Sometimes the media report a scientific study that makes you think, “Duh. Everybody knows that. Why waste time and money studying that?”
That was my initial reaction to the online publication by MedicalXPress (formerly the medical strand of PhysOrg) of “Domestic dogs display empathic response to distress in humans” (June 7, 2012).
As a live-in companion of four Golden Retrievers over the last twenty-five years, I don’t need to be convinced how emotionally sensitive and empathically supportive dogs are when people around them are in distress. When you’re down, few things could be more comforting than a cold nuzzle from a warm Golden.
Yet science quite properly discounts anecdotes, and all of the tales I could produce would not add up to empirical evidence. For that, you need a real experiment.
The MedicalXPress article reports just such an investigation, Empathic-like responding by domestic dogs (Canis familiaris) to distress in humans: An exploratory study, published in Animal Cognition and made available online by Goldsmiths, University of London (May 29, 2012).
Beyond its direct results, the study, conducted by Deborah M. Custance and Jennifer Meyer, is a fine example of the proper way to do observational research that depends on both interpretation of behaviours and tentative conclusions based on those behaviours.
Many writers criticize observational studies for their liability to error. Theoretical or confirmation bias, cherry-picking of data, confusion of correlation and cause, and overblown conclusions are just the most obvious conceptual reefs around which scientists must navigate.
Empathic-like responding … (ELR) begins to steer carefully around the dangers inherent in the methodology from the very first words of the study’s title. While the popular press may prefer headlines like “Dogs know when you’re sad,” ELR‘s laudable caution starts with “empathic-like.” Before we read even a paragraph, we know that we’re dealing with cautious and self-restrained researchers, and we can examine their work with some confidence that we won’t be wasting our time.
In the abstract, ELR begins with a summary of the findings: “An experimental protocol first used with human infants was adapted to investigate empathy in domestic dogs. Dogs oriented toward their owner or a stranger more often when the person was pretending to cry than when they were talking or humming.”
Immediately, the authors distance themselves from the sensational conclusion that dogs have empathy just like people do. They tamp down that sort of overstated conclusion: “The dogs’ pattern of response was behaviorally consistent with an expression of empathic concern, but is most parsimoniously interpreted as emotional contagion coupled with a previous learning history in which they have been rewarded for approaching distressed human companions.”
In other words, and with admirable scientific restraint, the authors report that their results could be construed to show the canine equivalent of human empathy, but that such a conclusion is not proved by the available evidence. In fact, later in their report, Custance and Meyer stress that their results rather point to an area for further research than they prove a specific conclusion on their own.
ELR uses a standard definition of empathy: ““the naturally occurring subjective experience of similarity between the feelings expressed by self and others without losing sight of whose feelings belong to whom.” While they acknowledge “emotional contagion,” the authors differentiate between personal distress, which is self-directed, and empathy, which is other-directed. Their study is structured to discriminate between these two response states. (For a full outline of their methodology, see the study.)
Sensitivity to others’ distress has been established in many animals, from monkeys to rats to cats. Post-conflict “consolation” behaviour has been observed in apes, rooks, and dogs, although it is difficult to pin down the exact nature of these actions.
Of particular interest for the present research is that, while most instances of empathy-like behaviours in animals are intraspecific, anecdotal reports abound that domestic dogs exhibit intraspecies empathy-like behaviours with humans.
To distinguish between merely habitual or self-directed empathy-like behaviours, the researchers exposed eighteen medium-sized domestic dogs of a variety of breeds to situations in which a stranger begins to cry. The dogs are in their own homes, and their owners are present.
What will the dog do? If the dog’s response is to seek comfort for itself, it is likely to approach its owner for support. If the dog’s response is merely the result of curiosity, then having the stranger hum rather than cry should provoke similar behaviour by the dog. If the dog misreads the crying, it could approach the stranger playfully, or in a dominant attitude, or even aggressively.
What happened? Fifteen of the eighteen dogs responded to the crying by approaching the stranger, while only six dogs responded to the humming. What’s more, thirteen of the fifteen dogs that approached the stranger did so in a submissive manner, that is, in a sympathetic manner. One dog approached alertly, and one dog approached playfully.
In raw numerical terms, then, 83% of the dogs approached the crying stranger, and 85% of those dogs did so submissively.
The majority of dogs in the present study behaved in a manner that was consistent with empathic concern and comfort-offering. The dogs responded to their owner and the stranger when they were crying in a markedly differently manner compared to when they were humming or talking.
And, “the fact that the dogs differentiated between crying and humming indicates that their response to crying was not purely driven by curiosity.”
What did Custance and Meyer conclude? Again, the cautious discipline of keeping your claims within the scope of your data is foremost: “Even if the dogs’ pattern of response exceeded what one would expect of personal distress and egotistic comfort-seeking, it does not automatically follow that they were empathizing in the sense of making a self-other differentiation.”
A more parsimonious explanation of their behavior is that they may have previously received positive reinforcement for approaching crying individuals. Any household dog who approaches a distressed human family member is likely to be positively reinforced by receiving affection. Through the process of generalization, any human who then cries in the presence of that dog is likely to initiate a conditioned approach response. Since the dog is nonetheless affected by emotional contagion the response will still tend to be submissive in its emotional tone. Thus, the behavioral outcome is a response to human distress that is consistent with an expression of empathic concern, but which may not actually involve the requisite self-other differentiation needed for it to count as true empathy.
Similarly, the authors concede that “there is no compelling evidence to suggest that the dogs’ behavior indicated sympathy or cognitive empathy. ” Since dogs can’t give overt verbal indications of their actual cognition, “it is difficult to imagine what behavior a dog could produce under such circumstances which could convincingly indicate mental perspective-taking.”
As a result, Custance and Meyer concede that “we in no way claim that the present study provides definitive answers to the question of empathy in dogs.” However, they feel justified to state that their study “sets out a profitable direction for further study.”
As I’ve indicated throughout this article, I very much like this kind of restraint. Even the most intriguing results must be framed in ways that are consistent with a sober and thorough consideration of their true scope.
Oh, and by the way, I’ll bet you a zillion dollars that both of the Golden Retrievers in this study responded with sensitivity, sympathy, and empathy to their crying strangers.
Count on it.