Sheep Discriminate Faces, So What's In It For the Sheep?
An Animal Rights Article from


Marc Bekoff, Ph.D., Psychology Today - Animal Emotions
November 2017

I don't find the results of the current study all that "surprising." People often stereotype other animals, including so-called "food animals," as being dumb and unfeeling individuals. They unmind them but interestingly don't diminish the cognitive and emotional capacities of the companion animals with whom they share their homes. Viewing other animals as lacking sentience and deep feelings opens the door to incredible abuse on the way to people's plates.


A recent discovery that eight captive sheep can discriminate familiar from unfamiliar human faces from a variety of perspectives has rocked the world of those who study animal cognition as well as biomedical researchers and those interested in saving sheep from being served up as human meals. Mass media has focused on this cognitive capacity previously thought to be unique to humans, centering on a study by University of Cambridge (UK) researchers Franziska Knolle and her colleagues called "Sheep recognize familiar and unfamiliar human faces from two-dimensional images" published in the journal called Royal Society: Open Science.

The entire essay is available online and numerous lay summaries are readily available. Basically, the researchers discovered that the eight sheep, when presented with a familiar celebrity's face—either Barack Obama, British newscaster Fiona Bruce and actors Emma Watson and Jake Gyllenhaal—could discriminate the face they recognized from faces with which they were unfamiliar. Then faces were then rotated so that the sheep viewed them from different perspectives and once again, the sheep also could recognize the celebrity's faces from tilted images. This ability has previously been demonstrated only in humans, and the sheep showed the same decline in response shown by humans in similar tests.

Are these results really all that "surprising?"

A study conducted by Jonathan Pierce and his colleagues published in 2001 called "Human face recognition in sheep: lack of configurational coding and right hemisphere advantage" using more sheep and conditions showed sheep could discriminate among human faces but are better at discriminating among the faces of other sheep. Commenting on the current study, Dr. Pierce notes, “I guess they have extended our work to show that sheep generalize viewpoints of the faces, which does require a rich representation of the identity.”

I don't find the results of the current study all that "surprising." People often stereotype other animals, including so-called "food animals," as being dumb and unfeeling individuals. They unmind them but interestingly don't diminish the cognitive and emotional capacities of the companion animals with whom they share their homes. Viewing other animals as lacking sentience and deep feelings opens the door to incredible abuse on the way to people's plates. And, just because the ability to recognize tilted faces has not previously been shown in animals other than humans, this does not mean that other animals lack this capacity. More comparative studies are sorely needed and the results of the research under discussion show that humans are not unique in this cognitive ability. It's best to keep the door open before declaring we're unique in different cognitive and emotional capacities.

Should transgenic sheep be used to study diseases from which they don't usually suffer? Bioethical considerations

Dr. Knolle and her colleagues conclude that their data "show that sheep have advanced face-recognition abilities, comparable with those of humans and non-human primates." They also write, "As well as providing novel ethological insights, this paradigm furthermore provides opportunities for investigating cognitive dysfunction. Indeed, face perception may be impaired at multiple levels in neurodegenerative diseases such as Huntington's disease (HD) [45] and Parkinson's disease [46], as well as psychiatric disorders such as autism spectrum disorder [47] and schizophrenia [48]...The face-recognition paradigm presented here would be ideally suited for studying cognitive decline in the transgenic sheep model for HD." (The numbers refer to references in their essay.)

In an interview I did with Sarah Gibbens for an essay titled "Sheep Can Recognize Human Faces" published by National Geographic, the question about the ethics of using sheep to study diseases from which they don't usually suffer, such as Huntington's disease. Huntington's disease is a rare and horrifically debilitating neurological condition that is irreversible. Having known someone who suffered from it showed me just how debilitating it truly is. Of this, there is no doubt.

Let me say that I can well understand why some people would favor creating sheep who suffer from Huntington's disease, for example, because of their advanced cognitive abilities and large brains. However, I also would like to see a more open debate about whether or not other animals should be created solely to learn about these disorders and how well the animal models really work.

One person who favors these studies notes, "The lambs do not suffer during this process. The sheep are not treated any differently....The lambs wouldn’t show any signs of the disease until they are five or six months old, the age at which they would be slaughtered for market anyway."

I’m a skeptic on how important these animal models are for learning about human diseases. I see both an ethical issue in engineering sheep with a degenerative disease and a biological issue in how effective results derived from animal studies can be for human patients. I’m more of a fan of studying people to learn about people. I fully realize that others do not agree with this position, for which I'm not alone in arguing. This is why open discussions are badly needed concerning the ethics of using other animals and how good animal models truly are.

What's in it for the sheep?

A good number of people emailed me after the National Geographic essay appeared and asked questions that boil down to something like, "What's in it for the sheep?" or "What about the sheep?"

Comparative research in the field called cognitive ethology (the study of animal minds and what's in them) is constantly generating new data. We know that all sorts of animals who are used by humans display rich and highly developed cognitive and emotional capacities, and these data are generating many wide-ranging debates about if and how they should be used and abused in a wide variety of venues, very often "in the name of humans." For example, detailed research shows that cows are bright and emotional bovine beings displaying abilities that some call "surprising" (for more discussion please see "Cows: Science Shows They're Bright and Emotional Individuals" and links therein), yet they're killed by the millions for human meals. [Some people might claim that Temple Grandin's so-called "stairways to heaven" have solved the problem of pain and suffering experienced by cows on their way to killing floors of slaughterhouses. Even if a tiny fraction of individuals have a "better life," it's still a life filled with enduring trauma before they arrive at a slaughterhouse and when they're waiting to be killed, and doesn't border on what anyone would reasonably call a "good life." All in all, the "Temple Grandin Effect" is not very effective at all. For more on how Temple Grandin's methods fail millions of individuals please see​ this essay and references therein.]

So, what's in it for the sheep? I'd like to think that the results of the study of facial recognition along with others that show clearly that sheep suffer and feel pain will be used on their behalf and result in the development of rigorous regulations on how they can be used for human ends. I honestly would like to see an end to their being used for food, for example, because of the incredible abuse to which they're subjected.

Where to from here?

The study of facial recognition in sheep has clearly opened the door for all sorts of discussions and debates about what we're learning in the general field of cognitive ethology. I look forward to more comparative research about the rich and deep cognitive and emotional capacities of other animals, and discussions about how we must use this information on the animals' behalf because we're not very good in doing this. [As in many other venues in which nonhumans are routinely and brutally abused, detailed information from scientific studies is not used on their behalf. Unfortunately, a "knowledge translation gap" still exists and what we know is not used on their behalf in far too many situations. Basically, the knowledge translation gap refers to the practice of ignoring tons of science showing that other animals are sentient beings and going ahead and causing intentional harm in human-oriented arenas. On the broad scale, it means that what we now know about animal cognition and emotion has not yet been translated into an evolution in human attitudes and practices.']

It's also essential to discuss whether or not other animals should be used in invasive biomedical research because they're readily available or because we can create individuals to serve our purposes, and whether or not they should be used in other human-centered venues.


Franziska Knolle, Rita P. Goncalves, A. Jennifer Morton. Sheep recognize familiar and unfamiliar human faces from two-dimensional images. Royal Society Open Science, 2017; 4 (11): 171228 DOI: 10.1098/rsos.171228


Return to Animal Rights Articles