Harnessing AI-based computer vision to evaluate biological and behavioral variables in dog-dog play interactions

Social play among mammals is thought to be crucial for the development of socio-cognitive and motor skills. Trait matching, or the tendency of individuals to seek partners who share similar biological and behavioral traits, could enhance the benefits of play by allowing mammals to accurately and reliably predict each other's behavior. As a result, this could lead to smoother social interactions. To test this idea, we explored the preferences of domesticated dogs during intraspecific open play bout sessions, as dogs are known for engaging in high levels of social play even into adulthood. In line with other studies on intraspecific play in dogs, we examined whether trait matching affected play duration and the average distance between dogs. We specifically predicted that trait matching would lead to longer play times and closer average proximity. We analyzed the effects of both biological (size, age, sex) and behavioral (head-to-head contact, head-to-tail contact, rearing history) traits on duration and distance. We did not find any evidence to support our predictions. Given the null results of our study and the recent mixed findings from other research on intraspecific dog play, it seems that other biological and behavioral factors (such as energy levels and personality traits) may play a role in trait matching. This, in turn, could influence the average duration and distance of play, although additional research is needed to confirm these ideas.