Close encounters of the cat kind: The influence of context and sex on facial signaling proximity in domesticated cats (Felis silvestris catus)

Animals, such as domesticated cats, establish and maintain various types of social bonds through frequent and affectionate interactions, which can include behaviors like allogrooming, allorubbing, and co-sleeping. In our previous study, we found that these behaviors are facilitated by rapid facial mimicry, a process that allows individuals to connect with one another both physically and psychologically. When measuring social bonds among animals, researchers have often relied on spatial proximity measurements, and recently, there has been a movement toward using AI-based approaches to improve accuracy in these assessments. However, such approaches have not yet been applied to the study of domestic cat affiliative behavior and social bonding, despite their large degree of social flexibility. In our current study, we use AI-based computer vision systems to investigate the relationship between social context (affiliative vs. non-affiliative), sex composition (female-female, mixed-sex, male-male), and spatial proximity during intraspecific cat facial signaling interactions. Based on our understanding of social bonding in domesticated cats, we developed two testable predictions: (P1) that cats will be closer during affiliative facial signaling interactions than during non-affiliative ones; and (P2) that adult sex composition affects communicative proximity, with female–female dyads showing the closest proximity, followed by female–male/male–female, and finally male–male dyads. Our results supported P2 but not P1. Interestingly, cats showed significantly greater spatial proximity during non-affiliative facial signaling interactions than during affiliative ones. Our study reveals the utility of AI-based approaches in the study of cat behavior, while also presenting a more complex and nuanced social profile of cats.