The extended social phenotype

Abstract Pet keeping is a uniquely human behavior among primates that may hold important clues about how we develop and express social abilities like empathy, attachment, and emotion recognition. In this commentary, we argue that close, emotionally invested relationships with pets should be more fully integrated into models of human social cognition. While foundational work in the field clearly recognized the relevance of both conspecific and heterospecific agents, most empirical studies have focused on interactions between humans. We suggest that sustained human–pet relationships offer an ecologically valid and biologically grounded context in which social-cognitive capacities are expressed, refined, and potentially shaped. We introduce the concept of the extended social cognition phenotype, proposing that social abilities are not limited to human–human interactions but also extend to emotionally charged, long-term bonds with non-human social partners. These bonds exhibit six core features, which each mirror aspects of human kinship: care-taking, naming, baby-talk, play, attachment, and exclusion from consumption. While similar in structure, pet bonds may differ in their conditionality, involving investment decisions shaped by perceived emotional reciprocity and cost. We explore how such relationships engage neurocognitive systems involved in human bonding, including oxytocin signalling and limbic network plasticity. Importantly, we highlight clinical evidence from neurodevelopmental and neurodegenerative disorders, such as autism spectrum disorder and frontotemporal dementia, where relationships with pets may serve as compensatory scaffolds for disrupted social cognition. In these populations, pets often provide stable, predictable, and emotionally salient interaction partners that may enhance motivation for social engagement and foster residual or alternative pathways for empathic and affiliative behaviour. These findings invite a reversal of the conventional framing: rather than viewing pet-keeping solely as a product of human cognition, we propose that it actively participates in shaping the human social mind across development and pathology.