Almost two decades ago, I argued against testing dogs for “food aggression” (what is now called “resource-” or “food-guarding”) because doing so was setting dogs up to fail. I argued that there were too many false positives, every dog — from the 120 lb. rottie to the 10 lb. chihuahua — had a little bit of the wolf in him, and shelters were notoriously unreliable venues to predict behavior in a home. Moreover, as an animal control director, I did not kill dogs for “resource guarding” their food because I found that adopters were tolerant of it. It was also easy to treat and often — outside of the shelter when the dog no longer lived in a world of perceived competition and scarcity — self-limiting. It resolved on its own.
Underscoring my own experience, in a subsequent survey of 96 dogs who were deemed food aggressive via standardized testing but adopted anyway, only one exhibited food guarding in the home and it went away after three months. In short, “temperament testing” dogs for food guarding was killing dogs and not protecting the public.
Eighteen years later, a comprehensive study of pre- and post-adoption of dogs who guarded their food in the shelter comes to the same conclusion. Ironically, the study occurred at the very shelter I used to run. It looked at five years of shelter data (long after I left) and found that while 15% of dogs were classified as guarding their food, many of them “do not guard food in their adoptive homes, and, even when dogs continue to display food guarding in the home, adopters do not consider it to be a major problem.” In cases where the initial adopters did return the dog, the dogs were readopted and tended to stay in the home. The conclusion: shelters should opt for “adoption rather than euthanasia for most dogs identified as resource guarders during behavioral evaluations in shelters.” If the dog is returned, the study authors further concluded, they should be readopted (and can be adopted) without incident.
The study, “Characteristics and Adoption Success of Shelter Dogs Assessed as Resource Guarders,” is here.
Though the study is welcome, it doesn’t go far enough.
As I’ve since argued, shelters should throw out the fake hand, the doll, the food bowl takeaway, and the loud knock on the door. The debate as to whether temperament testing in a shelter is effective, flawed, needs modification, or should be discarded should be over. We should get rid of temperament testing altogether. Thankfully, we don’t have to wait another 18 years for a study to confirm this, too.
A recent study proved that not a single behavior evaluation test used by shelters has any predictive ability and concluded that they are no better than a coin toss. In fact, a coin toss may be better: “a positive test would at best be not much better than flipping a coin, and often be much worse, because many of the dogs who test positive will be false positives.”
Another study that evaluated 25 years of research was even more emphatic: not only were some of the tests wrong as much as 84% of the time (a combination of poor tests and poor testing practices by pound workers), they can never be reliable because they are built on a “fatally flawed” premise: “that the provocations used at a single time during a dog’s stressful experience in a shelter will predict future behavior at a different time and place.” The study called for an end to their their use.
Unfortunately, just because they shouldn’t be used doesn’t mean they won’t be. Regressive shelters do not make decisions because they are right; they make them because they are easy. And it is easier to kill animals than do the work necessary to stop it. Take the dog in the video, for example. The Philadelphia pound intended to kill this dog back in July 2017 after provoking her in a “temperament test” and deeming her “food aggressive.” She would have been killed, too, had the video not gone viral and a rescue group not intervened. No surprise: she did not show any “aggression” after rescue.
If the killing is going to end, we have to demand — and take action to ensure — that it does.
Have a comment? Join the discussion by clicking here.