New Hope for Shelter Dogs

How Tufts-based behavioral research could help many pooches find their forever homes

photo booth images of a happy dog

All it takes is a little footage of sad-eyed puppies set to a Sarah McLachlan soundtrack to cue the waterworks and get many people pulling out the credit card to donate for shelter dogs. But what actually goes on at the humane organizations, animal control facilities and rescue groups across the country that do the work of caring for homeless dogs and matching them with new families?

To find out, Tufts Now recently spoke to Seana Dowling-Guyer, associate director of the Center for Shelter Dogs at Cummings School.

Tufts Now: How do dogs tend to wind up in shelters in the first place?

Seana Dowling-Guyer: In many regions of the country, dogs and puppies are brought into shelters because there are more of them than there are available homes. Animals end up being euthanized simply because there are far more of them than the shelters can care for humanely.

Here in New England, we are fortunate that spay and neuter rates are high. However, we still have plenty of homeless dogs, because our shelters have the time and space to keep surrendered dogs that were once considered “unadoptable”—for example, older dogs with medical problems—until the right home is found. The result is that the shelter population in our region has changed from a population of largely young, healthy dogs to somewhat older dogs and those that may have medical or behavior issues.

We know from research back in the 1990s that medical and behavioral problems are common reasons for dog relinquishment everywhere. As the cost of veterinary care rises, many shelters report an increase in the number of pet owners who are forced to consider giving up their pet because of a medical condition for which they cannot afford treatment.

What does the Center for Shelter Dogs do?

The center was established in 2008 at the Animal Rescue League of Boston through a grant from the Stanton Foundation, but it found a new academic home at Cummings School in 2014 as part of Tufts’ Shelter Medicine Program at the Center for Animals and Public Policy.

Back when the center was based at the Animal Rescue League of Boston, we primarily educated shelter staff and volunteers about shelter dog behavior. (Those educational resources can be found at When you’re in the shelter, your concerns are so immediate. It’s like the ER—you’re always reacting to crises, and you rarely get a chance to step back and think about the big picture.

Now that the center has moved to Cummings School, our mission is to conduct the rigorous science that provides evidence-based tools and techniques to help dogs in shelters, and to keep them from ending up in a shelter in the first place.

How does the center’s research help shelter dogs?

Teton poses for the camera, hoping to be adopted. Photo: Guinnevere Shuster/Humane Society of UtahTeton poses for the camera, hoping to be adopted. Photo: Guinnevere Shuster/Humane Society of Utah
Many shelters use behavior evaluations to assess a dog, especially when they don’t have any details about that animal’s history. The results may be used to match dogs to adopters and sometimes even to elect for euthanasia. However, more research is needed to show how predictive these evaluations really are of what a dog’s behavior will be like outside the shelter.

For some behaviors—basically those indicating a friendly, resilient type of dog—the commonly used evaluations appear to be pretty predictive. If you see friendly behavior in the shelter, you are likely to see it in the home. It’s the dogs that are having an issue in the shelter that may not show the same behavior in a home.

I’ll give a couple of examples. For our study published in July 2013 in Applied Animal Behaviour Science, we looked at food-related aggression in shelter dogs—a dog showing teeth, growling, snapping, lunging or biting when either the animal or a nearby food item is approached or touched by a person.

We took 20 shelter dogs that had shown some kind of aggressive reaction around food on the standard evaluation, and followed this small group of pets after they were adopted into homes. We talked to their owners a few months to a year later, and found that only about 55 percent of them reported seeing their dog display any of these behaviors around food—and, of those that did, most of their owners didn’t think the behavior presented any challenges to keeping the dog as a pet.

We also followed 77 dogs that hadn’t shown signs of food aggression, and found that about 16 percent of them displayed the behavior at home, and again most of their owners did not find their dog to have troublesome behavior. This matters to dogs’ lives because some shelters will euthanize for food aggression. That’s because these animals have been considered unsafe around people. The knowledge that they may never display that behavior in the home—or, very possibly, at least not in any dangerous way—once they’re settled, secure and bonded could change that.

In another study, published in December 2014 in Applied Animal Behaviour Science, we assessed a common evaluation technique of using a stuffed animal to determine if a shelter dog is aggressive or fearful around other dogs. We wanted to see if the results would align with how these animals later reacted to a real dog. Again, the happy-go-lucky dogs were pretty consistent with their evaluation and real-life behavior. But the other dogs were all over the place.

For our next study, we want to use several different live dogs to see how that influences the results. Other research has shown that dogs are incredibly sensitive to both canine and human social signals, so it’s important to consider our influence and the influence of any other dogs during an evaluation. Most times, shelters have to use a dog or two from their own kennels to evaluate how well a shelter dog may get along with others, but if that dog is stressed or doesn’t react well, it can affect the test dog’s results. However, that may not mean the dog couldn’t be adopted into a home to be friends with the right dog.

What other research is the center working on?

We survey shelter staff and volunteers about what kinds of dog behaviors they see and which ones they are most struggling with and what kind of community programs exist locally.

We also survey pet owners. For example, in 2013, we worked with Helia Zarkhosh, when she was pursuing her M.S. in animals and public policy at Cummings School, to investigate how photographs influence interest in adoption. We suspected that just as a profile picture can have a big effect on a person attracting dates online, a shelter dog’s photo can be key to finding a home through a site like

Our study, which we are hoping to submit for publication this year, investigated whether a shelter dog’s “attractiveness” differed by type of dog, as well as by what was or wasn’t in the picture. In this study, we took photos of different types of dogs with a person, without a person, in the shelter through the kennel bars, and outdoors. We then asked a random sampling of people visiting a pet adoption website about which dogs they liked best. Their responses differed in some cases depending on the dog, but overall nobody liked the pictures of a dog through the bars, and everybody liked pictures of a dog outside. Depending on the type of dog, it made a difference whether the person was there or not.

This summer, I’m working on another study with Jody Epstein, a current student in Cummings School’s M.S. in animals and public policy program, to study the effect that anxiety wraps—body-gripping garments for pets such as the “ThunderShirt”—have on shelter dogs’ stress behavior.

There has been some research, including by Professor Emeritus Nicholas Dodman, that shows anxiety wraps appear to relieve acute stress in dogs—the kind of panic a pet may experience during a scary moment like a thunderstorm or a local fireworks celebration. We want to know if these aids might also reduce chronic stress experienced by dogs that have been in a shelter for weeks or months. We’ll be working with local shelters across New England to enroll dogs so we can observe their behavior and test their cortisol levels, a physiological sign of stress.

How does stress affect shelter dogs?

The effects of stress on a shelter dog can be really obvious or really subtle. A stressed dog may exhibit a “stereotypy”—that is, behavior like pacing or jumping—or get really barky as a way to try to control its environment. For obvious reasons, most adopters steer clear of those behaviors.

However, a dog may also become more withdrawn as a result of stress. These animals shut down, and staff may label them as shy like it’s a personality trait. But it’s really stress-induced anxiety. You put that dog in a different environment, and you’d consider it a pretty friendly pet. So reducing stress for shelter dogs not only improves their welfare from a physiological standpoint, it also may help them get adopted more quickly.

Genevieve Rajewski can be reached at

Back to Top