The largest-ever study of dog behavior, tracking more than 47,000 pets, reveals how breed, size, sex, and age shape canine traits. It provides insights that could help identify early signs of illness and improve training and welfare.
In addition to breed-specific traits, dog behavior is, broadly speaking, inextricably linked to well-being. Everyday behaviors are an effective way of gauging a dog’s emotional state and their state of physical health.
A new large-scale study jointly led by researchers from Virginia Tech and the University of Washington has established a crucial baseline for understanding the link between dog behavior and health.
“Most importantly, with these data, we’re excited to now have a starting point from which we can continue to follow changes in the behaviors of tens of thousands of dogs as they age, which will ultimately help us understand how behavior and health are linked,” said the study’s co-lead and corresponding author, Courtney Sexton, PhD, a postdoctoral researcher in the Virginia-Maryland College of Veterinary Medicine.
The researchers used information from the Dog Aging Project (DAP), which follows tens of thousands of pet dogs in the US over their lifetimes. For the present study, they utilized data on 47,444 dogs enrolled between 2020 and 2023. Owners completed the Health and Life Experience Survey (HLES), which included a shortened version of the Canine Behavioral Assessment and Research Questionnaire (aptly abbreviated to C-BARQ). The researchers used a statistical technique called principal component analysis (PCA) to group dog behaviors into four broader domains: fear, attention/excitability, aggression, and trainability. Statistical models were used to see how behavior related to factors like breed, sex, size, spay/neuter status, life stage, region, and year of enrollment.
“When you have a data set this big, you really do have power in numbers,” Sexton said. “While we can’t understand all the factors, when we find statistical significance, there is likely something there worth thinking about in its real-world context.”
The researchers found that enrollment year affected behavior. Dogs that joined between 2021 and 2023 scored lower in trainability than those who joined in 2020. Whereas fear scores stayed fairly stable across years, and attention/excitability was slightly higher for dogs enrolled in 2022. Only in 2023 did aggression scores dip significantly compared to 2020, not as a consistent trend across years.
Mixed-breed dogs scored higher on fear, attention and aggression than purebreds, though not on trainability. Smaller dogs – those under 20 lb (9 kg) – were reported as more fearful, more aggressive, and more attention-seeking, but less trainable than larger dogs. Puppies, as you might expect, demanded more attention, were less fearful and aggressive, but also less trainable than adults (though still more trainable than senior dogs).
Female dogs were reported as more trainable than males. Spayed or neutered dogs were generally rated as more fearful, more aggressive, and less trainable compared to intact dogs. This contradicts the belief that spaying or neutering a dog reduces aggression. As you’d hope, service and therapy dogs showed less aggression, and obedience dogs scored higher on trainability than pet dogs. Some regional differences were noted. For example, dogs in the Midwest were rated as more attention-seeking than those in other areas.
“We saw that certain factors, such as a dog’s life stage, sex, and size had some influence on their behavior,” said Sexton. “Interestingly, we found that throughout the COVID-19 pandemic time period, there were not substantial changes in dogs’ overall behavioral profiles from year to year despite what we might have expected given the changes in environment and routines that many were experiencing at that time.
“What was interesting was that the difference in reported trainability between the first year and the last year was the smallest of any in the averages. It’s a small statistical significance but could show that dogs, or their owners, are bouncing back [from conditions imposed during the pandemic].”
One of the study’s largest limitations is that it relied on owner-reported data, which relies on owners’ perceptions, which may be biased or inconsistent. Additionally, household factors such as training style, family stress during the COVID-19 pandemic, and owner experience weren’t directly measured. There were few intact dogs included in the study, which may skew results. And, importantly, the study shows associations, not proof that traits like breed cause certain behaviors.
Nonetheless, the huge study gives scientists a starting point to track how dog behavior changes over time, especially as dogs age. Shifts in behavior – a change in fear or aggression, for example – could be early warning signs of illness or cognitive decline. These insights can help trainers, vets, and owners tailor approaches by breed, size, or life stage.
Further studies are needed to explore how factors such as a dog’s location or their health status influence behavior over time.
The study was published in the journal PLOS One.
Source: Virginia Tech
