shelter pets in need
Each order helps pets in need
Researchers have found that dogs can learn the names of objects, respond most strongly to the scent of a familiar person, and notice what people do around them. No study can tell us what a dog is thinking at a given moment. This page covers what researchers measured, in how many dogs, and what each result does and does not show.
They do things that researchers count as thinking, and the best-known test involved one dog. In a 2011 study in Behavioural Processes, John Pilley and Alliston Reid worked with a Border Collie named Chaser. The paper reports that she "learned and retained, over a 3-year period of intensive training, the proper-noun names of 1022 objects."
The authors ran four experiments. In the second, they mixed three commands with three names, and Chaser "understood the separate meanings of proper-noun names and commands." In the third, she learned three common nouns, which the paper describes as "words that represent categories." In the fourth, she learned words by "inferential reasoning by exclusion", which the paper defines as "inferring the name of an object based on its novelty among familiar objects that already had names."

One dog is a small group, and the study says only what Chaser could do. Alexandra Horowitz, who leads the Dog Cognition Lab in New York City, described the result in a NOVA question and answer: Chaser "learned 1,022 words" because her person used them "reliably and regularly" when talking to her.
Researchers cannot read a thought, so they measure what a dog does or what its brain does and ask what the result fits. Alexandra Horowitz, a professor at Barnard College, leads the Dog Cognition Lab. She and Brian Hare are among the scientists who study dog cognition, which the Aspen Institute describes as asking "what a dog knows, understands, and believes."
Research run at Emory University and reported by Mic, published in Behavioural Processes in 2015, used fMRI on 12 dogs that had been trained to stay still in the scanner without sedation or restraint. The dogs smelled five scents: their own, a familiar human, a strange human, a familiar dog, and a strange dog. The paper reports that "the olfactory bulb/peduncle was activated to a similar degree by all the scents" and that "the caudate was activated maximally to the familiar human."

The authors tie the caudate to positive expectations. The person whose scent was used was not the dog's handler, and the paper says the caudate response differentiated the scent "in the absence of the person being present." The authors wrote that the dogs "had a positive association with it." That is a measured brain response to a smell, and the study does not go further than that. Our post on why dogs choose to sleep on one person over another covers what research says about that choice in the home.
In some tests the results look like a human one, and in others they do not. The Chaser paper says she "acquired referential understanding of nouns, an ability normally attributed to children." Horowitz adds a limit in the NOVA question and answer: "Dogs are markedly better equipped to understand nouns and verbs. Adjectives, adverbs, and other tricky parts of speech appear to be largely beyond their ken." She also cautions that "dogs' behavior (and thus psychological condition) does not necessarily map identically to our own, no matter how much we might wish that this were so."
Dogs can sound as if they are trying to talk. Horowitz described one theory about barking: people may have selected for barkier dogs because they made better guards, and some dogs may have barked more because they were trying to mimic speech sounds. She added, "This is only a theory, and it doesn't mean that barks = dog speech."
Feelings are harder to test. Writing in The Guardian in 2015, John Bradshaw said that scientists who studied animal behavior once focused on what animals did and not on how they might feel. He describes dogs as "incredibly fast and flexible learners," and our blog has a list of facts about these swift learners.
One test looked at the "guilty look." In a 2009 study in Behavioural Processes, Horowitz videotaped 14 dogs. Trials varied whether a dog had the chance to disobey a command not to eat a treat while its person was out of the room, and whether the person knew what the dog had done. The results "revealed no difference in behaviours associated with the guilty look" across those trials. More of those behaviors appeared when the person scolded the dog, and the effect of scolding was stronger when the dog had obeyed. The authors concluded that the look is better described as a response to the person's cues than as an appreciation of a misdeed.

That is one behavior in 14 dogs, and the paper's conclusion is limited to what the look means.

Many dogs twitch their paws or make small sounds in their sleep. The American Kennel Club says dogs have similar sleep cycles to people: "Like people, dogs and other animals go through several sleep cycles. There are periods of wakefulness, followed by rapid eye movement (REM) sleep, and non-rapid-eye-movement sleep." It says REM sleep "is when you might notice your dog twitching or making sounds," and that "it’s impossible to have all the answers on what dogs might dream about."
The AKC also warns about waking a dog from a bad dream: "dogs can react aggressively toward the person waking them. This can be dangerous, especially for children." It advises waiting for the dog to wake up and then being there to comfort it.
If your dog groans or moans in its sleep, why dogs moan and groan explains the common causes and when a groan is worth a call to your vet.
At Animal Hearted, we love dogs, offer apparel and gifts for dog lovers, and help save shelter and rescue pets! Treat yourself and your furry friend by checking out some of our collections!
Horowitz calls dogs "near-anthropologists among us, watching us, observing our movements, noticing our habits." She gives an example: how a person gets up from the couch to go to the fridge looks different from how they get up to go for a walk, and "dogs notice that."
Some words, yes. Horowitz says dogs "can attend to specific words, if you use them carefully" and "can understand many words, if we talk to them clearly." Chaser is the example she gives, and the 2011 paper tested her on object names and commands.
One study suggests they can. In a 2014 PLOS ONE paper, researchers tested 36 dogs while their person showed affection to what appeared to be another dog, and compared it with the same attention given to non-social objects. The dogs showed "significantly more jealous behaviors" when the person was affectionate toward the other dog, such as snapping, getting between the person and the object, and pushing or touching the object or the person. The authors say the results "lend support to the hypothesis that jealousy has some 'primordial' form" that exists in at least one other social species besides humans.
Leave a comment