A personal admission: This research is killing me. It’s messing with every memory I hold dear about the late, great Bear—aka the best cocker spaniel ever—who was my pal and stalwart companion for 17 years, my solace after bad dates, first child and flower pup at my wedding. While not the brightest light in the forest, she was certainly the most enthusiastic and open, telegraphing her every emotion to the world. Snarfing a forbidden cookie, she was blissed; finding a blue jay’s feather on the lawn and carrying it around in her mouth as if it were the bird itself was a transcendent moment of pride and joy. And when she was bad and told to go sit in the corner, she would drag herself, head down, ears dusting the floor, eyes squeezed shut, totally dejected. I was always sure that she thought that if she couldn’t see me, I couldn’t see her either. I knew she felt just terrible, wracked with guilt, for whatever she’d done.
Of course, none of this is strictly true—not according to science, at least. I’m not alone in thinking this way; witness what a friend emails me: “I do believe I know how Charlie is feeling much of the time. He makes strong eye contact. He has a happy face and tail and a sad one. On the rare occasions he's bad (ate a soft nail file recently) and when I scolded him he knew by my tone that I was mad and he sulked away with his tail down and wouldn’t look at me. Guilty!” It won’t surprise you that my friend counsels kids; note her description of “strong eye contact.”
Owners are sure of their perceptions
As it happens, my friend and I are very much part of the mainstream: The vast majority of dog owners firmly believe that their furry pals not only exhibit a full range of primary emotions but secondary emotions as well. That’s what a study by Paul H. Morris and his colleagues found through interviewing owners of animals (including dogs, cats, horses, rodents, and birds). They asked the owners to identify the primary emotions animals experienced (anger, fear, surprise, joy/happiness, anxiety, love and curiosity) as well as secondary emotions (empathy, shame, pride, grief, guilt, jealousy and embarrassment.)
Among dog owners, 88% reported all of the primary emotions, with joy and love being identified by 99% and 97%, respectively; 87% reported sadness. Among the secondary emotions, jealousy (81%) and guilt (79%) were the most observed, with 64% of owners thinking their dogs exhibited empathy.
A second study, this time with 40 dog owners, asked participants if they thought their dog exhibited jealousy; to give examples and descriptions of jealous behavior; and to provide alternative explanations other than jealousy for what the dog did. All the participants identified the cause of jealousy as an intrusion in the relationship between owner and pet—22% mentioned cuddling someone else—either a person or another animal—as one trigger; more than half noted that their jealous pet pushed him or herself between the owner and the intruding party. And 67% of those asked couldn’t come up with an alternative explanation for such behavior other than jealousy.
But is jealousy really what’s going on? I certainly thought so when my daughter was born and Bear was jealous, heartbroken, and—according to my vet—depressed, having to compete with this intruder in her life. Bear was listless, ate little, and kept to herself for months. It was only when I stopped carrying the baby—the way I’d once cuddled Bear—and the two of them were on the floor together as my daughter crawled, that Bear became her old self again, realizing that she hadn’t lost me after all.
But is there an alternative explanation?
Is it all human projection?
That humans have long attributed emotions to dogs is clear from The Odyssey, the first extant masterpiece of Western literature—in which the first one to recognize the long-wandering and disguised Odysseus is his dog, Argos. Are we all suffering from some collective kind of anthropomorphism, projecting feelings and thoughts onto our hapless canine companions, or are Charlie and whoever is lying at your feet really feeling guilty or jealous?
It’s an excellent question, which Alexandra Horowitz and Julie Hecht of the Dog Cognition Lab at Barnard College are trying to answer. And what they have discovered is absolutely fascinating. Let’s look at guilt first, specifically that “guilty look,” since owners are adamant that each pup has her or her own way of expressing it. As one person wrote me, “If I get home and she’s not wagging at the door to greet me, it’s a sign she’s done something she’s not supposed to. It’s just a question of going through the house and finding out what it is. She’ll be lying on the living room, pretending to be invisible.” “My dog is a bluffer,” says another, “But if I get angry, he’s instantly contrite.”
Does a dog feel guilt?
... or, put differently, display guilt? Despite how convinced owners are that they do, the scientific jury is still out. One study, published by Alexandra in 2009, looked at 14 dog-and-owner pairs. The owners instructed the dog not to eat a treat and then left the room; upon returning, they were told by the experimenter whether the dog had been “good” (left the treat alone) or “bad” (eaten it). The owner then greeted the dog normally or scoldingly, and the dog’s behavior was observed. There were four trials but, unbeknownst to the owners, the experimenter switched up the information, saying the dog had eaten the treat when he hadn’t and vice versa. The bottom line? Dogs exhibited that guilty look whether they’d snarfed the treat or not. In fact, innocent dogs displayed more guilty behavior than their guilty brethren. The experimenters opined that the guilty look might be behavior either caused by scolding or in anticipation of a scolding.
So, are dog owners just reading into it?
For those of you still needing validation of your deepest canine beliefs, there’s hope from another experiment conducted by Hecht and others. In this experiment, owners filled out an extensive questionnaire and then their dog was put in a room that had a table with a piece of hot dog on a plate. The owner made it clear to the dog that the hot dog was for people, not puppies, thus establishing a rule which, if broken, should elicit that guilty look. The owner then left the room.
Later, judging just from the greeting owners received from their dogs when they returned, 40 out of 54 of them were on the money whether the pooch had gone for it, but 14 were not. So maybe it’s not projection after all, but actually canine guilt? The researchers surmised that once a dog figures out that guilty looks will get him or her some lenience, the pup is highly motivated to offer one. The fact remained that “guilty” behaviors were demonstrated by both innocent and guilty dogs alike.
So, are we projecting or is fair to say that our pooches both know us and know how to play us? You tell me.
Morris, Paul H., Christine Doe, and Emma Godsell,” Secondary emotions in non-primate species: Behavioural reports and subjective claims by animal owners,” Cognition and Emotion (2008), 22, 1, 3-20.
Horowitz, Alexandra, “Disambiguating the ‘guilty look’: Salient prompts in familiar dog behavior,” Behavioural Processes (2009(, 81, 447-452,
Hecht, Julie, Adam Miklosi, and Marta Gacsi, “Behavioral assessment and owner perceptions of behaviors associated with guilt in dogs,” Applied Animal Behaviour Science (2012), 139, 134-142.
Horowitz, Alexandra and Julie Hecht,”Looking at Dogs: Moving from Antropocentrism to Canid Umwelt,” pp.201-219 in Domestic Dog Cognition and Behavior, ed. A. Horowitz (Berlin: Springer Verlag, 2014).
Copyright © Peg Streep 2015
VISIT ME ON FACEBOOK