I feel your pain

20 September 2013 by Malcolm Campbell, posted in Biology

“We’re sick” Elliott in E.T. The Extraterrestrial, screenplay by Melissa Mathison (1950- )

Sometimes empathy can be a real pain.

This week was a rough week for our dogs. Lameness and illness struck in equal measure. Despite the fact that they are non-human members of the family, whenever the dogs are under the weather like this, it wreaks havoc on the emotional state of at least one of their human caregivers. Canine health woes create an atmosphere that sits between unease and malaise.

Of course, this empathic connection works in the positive direction as well – when the dogs are in high spirits, the mood is infectious and we revel in it too.

This incredible capacity for empathy is one of the many wonderful things about being human.

There’s a memorable scene in the movie “E.T. The Extraterrestrial” that showcases the human capacity for empathy. In this scene, the eponymous extraterrestrial is making itself at home in the residence of the young protagonist, Elliott.  At the same time, Elliott is attending school. The scene works movie magic in revealing the emotional connection between the two, despite their geographical separation. They simultaneously experience inebriation, surprise, passion, and rebellious indignation on account of the alien’s experiences.

Later in the movie, a dialogue occurs between government scientists and Elliott’s brother, Michael. In trying to determine E.T.’s ability to communicate, one of the government scientists asks, “Elliott thinks his thoughts?”. “No,” replies Michael, “Elliott feels his feelings.”

The empathic connection between the two lead characters, and our own empathic connection with Elliott, is a real strength of the film. None of us question Elliott’s emotional connection with E.T., as we can imagine ourselves behaving equivalently in the same position. Remarkably, it is also likely that none of us question E.T.’s emotional connection with Elliott. E.T. is a creature of whom we have no prior knowledge, who does not exist in reality, and yet we have no issues ascribing human emotions to E.T..

Anyone who lives with a non-human companion animal will not be surprised by the notion that humans ascribe emotions to non-humans. People do it all the time. “Oh look, the cat is happy.” “That’s one angry dog.” And so on. Happiness, anger, guilt, malice, you name it. People will tell you that their companion animal has any number of what we normally think of as distinctly human emotions.

Why is it that humans do this?

The most obvious answer to this question is that animals truly possess these emotions. There is a large, and growing, body of evidence to suggest that this is the case.

The other answer is that, for whatever reason, humans are wont to ascribe human psychological traits, including emotions, to non-humans.

An ideal way to test this hypothesis would be to determine if humans describe objects that are decidedly non-human as having human traits. It so happens that the outcome of such a hypothesis test was described back in 1944.  Using a cunning animation of geometrical shapes, Marianne Simmel and Fritz Heider of Brown University asked undergraduate students to share their perspectives on the actions of the shapes.

The video that Heider and Simmel used is wonderful in its simplicity. It involves two filled triangles, one slightly large than the other. It also involves a filled circle, and a box with a flexible opening, like a door. The animation involves the movement of the two triangles and the circle in the vicinity of each other, and in relation to the box. This simple explanation does injustice to the complexities of what are seen in the animation.

The video in question can be seen here:

Heider & Simmel Experiment

In their classic 1944 paper, Heider and Simmel report on three experiments they conducted with the animation. In the first experiment, 35 female undergraduate students were asked to develop a narrative based on their understanding of the animation. Notably, 32 of the students gave the triangles and circle human identities. Two other students described them as birds. Only one student described the shapes as simply shapes. This said, in all cases, the students provided the shapes with agency. In all cases, the shapes were said to control the motion of the door in the box. The box itself was generally seen as a room, which the shapes chose to enter or exit. The narratives commonly indicated that there was a fight between the two triangles; that the larger triangle chased the circle; that the circle ran away from the larger triangle; and, that the smaller triangle and the circle played with each other. That is, the shapes all were said to have motive or response ascribed to their actions.

Having established that the students saw the shapes as animate characters within a drama, a second group of students were asked to attribute personalities and motives to the shapes.  The students had no problem doing this. The large triangle was typically seen in a negative light, as an aggressive, quarrelsome bully. By contrast, the circle was generally viewed as timid, shy or fearful. The smaller triangle was cast in the most positive light, as a brave, defiant hero. The smaller triangle and the circle were seen as allies, who express joy when reunited. This contrasts with the larger triangle, whose interactions with the circle were seen as manipulative or coercive. Crucially, the interpretation of the social interactions between the shapes was altered by watching the animation in reverse. In a third experiment, a different group of students viewed the video run in reverse. Here, intent and motive were altered. Nevertheless, the same character traits were retained by the shapes.

Heider and Simmel’s experiments beautifully illustrate human’s amazing capacity to anthropomorphise. Even if objects are normally inanimate, let alone non-human, we have little difficulty in providing them with human agency – personifying them. Our capacity to personify, to anthropomorphise, is almost obscenely anthropocentric.

Why assume that non-humans will act in a human manner?

Evolution provides us with answers.

In the first instance, ascribing human agency to non-humans is one way to avoid conflict with other animals. We share with animals commonalities in the way we transmit neuronal signals, and in many of our physiological responses. Unsurprisingly, this body “wiring” means that we also share similar behavioural patterns. By and large, aggression generally looks like aggression irrespective of the species involved. Similarly, other basic behaviours, like fear, stealth, and playfulness, share similar attributes across many species. In a world where appropriate interpretation and action based on such behaviours can mean the difference between eating and not eating, or being eaten and not being eaten, it is perhaps not surprising that we ascribe our emotions and motives to other animals.

Another reason that may explain our tendency to anthropomorphise may lie in the evolution of our social nature. As is the case with a great number of other species, the capacity to socialise in groups has been an arguably great evolutionary advantage to Homo sapiens. Coordinated, collective action has enabled us to feed, house, clothe, and protect ourselves in ways that might be near impossible as individuals. The ability to understand and respond appropriately to the emotions and motives of others of our species has been key to our capacity to work together.

It may be that our anthropomorphising is a by-product of an exquisitely honed, innate sensitivity to cues that we use to infer emotion and motive. This acute, innate sensitivity allows us to interpret cues on-the-fly, and thereby react in a manner that reinforces our social bonds. Contagious yawning provides a beautiful illustration of this in-built mechanism.

Contagious yawning is prevalent in humans. In humans, 45-60% of healthy adults are susceptible to contagious yawning. Merely seeing another person yawning is enough to induce a yawn. In fact, sometimes the mere use of the word “yawn” is enough to induce a response. The response is rapid, and, as its name implies, it is contagious within groups. Critically, people are most susceptible to contagious yawning when it occurs within close social circles – the closer the individuals the higher the likelihood of “catching” a yawn. In keeping with this, clinical, psychological, behavioural and neurobiological studies support the view that contagious yawning is related to social interactions, particularly empathy. In fact, contagious yawning is increasingly being used to understand the nature of empathy.

Contagious yawning is not something we learn. It is an innate ability to pick up cues from another person. Moreover, contagious yawning is an innate ability that invokes an involuntary empathic response that reinforces our social bonds. Contagious yawning suggests that we are an innately empathic species. Should it be any surprise that we extend this innate ability to other species, or even other objects? Given that we are so finely honed for empathy, it is hardly surprising that this trait might spill over beyond our species and extend to others. It may be read as anthropomorphising, but at its core, it merely may be an extension of empathy.

Notably, we are not unique in our innate empathy, at least as far as contagious yawning is concerned. Other primates, including chimpanzees, bonobos and gelada baboons yawn contagiously when they observe other members of their species yawning. As with humans, with these other primates, the closer the social bond between individuals, the higher the likelihood of yawning when another yawned.  These observations emphasise the empathic nature of contagious yawning, as empathy is biased towards individuals that are most familiar, or socially close.

Strikingly, contagious yawning suggests that we are not alone when it comes to empathising with others outside our species.

It turns out that dogs are also contagious yawners. With dogs, the interpretation of contagious yawning is somewhat complicated by the fact that yawning is correlated with alleviating psychological tension or mild stress in dogs. Nonetheless, some have suggested that even this function of yawning may be correlated with empathy in dogs, as a means by which dogs reduce stress in social groups – a so-called “calming signal”.

Recently, Teresa Romero and colleague investigated contagious yawning and its link to empathy in dogs.  As with primates, including humans, dogs were more likely to catch the yawn contagion when the yawn came from another within their social circle. Critically, the “social circle” in question here was their companion human. Twenty-five dogs were studied. Of the thirteen dogs that yawned in the study, they yawned contagiously when humans had a “real” yawn with a yawning noise, as opposed to counterfeit mouth movements. In fact, these findings confirmed earlier research by Karine Silva and colleagues, who found that the sound of humans yawning was enough to induce contagious yawning in dogs. On the basis of the fact that yawning dogs maintained a consistent heart rate, contagious yawning did not appear to be a “calming signal” related to managing anxiety or stress.  In keeping with the contagious yawning being related to empathy, dogs were most likely to yawn when their companion human was someone with whom they were more familiar – someone with whom they had a social bond.

What is particularly striking about contagious yawning in dogs is that it appears to be more acute when the yawn is “caught” from a human, rather than from a member of their own species. That is, dogs appear to be particularly attuned to the human cue, and have an empathic response to reinforce their social bond with us. As their response to yawning seems to differ contingent on whether the yawn transmitter is dog or human, dogs appear to have bettered us in the empathy department.

Where we tend to anthropomorphise – ascribing human traits to a non-human – dogs interpret cues contingent on the species of origin. Dogs don’t superimpose the canine experience onto the human experience. They don’t “cynomorphise”, or “canine-ify” our actions. Instead, they are able to read the human state, and empathise according to our emotional status.  Clearly, this has been to their advantage over their history as companion animals. Tuning into, interpreting, and empathising with our emotional states has been key to reinforcing social bonds between our species, and made them the valued companions they are today. We could learn a thing or two from dogs about how to interpret, and empathise with the lives of our fellow animals.

Even during a week when empathy for our dogs is taking an emotional toll, it seems a small price to pay knowing that the rest of the time they are being far more attentive to us.


Buttner AP & Strasser R (2013) Contagious yawning, social cognition, and arousal: an investigation of the processes underlying shelter dogs’ responses to human yawns. Animal Cognition 1-10.

Campbell MW & de Waal FB (2011) Ingroup-outgroup bias in contagious yawning by chimpanzees supports link to empathy. PLOS ONE 6: e18283

Demuru D, Palagi E (2012) In bonobos yawn contagion is higher among kin and friends. PLOS ONE 7: e49613

Harr AL, Gilbert VR & Phillips KA (2009) Do dogs (Canis familiaris) show contagious yawning? Animal Cognition 12: 833-837

Heider F & Simmel M (1944) An experimental study of apparent behavior. The American Journal of Psychology 57: 243-259

Joly-Mascheroni RM, Senju A & Shepherd AJ (2008) Dogs catch human yawns. Biology Letters 4: 446-448

Norscia I, Palagi E (2011) Yawn Contagion and Empathy in Homo sapiens. PLOS ONE 6: e28472

Palagi E, Leone A, Mancini G & Ferrari PF (2009) Contagious yawning in gelada baboons as a possible expression of empathy. Proceedings of the National Academy of Sciences 106: 19262-19267

Platek SM, Mohamed FB & Gallup Jr, GG (2005) Contagious yawning and the brain. Cognitive Brain Research 23: 448-452

Romero T, Konno A & Hasegawa T (2013) Familiarity bias and physiological responses in contagious yawning by dogs support link to empathy. PLOS ONE 8: e71365

Silva K, Bessa J & de Sousa L (2012) Auditory contagious yawning in domestic dogs (Canis familiaris): first evidence for social modulation. Animal Cognition 15: 721-724

Images: All photographs by Malcolm M. Campbell.

Acknowledgement: This post was inspired, in part, by a wonderful piece by Jason Goldman in his superb Scientific American blogThe Thoughtful Animal”. In his piece, Jason beautifully describes the Heider and Simmel experiments, and their implications. Jason took a particular angle on his piece, leaving space for some of the ideas discussed here. Hopefully the two pieces complement each other, and any redundancies underscore the elegance of the research done by Heider and Simmel.


Leave a Reply

+ 5 = eight