Dogs comprise one of the most physically diverse species on the planet. But without humans, would dogs naturally range in appearance from Pug and Poodle to Great Dane and Weimaraner? Probably not. Have humans selected dogs for particular physical appearances?

The conventional wisdom is that dogs have been selected to look cute. Nobel laureate ethologist Konrad Lorenz proposed that human and non-human animals have particular physical features which encourage humans to nurture and attend to them. These features correspond with neoteny, or juvenilization, and could include “[a] relatively large head, predominance of the brain capsule, large and low-lying eyes, bulging cheek region, short and thick extremities, a springy elastic consistency, and clumsy movements” (Lorenz, 1950). The preference for baby-like features is called kindchenschema, or baby-schema. Pugs, Shar Peis and many others exemplify dogs who retain “cute,” juvenilized appearances into adulthood.

But could there be more to the story than simply preferring juvenalized features? For example, humans also ascribe meaning to the way dogs look, as shown by, among other things, American Kennel Club breed standards. The expression of the Great Pyrenees is described as “elegant, intelligent and contemplative,” and the Chihuahua is said to have a “saucy” expression. The meanings we ascribe to dogs are very much intertwined with the way dogs look. And of course, this phenomenon can be explored in fun studies. 

Researchers in Alexandra Horowitz's Dog Cognition Lab in New York City investigate the general topic of anthropomorphic attributions to dogs. For example, does the so-called “guilty look” map to a dog’s knowledge of disobedience or is it related to owner behavior? (Here’s a short video showing one condition from Alexandra Horowitz’s 2009 study on the “guilty look”.)

I had the pleasure of hearing Julie Hecht (see also) recently present their research at the 3rd Canine Science Forum in Barcelona, Spain (see also). In their study, participants viewed two nearly identical images of mixed-breed, adult dogs and implicitly selected which image in each pair they liked best. Unbeknownst to the participants, one image in each pair had been slightly altered to explore a particular physical feature. Some features related to neoteny, while others related to human-like traits. Peoples’ choices indicated their preferences, or lack of preferences, for particular physical features.

They found that the story does not begin and end with neoteny. Participants preferred some but not all characteristics of neoteny. For example, larger eyes were preferred over smaller; but an enlarged cranium was not. Participants also preferred human-like traits such as distinct smiles or colored irises.

They dug deeper and found that preferences differed based on preconceptions about animals. In contrast to self-ascribing “animal people” those who self-characterized as “non-animal people” did not have a preference either way for eye size, or a distinct smile.

The physical features of dogs to which humans are attracted aren't trivial matters. These findings can have important welfare implications. Do our preferences for particular physical features always benefit our canine companions? While “animal people” seem drawn to dogs with larger eyes, big eyes can be associated with health issues like brachycephalic ocular syndrome and exposure keratitis. And although participants didn’t implicitly prefer larger nostrils, what if that physical attribute would enhance many dogs' ability to breathe?

Clearly, humans have implicit preferences about the physical features of dogs and these preferences can be systematically investigated. I look forward to more research on this fascinating topic. 

I thank Julie Hecht, Alexandra Horowitz, and Paul McGreevy for their input on this essay.


Hecht, J., & Horowitz, A. 2009. Physical prompts to anthropomorphism of the domestic dog. Third Canine Science Forum, Barcelona, Spain.

Horowitz, A. 2009. Disambiguating the “guilty look”: Salient prompts to a familiar dog behavior. Behavioural Processes 81: 447–452.

Horowitz, A. C., & Bekoff, M. 2007. Naturalizing anthropomorphism: Behavioral prompts to our humanizing of animals. Anthrozoös 20: 23–35.

Lorenz, K. 1950. Ganzheit und Teil in der tierischen und menschlichen Gemeinschaft. Reprinted in Studies in Animal and Human Behaviour, 1971, vol. 2, 115–195, ed. R. Martin. Cambridge, MA: Harvard University Press. 

You are reading

Animal Emotions

Fishes Show Individual Personalities in Response to Stress

Trinidadian guppies display complex individual and consistent coping strategies

Choose a Dog Trainer as Carefully as You Would a Surgeon

Dog training is an unregulated industry in which dogs and humans suffer.

Do Wolves Understand Cause and Effect Better Than Dogs?

The headline of an essay asserts they do, but dogs aren't dumb-downed wolves