There’s a story that animal-behaviour experts like to tell about a horse named Clever Hans.
Clever Hans toured Berlin in 1904, amazing onlookers with his feats of cognitive skill.
He could do math, identify paintings, and form whole sentences — all by tapping his hoof on the ground. Skeptics were allowed to test the horse and came away convinced that, yes, he really could do these things.
But in 1907, a biologist decided to conduct a different kind of experiment, placing a screen between the questioner and Clever Hans. Sure enough, when he was unable to see the face of his questioner, the horse could no longer perform on demand.
Clever Hans was almost certainly smart, but he wasn’t mathematically gifted — he was able to read people’s facial expressions. Unbeknownst to his trainer, the horse had been watching his facial cues and using them to determine when he had tapped the correct number of times. If the person asking Hans the question didn’t know the correct answer, the horse was at a loss.
The story serves as a warning for animal researchers: everything may not be as it seems, and results are only as good as the study that produced them. When humans make judgments about animal behaviour, they often also make assumptions. And animals, especially dogs, can be very good at reading humans, so researchers need to be extremely cautious when setting up their experiments.
“We're not that good at understanding dogs, but dogs are good at understanding us,” says Daphna Buchsbaum, principal investigator at the University of Toronto’s Canine Cognition Lab. “When we naturally interact with them, we don't always correctly know what they see and think about a situation, but they're often very good at fitting into the day-to-day communicative environment of people and understanding what we mean when we're trying to tell them something.”
We may think we know what dogs are thinking, but we’re inclined to be biased observers — to map human emotions and responses onto our pets.
In a 2009 study, researcher Alexandra Horowitz investigated the claim that dogs look guilty when they’ve done something wrong.
Owners would put food on the floor, tell their dog not to eat it, and then leave the room. In some cases, a researcher would remove the food before the dog could eat it; in others, the food stayed where it was.
It made no difference whether the dogs actually ate the food: those that disobeyed and those that didn’t had the same reactions — looking away and cowering, for example — when confronted by an angry owner. Horowitz’s findings suggest that that the pets aren’t feeling remorseful; they’re afraid of being punished and therefore are trying to appease their potential punisher.
“We anthropomorphize,” says Lee Niel, principal investigator at the Ontario Veterinary College’s Companion Animal Behaviour and Welfare Lab in Guelph. “We think about the situation in terms of what we would expect humans to do in some situations, and that's not necessarily what's going on.”
When Niel and Buchsbaum design their studies, they try to control for the human tendency to make assumptions. One of Niel’s current experiments involves finding out how able people are to recognize signs of fear in puppies.
“Many times, people interpret barking behaviour as their puppy being excited, but the research is showing their puppy can be scared,” says Niel.
The canine subjects are exposed to certain stimuli — remote-controlled cars, for example — and their reactions are recorded. The human subjects are asked whether the puppy is showing fear, but they don’t know what it’s seen: if they knew about the remote-control car, they might be more likely to think the dog was scared.
“That's critical to this type of research. If you really want to understand what's going on for the animal from an objective point of view, you need to remove that external bias,” says Niel. “You're setting things up right from the beginning to be able to clearly, objectively state what the outcome is and then what that means.”
Both labs are open to the public and in need of dogs to participate in upcoming studies.
May we have a moment of your time?
Our public funding only covers some of the cost of producing high-quality, balanced content. We depend on the generosity of people who believe we all should have access to accurate, fair journalism. Caring people just like you!
Get Current Affairs & Documentaries email updates in your inbox every morning.