I had a male Siamese mix. He looked, sounded and acted like a normal seal point but was built like a rock. I never had him neutered and he never got a scratch (if you know anything about cats you know that is unusual, all males get in fights and they get scratched and infected). He sounded like a cougar when he was facing off with another cat and I think he scared the bejesus out of them, he scared people that heard him.
Physmet
not rated yet Dec 19, 2010
I have a dog, so I know that this result seems obvious. On the other hand, why is it the small dogs can cause larger dogs to back off? I think that may be part of the curiosity. Maybe they figured that the growl could fool the other dogs about their size?
geokstr
2.3 / 5 (3) Dec 19, 2010
On the other hand, why is it the small dogs can cause larger dogs to back off? I think that may be part of the curiosity. Maybe they figured that the growl could fool the other dogs about their size?

Pah. A big dog backs off from a little yapper mostly out of amusement, not fear. I'm quite certain they know that they could tear the yipper apart in short order. The little dog is trying to make itself seem ferocious by being bombastic, but its bark is still only a yip.

But various species have all sorts of tricks to make themselves appear bigger than they really are. A threatened cat's hair stands on end and it arches its back to make itself look bigger. Others can puff themselves up with air to make themselves look larger.

Doesn't always work though.
Skeptic_Heretic
not rated yet Dec 19, 2010
Pah. A big dog backs off from a little yapper mostly out of amusement, not fear. I'm quite certain they know that they could tear the yipper apart in short order. The little dog is trying to make itself seem ferocious by being bombastic, but its bark is still only a yip.
That's not accurate in all cases. Dogs are psychologically hardwired to respond to particular signals given by other dogs. A little yapper appears unstable or neurotic. Bigger dogs that back off are backing off due to instinct, not amusement.

This is akin to how a mountain lion and a racoon won't fight it out. The mountain lion will certainly kill the racoon, but the racoon may get the mountain lion good enough to ensure death through infection.
rgwalther
not rated yet Dec 19, 2010
My dogs respond to TV and recognize that it is TV as opposed to reality
Appreciate your response. Sorry I ran out of space and became embarassed over my dog fixation. My dog does recognize TV but ignores TV info as useless (no threat, scent, food etc?). There are ethics/intelligence lessons there somewhere.
I've had many dogs over the last 60 years. Were they intelligent when I was a child/youth, I was too stupid/busy to notice.
The last 24 years I have had two very different dogs(a mostly scotty & a half-breed shepherd) endless cats and lots of children around. I learned to pay attention. The scotty was a single-minded caretaker of any creature he felt was young(kids) or stupid(kids, puppies, kittens).

Last 5+ years, kids grown, self sedentary, I've watched a dog grow & develop amazing personality and intelligence. All positive instincts on high alert/sophistication. Wonder the results had he been actively trained.
Again! On and on about the dog.
Merry Christmas

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.