Every night, suburban parks fill with a mix of pooches of all shapes and sizes letting off some steam as their doting owners watch on.
- US scientists analysed a survey of more than 18,000 dog owners and sequenced the genome of more than 2,000 dogs
- They found that breed stereotypes are a poor indicator of individual dog behaviour
- They say the findings challenge current assumptions surrounding dog breed stereotypes, used to explain why some breeds are more aggressive, obedient or affectionate than others
Parks have recently swelled with young dogs and first-time owners as pet ownership boomed through the pandemic.
Some dogs are very social, others bark at strangers, some will fetch, others make their humans do all the work.
While it might be tempting to attribute the behaviour of these canine companions to their breed, a new study published in the journal Science suggests breed is not a good predictor of an individual dog’s behaviour.
“I feel like sometimes people think of the breeds as sort of like catalogue shopping,” said study co-author Elinor Karlsson of the University of Massachusetts.
“You can go to those websites and they’ll tell you, ‘If you want a dog that’s going to be friendly with your kids, you should get these breeds,'” Professor Karlsson said.
“But you could get a dog from one of those breeds and it just happens to be a dog that doesn’t like children.
According to the study, age or gender, rather than breed, were the best predictors of behaviour for some traits like activity level and playing with toys.
But there were still some subtle differences, with some breeds more likely to display traits that harked to their ancestral roots.
Once were wolves
Dogs evolved from ancestral wolves thousands of years ago and were gradually selected for skills such as hunting, guarding or herding.
The notion of breeds — selecting for physical and aesthetic traits such as colour, size and type of fur — only goes back around 160 years to the Victorian era.
Modern breeds are ascribed qualities by organisations such as the American Kennel Club based on their ancestral functions as herders or hunters.
Border collies, for example, are described as “affectionate, smart and energetic”, whereas beagles are “friendly, curious and merry”.
To find out if these stereotypes had any genetic basis, the team surveyed the owners of more than 18,000 dogs who participated in a citizen science project called Darwin’s Ark.
The owners were asked a series of questions designed by dog trainers to tease out how their dogs stacked up in eight categories, such as their ability to follow directions, control their impulses, play with toys, socialise with humans or dogs, or how easily they were provoked by uncomfortable stimuli (“agonistic threshold”).
“People are very good at telling us about their dog’s behaviour. They spend a lot of time watching their dog and if you ask them, they will tell you really accurately,” Professor Karlsson said.
The team cross-referenced this with DNA samples from 2,155 dogs.
While around 600 of the dogs were purebreeds, most involved in the sample were mutts, said lead author Kathleen Morrill of the University of Massachusetts.
The mutts also helped rule out biases about breed stereotypes from the owners, who were often unable to correctly identify the mix of breeds in their dog.
But genetic testing can reveal what percentage of a breed a mutt carries.
For instance, Jack (above) is one-quarter American pit bull terrier.
The study found breed only explains 9 per cent of the behavioural variation in individual dogs.
“We basically showed that every behaviour was seen in every single breed,” Professor Karlsson said.
In saying that, though, there were some traits that were more likely to be passed down through the generations of some breeds of dogs.
Border collies and mutts with a greater percentage of border collie were good at following directions, while beagles, bloodhounds and Siberian huskies were more likely to howl than other dogs.
The team identified 11 points on the genome that were associated with behaviours such as howling and human sociability.
Many of these so-called motor characteristics, such as howling, pointing, retrieving and playing with toys, are throwbacks to when dogs evolved from wolves.
Is there an aggressive type?
Although a number of breeds, such as American pit bull terriers, are banned or restricted in Australia, no breed in the study stood out as being aggressive.
“[Agonistic threshold] was actually the behavioural factor that we found almost no evidence of any breed differences for,” Professor Karlsson said.
Instead, these behaviours appeared to be influenced by environmental, not genetic, factors.
“You could be born with the best genetics in the world, but if you don’t have the right experiences, for example, in early puppyhood, that could totally change your personality,” she said.
What about working dogs?
The study included information about 78 companion dog breeds, but it did not go into the behavioural traits of working dogs.
“These dogs are currently being selected to do a job, people don’t care what they look like,” Professor Karlsson said.
“We would expect to see a lot more behavioural distinctiveness in those populations,” she said.
But there were very few iconic Australian working dogs like kelpies in the owner database, she added.
It is very hard to study the genetics of behaviour, said Claire Wade, an animal genetics expert at the University of Sydney who studies purebred kelpies.
“We know that these complex traits are likely to be influenced by hundreds, if not thousands, of different genes,” Professor Wade said.
“There’s no question you get a lot of variation across breeds, but I think it comes from so many places affecting each trait that it’s going to be hard to pin down something that’s biologically meaningful.”
There’s a lot of nurture in there, as well as nature.
“How many of those behaviours are innate — just the way the dog is — versus trained behaviour?” Professor Wade said.
And perception of a dog’s behaviour can depend upon circumstance, or where the dog is when it’s being watched.
Professor Wade said trying to tease out the answers to these questions would be even harder with mixed breeds.
“If you do have major genes that affect something, are they dominant or recessive, are they affected by environment, are they masked by genes contributed by the other breed?
“But if you throw large numbers at it [as they’ve done in the study], then you can find biological meaning at the end of it, but I don’t think they are there yet [in our understanding of the genetics of behaviour].”
Back at the dog park …
However, while the genetics of behaviour was “still a mystery”, Professor Wade said the data collected in the study on how aging affected a dog’s behaviour was very interesting.
“When they are young, they want to play with everybody, but that really drops off by the time they are four or five. Then they end up not wanting to play with anyone.
“We kind of know that, but yet we all drag our dog off to the dog park hoping it’s going to get along with most dogs.
“And what that shows is in the case of most dogs, that’s probably not the right thing to do.”