Cats Are Paying Attention to Your Feelings
Last Updated on Tuesday, December 08, 2015 12:20 AM
Published on Sunday, December 06, 2015 12:41 PM
Written by Moriah Galvan, MSc and Jennifer Vonk, PhD
Affectionately known as "man's best friend," domestic dogs are the apex of human companionship, selectively bred for tameness, obedience, and a human-centric disposition.¹
Research into the dog-human relationship yields a plethora of results highlighting the mutual understanding between humans and their pet dogs, which further cements the domestic dog as the affable sidekick in most popular media. Surprisingly, the cat-human relationship has not received the same attention from scientists, even though cats are the world's most popular pet.² ³
A search of comparative scientific research on cat behavior yields few results, with most research focusing on the biological traits of cats more than their behavior. We wanted to understand the human-cat relationship, and more specifically, investigate a topic that is currently generating debate in the canine literature: the ability of domesticated animals to understand and respond appropriately to human emotion cues, and why, or if, with regard to cats, such behavior occurs. Aside from their popularity in human households, cats were of interest to us for a variety of reasons: 1. Their history of domestication differs from that of dogs in several important ways; 2. They did not evolve to live in large, complex social groups as dogs have.
All domestic cats can be traced back to a single lineage of wildcats, F. sylvestris lybica, in the Middle East.⁴ These wildcats were asocial hunters, and likely self-domesticated to early human encampments to prey on rodents attracted to human waste. Humans tolerated the existence of the wildcats because they provided pest control. Unlike wolves, who also self-domesticated to human settlements, that could be tethered and selectively bred for guarding behavior, working service, and obedience to humans, domesticating wildcats was difficult as they could escape human confines and never provided a "working" service to humans.⁵ Even today, many cats are still allowed to roam unsterilized outside, which makes controlled breeding for desired traits, like gentle dispositions and/or obedience, more difficult.⁶ However, progress with the domestication of cats has occurred and will likely continue to proliferate in the future. But, keep in mind that most of the distinctive cat breeds we see today have been around for only a few hundred years. Differences in the domestication of dogs and cats can readily be observed when comparing the diversity of dog breeds to that of cats. Most cat breeds express more similarities than differences, whereas dog breeds range from the miniature, fluffy toy dog, such as a Lhasa Apso, to a large and shorter-haired working dog, such as the Doberman Pinscher or Great Dane. Dog breeds are bred selectively to fulfill particular roles; some of which may involve the need to respond to human communicative cues. Cats do not currently assume working/service roles and therefore may not have undergone the same selection pressure to understand human communicative cues.
So why would cats respond to human cues? Rather than being a co-evolved trait - in which dogs, but not cats, would hold the upper hand through their extensive coevolution with humans - we argue that understanding and following human cues is a learned behavior.⁷ Cats, just like dogs, live in human homes. It is necessary for them to understand human behavior because it fosters a mutualistic relationship. When humans are happy, the cat stands to benefit, through treats, extra back rubs, and a warm lap to sit on. When humans are angry, cats could experience negative consequences, like being yelled at or removed from the area that the human occupies. Therefore, regardless of the time span or co-evolutionary processes that may have occurred, the domestic pet's ability to understand humans is critical to a sustainable interspecies relationship. The two-stage domestication hypothesis outlines how this process can occur.⁸
The two-stage hypothesis proposes that two key elements are necessary for a non-human animal to sensitize to human cues.⁹ First, the nonhuman animal must be willing to accept humans as companions, a process that usually occurs early in development. For cats, this is between two and eight weeks of age. Second, this animal must have sufficient reinforcement for following and attending to human cues, such that they are increasingly likely to follow future cues. Research on domestic cats shows support for both of these stages. Support for the first stage comes from research on kittens. Kittens socialized at an early age actively seek out humans later in life.¹⁰ ¹¹ And, they have the ability to imprint on and recognize specific handlers. Research on adult cats finds them capable of differentiating their owner's voice from that of a novel human.¹¹ ¹² ¹³ For the second stage, a landmark study directly comparing the abilities of dogs and cats found cats as capable of following human gestural cues, such as a point directed at a bowl of food, as dogs.¹⁴ This leads us to believe that attending to human behavior, and learning its consequences, is within the cognitive abilities of domestic cats.
We conducted two experimental studies to investigate whether cats could differentiate human emotion cues. Picture A shows Smokey, one of the test subjects. Experiment 1 used only facial and postural cues for two emotions, happiness and anger¹⁵, exhibited by both a person the cat was unfamiliar with and the person that the cat was most familiar with. In picture B, the researcher demonstrates the angry and happy postural and facial cues used in the experiment. We included 12 sterilized adult cats, 7 male and 5 female, of mixed breed. For each trial, the cats, one by one, were placed in a pre-determined starting location two meters from the human, familiar or unfamiliar, in a room of the house they were comfortable in. The human sat cross-legged and presented the cues statically for three minutes, without using any additional cues, such as vocal or gestural cues. If the cat approached, the human did not offer any reaction, but merely maintained the given emotion expression. To analyze the results, we measured the time it took each cat to approach and "make contact" with the human, the amount of time spent in physical contact with the human, as well as the types of behaviors the cat displayed during the trial. These were later classified as negative, positive, or neutral-type behaviors as defined by earlier research.¹⁶ These measures were selected based on prior research with dogs. This research revealed that dogs displayed more avoidant behaviors and were significantly less likely to approach actors displaying angry expressions compared to when the actors displayed happy expressions.¹⁷ We reasoned that cats would display more positive behaviors, approach more quickly, and spend more time overall with a human in the happy condition than in the angry condition in which we expected the opposite - more negative/avoidance behaviors, slower approach and less time in contact. The factor of familiarity with the human was included to parse apart whether cats needed extensive experience with a specific human expressing varying emotions in order to recognize them, or if any learned associations between facial expressions and human behavior could be generalized to other humans. We also expected that we would see larger effects of the emotion on the cats' behavior when they were responding to their familiar caretaker, based on prior findings that both cats and dogs seem more attentive to owners compared to strangers.¹⁸ ¹⁹ ²⁰
The results of this first experiment determined that the cats spent more time with the familiar person and displayed significantly more positive behaviors, e.g., rubbing, purring, having a relaxed body posture, in the happy condition, versus the angry condition. But, this effect was observed only when they interacted with the familiar human. For the unfamiliar human, the cats did not alter their approach or behavior. This seems to indicate a learning effect, because even though the emotion expressions did not differ between the familiar and unfamiliar human, the cats seemed to respond differentially to the emotion cues only when presented by someone they had experience with; however, it may also be the case that the cats have greater overall interest in the emotion expressions of their owners. This result aligns well with the two-stage domestication hypothesis in that the cats had previous experience with the familiar human and had potentially been rewarded for attending to these emotion cues - such as receiving praise and cuddles when the human was happy - but had no history with the unfamiliar human and therefore did not behave differentially. Picture C shows a test cat approaching the researcher during the happy portion of the experiment.
Given the relatively subtle effects of human emotion on cat behavior in the first experiment, in Experiment 2 we attempted to make the emotion conditions even more salient by adding vocal cues in the form of conversations. This context more realistically embodied a real life interaction with a familiar and unfamiliar human. We included 26 sterilized adult cats, 15 male, and 11 female, of mixed breed. Each cat had two opportunities to interact with their owner and a different unfamiliar person (the researchers): once with a positive conversation taking place, and once with an angry conversation taking place. For each test, the cat was in his/her carrier on a table in the testing room, at a pre-measured distance from the owner and researcher, who were facing each other. See picture D. For the positive emotion trial, the owner and researcher conversed happily, while maintaining a relaxed posture while smiling at one another. During the negative emotion trial, the owner and researcher conversed angrily, while maintaining a stressed and tense body posture and displaying a frowning or grimacing facial expression. Voice volume was controlled across both trials. The cat's carrier door was opened at the beginning of each trial, allowing the cat to respond to the trial in any way he/she desired.
The measures we selected to analyze from the results of this second experiment were much the same as in the first experiment: we measured the time it took the cat to exit his/her carrier, the amount of time spent in contact with either the owner or the researcher, the behaviors the cat displayed - these were later classified as negative, positive, or neutral-type behaviors as defined by earlier research²¹ - and additionally, the cat's gaze direction throughout the trial. Gaze direction was important because human children often use their caregiver as a reference point when assessing an uncertain situation; this is known as social referencing. Dogs have been shown to engage in social referencing as well.²² We found that there were no significant differences based on the emotion of the trial for the cats' general behavior. These results are unsurprising given that cats are highly neophobic, and this experiment took place in a potentially scary research lab, where the cats may have been too nervous about the experience in general to be paying careful attention to what the humans were doing. Interestingly though, the cats gazed at their owner significantly more than at the researcher in both trials. This observation suggests that, although the cats may not have been picking up on the "emotion" of the study, they were looking to their owner to help them understand the experience. This observation confirms that cats also engage in social referencing, a result that was recently discovered by another research team.²³ Social referencing should occur only in species that learn from the emotion cues of others, so it is interesting to find such a behavior in a species previously considered "asocial."
There are two key results from the research we conducted. First, cats can differentiate human emotion cues in familiar humans, but the ways in which they differentiate are subtle. Second, cats display social referencing behavior, that is, they gaze significantly more at a familiar human in a novel situation for information on how to react.
We can confidently say your cat is paying attention to you and they are understanding you, at least with regard to gestural cues, emotion cues, and social referencing.²⁴ ²⁵ What we need to continue to study is how and why this behavior occurs. Our research supports the two-stage domestication hypothesis, but current comparative scientific research has other competing theories with new supporting research appearing regularly.²⁶ In order to strengthen our results, and also confirm our theory about why this behavior occurs, additional studies with larger samples of cats - with even more diverse backgrounds - will need to be conducted. We hope that one day our research will contribute to an entire body of work that helps us fully understand cats, which will lead to better training methods, welfare practices, and standards of care. In the meantime, we hope to offer you, cat-lovers, and all animal enthusiasts some cat-munition for your next "whose pet is better" debate.
Moriah Galvan received her Masters of Science in Psychology in 2014 from Oakland University in Rochester, Michigan. She works as a researcher at a market research firm, but spends most of her days looking for a career in animal welfare, her one true passion.
Jennifer Vonk received her PhD in comparative psychology from York University in Toronto, Ontario, Canada. Since then she has held faculty appointments at the University of Southern Mississippi and Oakland University, where she currently directs the Laboratory of Cognitive Origins. She and her students study the cognitive abilities of a range of species from frogs, cats, bats, bears and gorillas, amongst others.
4. C Driscoll, M Menotti-Ryamond, AL Roca, K Hupe, WE Johnson, E Geffen, EH Harley, M Delibes, D Pontier, AC Kitchener, N Yamaguchi, SJ O'Brien and DW Macdonald, "The Near Eastern Origin of Cat Domestication
317, no. 5837, July 2007, 519-522.
8. C Driscoll, M Menotti-Ryamond, AL Roca, K Hupe, WE Johnson, E Geffen, EH Harley, M Delibes, D Pontier, AC Kitchener, N Yamaguchi, SJ O'Brien and DW Macdonald, "The Near Eastern Origin of Cat Domestication
317, no. 5837, July 2007, 519-522.
11. EB Karsh, "The Effects of Early and Late Handling on Attachment of Cats to Humans," Conference Proceedings, 1983, Globe Press, St. Paul.
12. EB Karsh and DC Turner, "The Human-Cat Relationship," In D Turner and P Bateson (eds.), The Domestic Cat: The Biology of its Behavior, 1988, Cambridge, UK, Cambridge University Press.
15. P Ekman and W Friesen, Unmasking the Face: A Guide to Recognizing Emotions from Facial Expressions, 1975, Malor Books, Englewood Cliffs.
16. P Leyhausen, Cat Behavior: The Predatory and Social Behavior of Domestic and Wild Cats, trans. B Tonkin, 1979, Garland Publishing, New York.
20. EB Karsh, "The Effects of Early and Late Handling on Attachment of Cats to Humans," Conference Proceedings, 1983, Globe Press, St. Paul.
21. P Leyhausen, Cat Behavior: The Predatory and Social Behavior of Domestic and Wild Cats, trans. B Tonkin, 1979, Garland Publishing, New York.