Humans' mask wearing has limited effect on family dogs' behaviour in standard test situations

COVID-19 changed our lives in many aspects. Among the most spectacular changes was probably the mandatory wearing of masks, which was proven to negatively influence human social interactions and communication. The various COVID protocols, however, not only affected humans but also had a huge impact on companion animals, such as dogs, living in human society. For example, it is particularly alarming, that throughout the pandemic the number of registered dog bites increased significantly. The phenomenon has been explained by the generally elevated stress level as well as family members and dogs spending more time together in restricted closed space. On the other hand, the communication deficit caused by the constant usage of masks cannot be ruled out as a further contributing factor. In the current study, we aim to test the effect of human’s mask wearing on dog’s behaviour in a range of situation. To this end we used previously validated standard test situations (responsiveness to human pointing, basic obedience, spontaneous following, emotion recognition, threatening approach) in which the human experimenter was present either with or without wearing a mask. N=21 family dogs were tested in a within subject design, with a minimum of 3-day difference between the two occasions. Tests were carried out in different randomized sequences for each subject to eliminate the order effect. The order of the masked and maskless occasions was randomized as well. We found that the mask-wearing of the experimenter did not influence dogs' performance in cognitive tests (responsiveness to human pointing, basic obedience, spontaneous following, emotion recognition; all p>0.05). The only significant difference observed was in dogs' reactions to the masked versus non-masked experimenter during the threatening approach situation. Reaction scores were coded on a 1–5 scale, ranging from friendly to aggressive. The number of more aggressive responses was significantly higher for the masked experimenter compared to the non-masked. These results have two important implications. First, it seems that (at least in the situations investigated) dogs' cognitive performance is not affected by the experimenter’s mask wearing, supporting the notion that research conducted during COVID protocols enforcing mask-wearing remains valid. Second, and perhaps more importantly, dogs seem to react with more aggression towards unfamiliar people wearing masks in ambivalent situations. Therefore, special attention needs to be devoted to dog attacks when these regulations are in place. Further research should address additional factors, such as the familiarity of the interacting (masked) human.