In an election year, we seek solace in constants: the tastiness of pizza, strong coffee, a dog’s trusty nose. But is a dog’s nose straightforward? While “B” dogs like Beagles, Basset Hounds, and Bloodhounds were selected for scenting and “live to use their nose,” others like the Siberian Husky and Afghan Hound were not selected along these lines. Surely this plays out in practice: those selected for nosiness should outperform the non-nosey dogs in scent-based tasks. Duh, right?
But it doesn’t always play out that way. In fact, many of the dogs participating in scent-based work are not actually scenting breeds. A 2014 study published in Forensic Science International by Tadeusz Jezierski and colleagues found that German Shepherds (from the AKC herding group) were more accurate than Labrador Retrievers, “Terriers” and English Cocker Spaniels in drug detection. More recently, Nathaniel Hall and colleagues published a paper in the Journal of Comparative Psychology finding that Pugs — continually making the news for their breathing difficulties — outperformed German Shepherds and Greyhounds in an odor discrimination task.
What’s going on? For one thing, dog handlers know that performance in scent detection and odor discrimination is affected by factors like trainability, obedience, stamina, and motivation. Deborah Palman, a retired state game warden, explains in Cat Warren’s book, What the Dog Knows: The Science and Wonder of Working Dogs, “Dogs of any breed with the proper temperament that are trained the way top bloodhounds are can probably do as well as the best bloodhound.” Olfactory prowess is but one part of the package. Bears have great sniffers too, but you don’t see them checking your luggage at the airport.
Although many dogs do well on sniffer tasks, the question still remains: are dogs that were selected to excel in olfaction-based tasks actually better than dogs not selected along these lines? What is a dog’s scenting behavior prior to training? Zita Polgár and colleagues at the Family Dog Project in Budapest set out to investigate whether a simple procedure requiring no pre-training — what they call the Natural Detection Task — could reveal a dog’s propensity to orient toward a known scent. Their study is available open access on PLOS One.
Dogs, wolves and noses
The researchers investigated nose performance in four different groups: would breeds selected for their scenting abilities, like the Beagle and Basset Hound, fare better than non-scenting breeds, like the Whippet and Miniature Pincher? How would short-nosed breeds, like the Boston Terrier and Pug, perform? And for the fourth group, how do dogs compare to hand-raised wolves, dog’s closest living relative?
I smell it!
In each trial of the Natural Detection Task, dogs and wolves walked on leash past four identical pots. Unbeknownst to them, a container was under each pot, one of which contained a familiar, desirable treat (raw turkey meat). The million-dollar question: would they indicate that one pot was not like the others?
To examine detection thresholds, dogs and wolves were exposed to increasing levels of difficulty as less scent was let out of the baited container. In Level 1, the container under the pot had no cover; easy peasy. In Level 2, the top of the baited container had 5 small holes; pretty easy, and by Level 5, the most difficult, the baited container was completely covered. Each level had four trials for a total of twenty trials.
Since dogs and wolves are not one-trick ponies, they indicated in a number of different ways: “placing paw on pot; refusing to move away from pot; attempting to turn pot over; vocalizing while next to pot; significantly increasing tail wagging speed while sniffing pot; or sitting next to pot.” Any of these behaviors indicated, “BINGO! SOMETHING IS UNDER HERE! THIS POT! OH YEAH!” A correct indication resulted in the dog or wolf getting to eat the food. Take a look:
The nosey dogs
The status quo prevailed: breeds selected for scent work demonstrated a higher olfactory acuity than other dogs. While there was no difference between the groups at the first four levels (the easier levels), at Level 5 — when the test was most difficult and the baited container had no holes — scent breeds and wolves performed better than non-scent and short-nosed dogs. And while the percent of 'No choice' remained relatively constant for the scent-breeds as the task became more difficult, the non-scent breeds displayed a greater percent of ‘No Choice’ as the trials progressed, particularly at the most difficult level.
But simply being in the scent group does not assure olfactory prowess. The Natural Detection Task also captured individual differences. Each group had poor performers: one of the Beagles (in the scent group) got 8/20, a Hungarian Greyhound (in the non-scent group) got 3/20, and a Bullmastiff (in the short-nosed group) also got 3/20.
On the other hand, some dogs in the non-scenting group deserved the gold, including a Hungarian Greyhound and a Greyhound mix who both got 20/20 and a Whippet who got 19/20. In the short-nosed group, a Boston Terrier also got 19/20.
A test like the Natural Detection Task highlights general trends, but also individual differences. While scent breeds generally performed at a higher level than the other dogs, one Beagle wasn’t so hot, while a Boston Terrier was a rockstar. Regardless of their performance, I would like to meet them, because they are dogs.
Previous Dog Spies posts on the dog’s sniffer:
Hall NJ, Glenn K, Smith DW, Wynne CDL (2015) Performance of Pugs, German Shepherds, and Greyhounds (Canis lupus familiaris) on an Odor-Discrimination Task. J Comp Psychol 129: 237–246.
Jezierski T, Adamkiewicz E, Walczak M, Sobczyńska M, Górecka-Bruzda A, Ensminger J, et al. (2014) Efficacy of drug detection by fully-trained police dogs varies by breed, training level, type of drug and search environment. Forensic Sci Int 237: 112–118.
Polgár Z, Kinnunen M, Újváry D, Miklósi Á, Gácsi M (2016) A Test of Canine Olfactory Capacity: Comparing Various Dog Breeds and Wolves in a Natural Detection Task. PLoS ONE 11(5): e0154087.