Factors that impact the categorization of dogs' facial expressions

Dogs possess emotions, such as happiness, sadness, fear, surprise, anger, and disgust and studies have demonstrated that dogs use facial expressions to portray their states of emotionality (Anderson & Adolphs 2014 as cited in Bremhorst et al., 2021; Bloom & Friedman, 2013; Csoltova, & Mehinagic 2020; Kaminski et al., 2019; Kujala, 2018; Kujala et al., 2017; Siniscalchi et al., 2018). Researchers have also found that humans can accurately categorize dogs' emotions, using their facial expressions (Bloom & Friedman, 2013; Kujala et al., 2017). However, limited research has explored how factors like morphology and breed attitudes impact humans' categorization of dogs' facial expressions. Using stimuli pictures, this study examined how breed and morphology impacts our categorizations of dogs' facial expressions. Participants (35 female, 10 male, and 1 non-binary) met individually with the student researcher via ZOOM. During this session the participants first completed a demographic questionnaire and the Coleman Dog Attitude Scale (used to measure humans' attitudes towards dogs). Next, two sets of stimuli pictures were presented. For the first set of stimuli pictures individual photos of dogs' faces were shown on the screen, and participants were asked to indicate how likely they were to approach each dog. For the second set of pictures, the faces of two different dogs were presented on the screen at the same time, and participants were asked to choose which dog they would approach. Despite participants overall self-reporting that certain facial features are not important to them, results indicated that breed and morphology do play a role in humans' categorization of dogs' facial expressions. Participants indicated they were most likely to approach Cocker Spaniels (91.30%), Pugs (89.10%) and Beagles (87.00%) and least likely to approach Dobermans (with Pinned Ears) (43.48%), American Terriers (37.00%), and German Shepherds (30.43%). Relaxed/Friendly expressions (open mouth and relaxed eyes) were also preferred over neutral expressions (no facial tension, closed mouth and relaxed ears, with eyes focused directly forward) with significant differences found between the Friendly Golden Retriever compared to the Neutral Golden Retriever, X2 (1, n = 46) = 3.19, p < .001, Friendly German Shepherd and Neutral German Shepherd X2 (1, n = 46)= 75.6, p < .001 and Friendly Rottweiler and Neutral Rottweiler X2 (1, n = 46) = 12.1, p < .001 (See Table 7-9). As well, Friendly Golden Retrievers were perceived more positively than Friendly German Shepherds, X2 (1, n = 46) = 18.9, p < .001, and Friendly Rottweilers, X2 (1, n = 46), = 25.0, p < .001 (See Table 7). These findings indicate that humans may possess specific breed biases and additional research should explore the nature and origin of these attitudes. As well, human biases towards breed-related morphological differences could lead to significant welfare implications and potential differences in treatment and care of specific breeds.