Dogs have fantastic noses and can identify their owners by smell. Can humans return the favor? Can dog owners identify their dogs by smell? Some scientists put this to the test. Lu and Tirth discuss this scintillating study on S01E11 of the Recreational Science podcast (timecode 21:02):
Tirth: Lu, what do I say all the time? Dogs are better people than people are.
Lu: Oh, is that right?
Tirth: Yeah. We’ve talked about this before.
Lu: That dogs are the best people?
Tirth: Dogs are the best people.
Lu: I see. I agree.
Tirth: You know, it’s well known that dogs have a very strong sense of smell. They can identify people, other dogs, other objects just by their smell. They can even find their way home if they’re lost, right?
Lu: Well, they can also sense the earth’s magnetic field.
Tirth: Yeah, that’s probably a big part of it, but the question is, can humans recognize dogs by their smell?
Lu: Can humans recognize dogs by the dog’s smell?
Tirth: Yes. The title of the study is “Discrimination of Dog Odors by Humans.”
Lu: So we established that dogs make the best people. This study is asking, can people make good dogs?
Tirth: Yes, exactly.
Lu: Okay, good, good. I like that.
Tirth: This was published in the journal Perception.
Lu: Oh, good journal.
Tirth: So the researchers recruited 26 dog owners, gave them a blanket, and told them, “hey, lay this blanket down wherever your dog sleeps in your house and leave it there for three days, so the dog’s odor will be transferred to the blanket.” Then they had the dog owners bring the blankets back, mixed them all up, and asked the owners to do a blind smell test. Specifically it was a pair wise test. So each dog owner was given two blankets, one of them was their own dog’s blanket, the other one was another dog’s.
Lu: And they were asked to guess which blanket belongs to their dog?
Tirth: Yes, exactly. Do you want to put your bets down? What do you think the result was?
Lu: Well, first, I want to ask, other smells from the house could potentially get on the blanket, right? Like, if you use a lot of perfume or you cook a lot of chili, some of that could get on the blanket. Owners can recognize that instead of their dog’s smell.
Tirth: Yes. Beautiful question. The researchers actually accounted for that as well. They gave each participant a second blanket and told them to hang it up somewhere in their house, out of the reach of the dog. They then tested the participants’ ability to recognize their house by smell.
Lu: Okay, so two different tests.
Tirth: Two different tests, yes.
Lu: A very well-controlled study.
Tirth: Yes, because they pointed out – exactly like you said – it’s possible that when smelling the dog’s blanket, the owners are recognizing their house’s smell, rather than their dog’s smell. So they wanted to control for that. There’s a term for this, by the way, we call these confounders or confounding factors.
Lu: So for the dog blanket test, dog owners were given two blankets, one from their own dog, and one from another dog.
Tirth: Yes, and not just that, they try to match up the breeds of dogs as best as they could. So for example, if your dog is a golden retriever, the second blanket you got was from either a golden retriever or a very similar breed.
Lu: Wow, well controlled. I’m going to say the owners were not able to tell which of the two blankets belonged to their dog. I’m going to say they guessed about 50% correct.
Tirth: Okay. And then what about the control test of guessing their house smell?
Lu: I’m also going to say they were not able to guess which blanket was placed in their house. I’m going to say they again guessed 50% correct.
Tirth: Okay. Before I get into that, I just want to point out that the researchers told the dog owners not to groom their dog a month before the experiment so that chemical/shampoo smells don’t get on the blanket.
Lu: Okay, so no bath?
Tirth: No bath, nothing.
Lu: Wow. For a month?
Tirth: For a month.
Lu: Very rigorous. Very good scientists.
Tirth: Well, here are the results, Lu. And this is going to surprise and shock you because you have very little faith in people. A whopping 88.5% of dog owners were able to recognize their own dog just by smelling the blanket.
Lu: Wow. Okay. Wow.
Tirth: But on the flip side, only 69% of participants correctly identified their own home by smelling a blanket.
Lu: Well, that’s still pretty high though.
Tirth: Yes, but importantly it’s lower than the dog blanket accuracy.
Lu: Okay, interesting.
Tirth: I just wanted to add a mini follow-up to this. There’s a second study where they basically did a very similar experiment with cats. Because cats don’t lay down in one place and are not as predictable as dogs, they told the owners, take the blanket and rub it over your cat several times so you get the cat smell on it. But alas, only 52% of the participants recognize their own cat’s blanket. So no better than chance.
Lu: Wow. People don’t get close enough to their cats, I guess.
Tirth: Well, their argument is that cats can like run away, but also cats groom themselves all the time. So they’re cleaning their own odor.
Lu: So maybe they don’t impart any smell onto the blanket.
Tirth: Yeah, exactly.
Article citation
Wells et al., 2000. The discrimination of dog odours by humans. Perception. https://journals.sagepub.com/doi/abs/10.1068/p2938
Leave a Reply