Jaining Li and Nichol Wong, at the Education University of Hong Kong, published a study in October 2025 in Scientific Reports that looked at exactly this pattern, pet owners who lean on animals specifically in place of people, and what it does to their wellbeing. The answer wasn’t the comforting one I expected going in. This is a report on that research and related studies, not advice, and if isolation or loneliness has been sitting heavily for a while, a doctor or therapist is a better resource than an article.
What the study actually found
Li and Wong compared pet owners against non-owners and looked specifically at living arrangement and loneliness as the mechanism connecting pet ownership to wellbeing. Among people living alone, pet owners reported less loneliness than non-owners, and that reduced loneliness explained most of the wellbeing benefit pets seemed to provide. So far, that’s the reassuring story most people already assume.
But the researchers also measured something more specific: how much of a person’s attachment to their pet functioned as a substitute for human relationships, rather than a supplement to them. That interpersonal substitution dimension didn’t predict better wellbeing. It predicted worse wellbeing, and the entire effect ran through increased loneliness. Leaning on an animal specifically in place of people wasn’t a workaround for loneliness. In this data, it was associated with more of it. The study is cross-sectional, a single snapshot rather than a study following the same people over time, so it can show that substitution and loneliness travel together without proving which one comes first.
Why later life specifically
There’s a reason this pattern shows up so often in discussions of people over 60 rather than people in their twenties. Retirement removes the incidental daily contact a workplace generates without anyone having to try. Widowhood, when it happens, removes the single most central relationship in a person’s social world at once. Adult children move away, old friends move into care facilities or die, and the pool of people available for the kind of low-effort, frequent contact that keeps loneliness at bay shrinks in ways that are largely outside anyone’s control. A pet’s company doesn’t ask anything of a shrinking social calendar. It’s simply there, on a schedule that doesn’t depend on anyone else’s availability, which is exactly the quality that makes it appealing and exactly the quality that can’t substitute for what a shrinking human network used to provide.
Not everyone gets the same trade
A separate study, published in 2025 in BMC Public Health, sampled 232 pet-owning adults over 45 in Taiwan and found the benefit of time spent with a pet depends heavily on who’s doing the spending. For people lower in neuroticism, more time with their pet was associated with higher wellbeing. For people higher in neuroticism, more time with their pet was associated with lower wellbeing, the opposite direction. Anxious attachment style worked as a similar moderator. The researchers noted that people high in attachment avoidance tend to keep some distance from their pets in the first place, which is a detail worth sitting with if the picture forming here is “socially withdrawn people simply substitute an animal for a person.” The people most likely to actually be uncomfortable with closeness don’t reliably direct that discomfort toward pets either.
Pets aren’t only a withdrawal mechanism
A large study complicates the substitution story from the other direction. Lisa Wood and colleagues, publishing in PLOS ONE in 2015, surveyed 2,692 residents across four cities in Australia and the United States and found pet owners were significantly more likely than non-owners to get to know people in their own neighborhood. Dog owners specifically were about five times more likely than owners of other pets to have met someone through their animal, and around a quarter of pet owners who met people this way went on to consider one of those people an actual friend, not just an acquaintance. In that role, a pet was a bridge toward more human contact, not a replacement for it. That sits oddly next to the substitution finding, and I think the honest reconciliation is that a pet can function either way, bridge or replacement, and which one shows up probably depends on the person and the relationship, not on pet ownership as a single, uniform category.
Why the appeal is genuine either way
None of this requires treating a preference for animal company as some kind of deficiency. Pets offer something specific that human relationships often don’t: unconditional attention without social comparison, judgment, or the ordinary friction of reciprocal obligation. That’s a real, legitimate appeal, and it doesn’t need a substitution mechanism to explain why someone over 60, having spent decades managing complicated human relationships, might genuinely prefer the more straightforward version an animal offers. If loneliness itself has become the harder part of that, rather than just a preference for a quieter kind of company, a doctor, a local seniors’ service, or a community program is a more useful next step than anything in this article.
What I’d take from this
The distinction that actually seems to matter isn’t whether someone loves their pet a great deal. It’s whether that attachment is standing in for human connection or running alongside it. The research so far, still cross-sectional and still limited to a handful of samples, suggests those are genuinely different situations with different outcomes, even when they look identical from the outside, which is one person, one dog, and a quiet house.
