Skip to main content

Verified by Psychology Today


What Dogs Understand but Cats Cannot

Many cats aren't even sociable enough to allow testing.

Key points

  • There is an ongoing debate over the relative intelligence of dogs versus cats.
  • The history and nature of dogs suggests that they will be more responsive to social signals than cats.
  • Many cats are not even sociable enough to allow testing.
  • Although dogs respond to a pointing gesture accurately, few cats in a recent study responded above chance.
kevin turcios / Unsplash
kevin turcios / Unsplash

The most popular domestic pets in America are dogs and cats. Dogs are somewhat more popular overall, with 38 percent of households in the U.S. having a dog while 25 percent of households have a cat. The number of actual animals is a bit more evenly matched since people who own cats are more likely to own more than one; there are about 77 million dogs and 58 million cats. Any conversation between dog and cat owners is almost inevitably bound to bring up the question of the relative intelligence of the two species.

Social Intelligence

A recent research report by a group of Hungarian researchers headed by Attila Salamon at Eötvös Loránd University in Budapest has revisited the question of dog-vs.-cat intelligence, focusing on the issue of whether dogs or cats understand human communication gestures better. Technically this refers to the relative "social cognitive ability" of canines and felines.

In many respects, evolution and the history of the domestication of cats and dogs should give us a clue as to the likely outcome of this comparison. From some 14,000 years ago—a likely estimate of when domesticated dogs began to appear—dogs have worked in close association with people. They assisted in hunting, guarding, and herding, as well as becoming close companions with humans. All of this requires some sort of interaction and communication. Dogs had a genetic leg up when it came to such behaviors because the ancestors of dogs lived in close family groups and packs that had a system of complex social interactions. Furthermore, it seems likely that the dogs that responded most readily and accurately to human communicative signals would be appreciated more and thus better cared for, and would be more likely to be bred. Thus a sort of seat-of-the-pants applied genetics would ultimately sort out dogs so that those who had inherited good social cognitive and communication abilities would begin to dominate the species.

The domestication of cats came later, probably beginning around 7000 years ago when organized agriculture was becoming important for humans. Now the growing of grain was carried out more systematically, and granaries stored the wheat, barley, and rye following the harvest, thus providing a food supply throughout the year. Rats and mice soon discovered these food reserves. Unfortunately, these rodents do not simply eat some of the food, but they foul much of the remainder making it not fit for consumption. Because of this, cats were a godsend. Working in almost complete independence of human guidance, cats protected the granaries and fed themselves on the rats and mice.

While dogs were living socially and dependently with people, it was not so with cats. The ancestor of the cat was a strongly territorial, mostly solitary animal with minimal contact between individuals except during reproductive periods. In essence, the newly domesticated cat only had a weak social association with humans, although their territories often included human settlements. It is only in the last few hundred years that cats have become companion animals, living more socially and interactively with humans.

Testing Social Intelligence

A common test of social cognition involves pointing. By the age of two, human children recognize that pointing is a communication gesture indicating the direction of something of interest. For animals, a pointing test is quite simple: First, the animal is brought into a testing room and shown a bowl that contains a bit of a treat, which it is then allowed to eat. This is done for several trials. Next, two identical bowls are put down, one to the experimenter's right and the other to her left, and she points to one container. The experimenter always points to the container containing the treat. If the animal recognizes that this is a communication gesture they go to the container with the treat and get to consume it. A successful trial is one in which the animal goes to the container that was pointed to. An error would be going to the wrong container, or failing to make any choice at all.

For dogs, laboratory testing was a breeze. They all rapidly became comfortable in the lab, and all made a choice when the experimenter pointed her finger. If the experimenter held her finger pointing at one container all of the dogs in the group (except one) chose accurately at a rate well above the level which would indicate simply guessing.

Cats are Harder to Test

For cats, the situation was much more difficult. When it comes to simply habituating to the lab, so that the cat would be calm, take food, and interact with the experimenter, this research team found that cats are really difficult subjects. In the first set of tests, 60 percent of the cats failed to calm down enough to be tested. They simply couldn't handle it.

The researchers concluded that the territorial nature of the cats was such that they were at a disadvantage in the unfamiliar environment of the laboratory. So they reran the tests in the cat's usual home. Now a higher percentage of the cats proved to be testable but still 40 percent had to be dropped from the study because they wouldn't habituate to the presence of the experimenter, even in their own home. That meant that they didn't respond, refused to make any choices, hid behind furniture, or even bit the hand of the experimenter.

Cats Don't Get the Point

The performance of the cats was rather poor even for those select individuals who could be tested and did make choices. To quote the experimenters, "If you count failing to choose as an error cats did not perform significantly above chance."

Remember, these felines were not simply representative of all cats, but as the experimenters note, "We acknowledge that the cats in our study were probably more sociable than typical cats due to the requirement to interact with the experimenter during the habituation process."

The accuracy of the cats was abysmal, even using a relaxed criterion for scoring (namely not counting a failure to respond as an error). The experimenters summarized their overall performance: "Only three cats (7 percent) made a choice at least half of the time and were above chance at home and in the lab." That means that the remaining 93 percent of the group of cats tested were guessing, rather than gathering information about where the treat was based on where the experimenter was pointing.

Although these results do not resolve the issue as to whether dogs or cats are the more intelligent overall, it certainly indicates that when it comes to social cognition and the ability to respond to human communication, the performance of cats lags well behind that of dogs—even when you take into account the fact that many cats simply refuse to engage in the testing process.

Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission.

Facebook image: DimaBerlin/Shutterstock


Salamon A, Uccheddu S, Csepregi M, Miklósi Á, Gácsi M. (2023) Dogs outperform cats both in their testability and relying on human pointing gestures: a comparative study. Scientific Reports, 13, 17837.

More from Stanley Coren PhD., DSc, FRSC
More from Psychology Today
More from Stanley Coren PhD., DSc, FRSC
More from Psychology Today