Empathy Affects How People Perceive Human and Canine Facial Expressions

Psychological factors, including empathy and personality, affect how humans perceive canine and human facial expressions, according to a study recently published in PLoS ONE.

Facial expressions are important for nonverbal social communication in both humans and dogs. These expressions allow for an understanding of others’ emotions, moods, and attitudes. Interestingly, dogs and humans share similar muscles that control facial expression.

Humans generally believe that dogs experience emotions. Previous studies have reported how humans can perceive a dog’s emotions based on its bodily movements. However, perceiving emotion from a dog’s facial expressions has proven more difficult.

Psychological factors, such as empathy and personality (eg, extroversion, neuroticism), affect how humans perceive human facial expressions. It is not well known, though, if these factors influence human perception of canine facial expressions.

For this study, 34 volunteers were shown images of 30 dog faces, 30 human faces, and 20 control images of household objects or pixels; the dog and human faces were categorized previously as pleasant, threatening, or neutral. The volunteers rated each image’s valence (intrinsic attractiveness or aversiveness) and arousal and noted how many of 6 emotions (happiness, sadness, surprise, disgust, fear, anger/aggression) each image contained.

After viewing the images, volunteers completed questionnaires related to empathy toward humans and animals, personality (extroversion, neuroticism), and dog expertise and exposure. Responses were used to evaluate the effects of psychological factors on ratings.

Valence Ratings

Valence ratings were higher for pleasant human (PH) faces compared with pleasant dog (PD) faces and for neutral dog (ND) faces compared with neutral human (NH) faces. Valence ratings were lower for threatening dog (TD) faces compared with threatening human (TH) faces.

Arousal Ratings

For human and dog faces, arousal ratings were significantly higher for threatening than for pleasant or neutral faces; images of pixels and objects elicited little to no arousal. Notably, arousal ratings were higher for PH faces than for PD faces.

Emotion Ratings

Generally, emotions were rated similarly for dog and human faces. Emotions receiving the highest ratings, according to categorization, were happiness (pleasant), anger/aggressiveness (threatening), and sadness (neutral).

Comparing human and dog faces, the authors made several observations:

• Happiness ratings were higher for PH faces than for PD faces.
• Sadness ratings were higher for NH faces than for ND faces.
• Fear and anger/aggression ratings were higher for TD faces than for TH faces.

Results of these comparisons suggest an ecologically meaningful bias to perceive conspecific faces positively and non-conspecific faces negatively.

Effect of Psychological Factors

Emotional and Cognitive Empathy

Emotional empathy (sharing emotion) exerted a clear influence on how volunteers perceived threatening and pleasant human and dog facial expressions. For example, emotional empathy correlated positively with anger/aggression for both species. It did not, however, affect the perception of neutral faces for either species.

Cognitive empathy (cognitive reasoning without sharing emotion) influenced the perception of human faces but not dog faces. Interestingly, it correlated negatively with arousal for objects and pixels. These findings, the authors wrote, “suggest that cognitive empathy can fine-tune the estimation of emotions from human faces … but also has a ‘reality-check’ effect on considering inanimate objects’ emotions.”

Personality

Extroversion correlated positively with anger/aggressiveness for NH faces. Neuroticism correlated negatively with happiness for PH and NH faces.

Dog Expertise and Exposure

Dog expertise and exposure correlated positively with valence and happiness for ND faces and with higher arousal for PD faces.

Conclusions

Taken together, the study’s findings demonstrate that humans perceive canine and human facial expressions similarly and that psychological factors affect this perception. The authors speculated, though, on whether anthropomorphism plays a role in perceiving animal emotion.

Dr. JoAnna Pendergrass received her doctorate in veterinary medicine from the Virginia-Maryland College of Veterinary Medicine. Following veterinary school, she completed a postdoctoral fellowship at Emory University’s Yerkes National Primate Research Center. Dr. Pendergrass is the founder and owner of JPen Communications, LLC, a medical communications company.