No, Science Doesn’t Say Breeds Have Distinct Personalities
January 18, 2019
There’s been a lot of
Pennisi’s article is about a study by University of Arizona researchers called Highly Heritable and Functionally Relevant Breed Differences in Dog Behavior. The problem is that the title of Pennisi’s article does not accurately reflect what the study says at all. The headline doesn’t even represent what the article says.
Not only did many people share the story on social media without reading the article or the study,
This is a lesson we should all have learned by now – headlines don’t tell the whole story or even an accurate story.
So What’s Really Going On with This Study?
The study focuses on dog behavior and breed traits, not personality. Personalities are entirely dependent on individuals. Breed traits do not make up the whole of who a dog is or what we can expect a dog to do.
The article makes frequent comparisons to human behavior, so let’s do that here – do siblings have the exact same personalities and respond in the exact same way to specific situations and stimuli?
The researchers openly admit their study falls short of exploring dogs’ individual personalities:
“One limitation of this study is that genotypic and phenotypic data were not collected from the same subjects, but rather aggregated across independent datasets.”
This is data without context. Data without context is easy to misinterpret and apply to one’s own preconceived notions. That’s exactly what’s being done by breedists (people who insist they can predict everything about a dog based on the way it looks and how it is categorized by kennel clubs) and it’s also being lauded by anti-dog advocates who insist that dogs with the label “pit bull” are a menace to society.
The truth is all this study shows is a correlation, not
To bring people back to reality, Marc Bekoff, Ph.D. put the study into context in an article for Psychology Today called “Dog Breeds Don’t Have Distinct Personalities.”
“… a correlation of some traits with a breed/breed mix doesn’t mean there is a causal—cause-and-effect—relationship between or among them.”
Again, breed traits, which most certainly exist, do not dictate personality. Bekoff points out something many of us have personally observed:
“When I watch dogs, I focus on individual differences among them, because no two dogs are the same. I love when people tell me that they live with two dogs from the same litter and they’re as different as night and day.”
He goes on to mention that breedism and the misinterpretation of breed traits can increase return rates for shelters and rescues. People adopt dogs believing they will be carbon copies of their breed stereotype only to find that the dog is an individual doesn’t meet their breed-based expectations.
We didn’t want to address this issue without giving Science Magazine a chance to respond. We expressed our concerns about the title and some of the implications in the article.
Their online news editor David Grimm, Ph.D., explained that the magazine often covers pre-published and non-peer reviewed studies and that the article did make it clear that correlation doesn’t equal causation.
He told us:
“As you may recall, one of the experts, Heidi Parker, echoes your and Marc’s [Bekoff] concerns: ‘This paper doesn’t call out any particular breed for its behavior.’ The article is also careful to point out that DNA explains only about 15% of a breed’s personality.”
But like we stated in the beginning, that’s not at all what the title of the piece says and thus, people now have a false understanding of dog behavior and genetics. Grimm never addressed the issue of the inaccurate title.
Bekoff, who was included on our communications with Grimm, responded:
“The title “Dog breeds really do have distinct personalities—and they’re rooted in DNA” is over the top based and misleading based on the data that are presented — it’s catchy and all that but some people have walked away saying things like, ‘I knew that’ when the data are equivocal at best.
The fact remains that dog breeds do not have distinct personalities and if there are any correlations, which do not at all imply causation, they are weak — and while they may be rooted in DNA — where else could they be rooted — the title is misleading and that’s what caught my attention and the attention of the many people who wrote to me asking what I thought about it all.”
So What’s the Problem with a Catchy Title?
We know that to make it in the internet age articles need to have catchy titles. Believe us, we get it (this article is currently being written by someone who used to be a viral editor for a major dog publication – I SO get it), but your titles need to be accurate.
We know, and as has been demonstrated by the viral nature of the Science Magazine write up, people often share articles based on headlines and take away whatever impression the headline gives them as the truth.
Because of this, Bekoff isn’t the only scientist balking at those who take the study out of context. Dr. Ádám Miklósi is concerned that the misrepresentation and failure to recognize the common variations among dogs from the same breed, even from the same litter, can have long-lasting and negative consequences for individual dogs and their owners – one of which, we mentioned above. Other consequences include breed-specific legislation and policies.
We’re not only calling out one publication here, we’re calling out everyone who shared the article without reading it, thinking critically about it, or putting it into context. We need to be accurate about how we talk about dog behavior and genetics because it has serious consequences for all dogs – not just dogs who are routinely discriminated against. Every dog deserves to be seen for who they are, not who we think they are.
Remember that the takeaway here is that data means nothing without context and that, as Bekoff says:
“breeds do not have personalities, individual dogs do.”
Let’s think critically before we share articles on social media. Let’s think critically before we believe articles on social media (and yes, that includes this one! We like thinkers.)
Get Involved with Our Work
Give to the Cause
Host An Event
Post on Social Media
Become a Corporate Partner and Sponsor a Program or Event
P.O. Box 624
Bangall, NY 12506