Original research paper:
Pongrácz, P., Szapu, J. S., & Faragó, T. (2018). Cats (Felis silvestris catus) read human gaze for referential information. Intelligence.
For humans, gazing is usually just another form of ‘pointing’. However, this is not such a simple thing if you think about. When someone indicates with pointing that ‘something interesting is there’, or ‘I want that something’, this is not simply putting a mark on an object in the vicinity. In case of humans, showing/pointing involves the communicative intent of the signaller that is also processed by the receiver. The interesting part comes when we provide pointing cues to animals, and they are able to follow these (usually by finding something at the ‘end’). Do they also conceptualize our intent to show them something? Or did they just learn that it is worthy to go that way where we point at? Or in the simplest case: do they simply choose something that is near to our pointing hand/finger? Gazing is even more special because a head that is turned to a given direction is really not such a big change to the silhouette of the body compared to, for example, a sideward pointing arm. Therefore, to be able to follow a gazing cue the animal more likely has to possess some kind of experience, knowledge about the way how humans use their face/eyes, they have to know that when we ‘look at’ something that may likely have some kind of importance.
Inter-specific communication (in our case: human-animal communication) is very interesting, because in general communication evolved for signalling purposes among conspecifics. Therefore, when we find evidence that there is an effective communication going on between humans and animals, we can investigate what could be the driving force behind such rare phenomena. Gazing cues (when someone wants to ‘show’ something to another person) are very important in human communication, but animals use them very rarely – for example because gazing can be an agonistic signal, or simply they just do not share this kind of referential information with each other. It was found earlier that dogs are very good in gaze following, and the researchers hypothesized that during their domestication dogs benefited from being able to read various human-specific forms of communication, because dogs were selected for co-exist and cooperate with humans. CATS at the other hand were directly not selected for any kind of cooperative work activity, however they are still very successful and ‘satisfying’ as companion animals. Contrary to dogs, cats were domesticated from an ancestor with solitary lifestyle, therefore in their case it is especially interesting to investigate, whether they possess comparable socio-cognitive skills to dogs, which enable them to be a successful companion for humans. So our goal was not mainly to compare dogs’ and cats’ capacity to follow human hazing, but to see whether cats can do it. If they can, this can add to our theory that domestication and becoming a companion animal may require the development of human-compatible skills of communication in various animal species.
Today it is widely accepted among ethologists that dogs are excellent subjects for comparative cognitive research, because their natural environment is the HUMAN GROUP, therefore it is not only easy to test them, but also it is very exciting what kind of evolutionary processes made dogs capable to blend in so well to this new social structure. Cats however provide a less clear-cut picture for the first glance. There are opinions still that cats are truly not even fully domesticated. Probably many scientists still shares the common belief that cats are less interesting for cognitive research because they do not really care whether humans want to interact with them or not – cats live at our property, but cats “don’t live WITH us”. Furthermore, it is less convenient to work with cats than with dogs. Not to mention that cats also have a ‘bad reputation’ for many scientists as a species that can easily damage ecosystems by its predatory behavior, so this may also be a reason why not to test whether they have sophisticated cognitive skills or not.
In the present study we tested companion cats in the homes of their owners. An unfamiliar experimenter provided two types of gazing cues (head turns) in a two-way choice experiment, where the cat had to find hidden food in one of the two pots placed left and right in front of the kneeling experimenter. The experimenter always looked at the baited pot. The gazing cue was either a sustained, or a momentary one. Parallel to this, two types of attention eliciting signals were give a priori to the gazing cue: ostensive signals (the experimenter called the cat’s name, and talked to it); or non-ostensive signals (the experimenter emitted clacking sounds, normally not used to call cats’ attention in Hungary). When the gazing cue was given, the cat’s owner released the animal that was allowed to choose and investigate one pot. As a total, 24 trials had to be completed by the subjects (6 from each combinations of gazing cue and attention eliciting). Briefly, we found that companion cats are very successful in human gaze following, no matter whether it was a sustained or a momentary head turn. Their performance was not affected by the presence of verbal (or ‘ostensive’) attention eliciting. However, ostensive cues made it easier to establish eye contact with the cat a priori to the gazing signal. The success rate of cats was similar to the performance of dogs in other studies. At the other hand, the drop-out rate of subjects (who did not cooperate, or gave up testing) was much higher in case of cats than it would be with dogs. For successfully testing 41 cats we had to start the procedure with near 100 subjects.
Gaze following in cats is almost surely not an innate capacity, however the fact that they can spontaneously (without specific training) learn to pay attention and follow human gazing is most likely based on their enhanced sensitivity for human socialization, and this is because of the process of evolution since their domestication. Intelligence in animals (or in other words, the biological definition of intelligence) means that the individuals are capable of using their learnt skills in various ways, even in new situations. It is unfortunate that people are quick to stick bad stereotypes to cats, especially when comparing them to dogs. Our research showed (again) that cats are very well adapted to the role of being a problem-free companion of humans, which includes their capacity to learn spontaneously from human behavior. People are willing to share their resources with animals that they like – because the animal is useful, or the animal is cute etc. However, if an animal has hard-to-tolerate behavior, it will not fare well on the long run. Cats blend in very well to the various lifestyles of humans, and for this they need the capacity of showing human-compatible social skills. Since domestication, intelligence in cats could be at least partly selected to serve this process of adaptation to humans.
Cats in Reuters: