Picture a living room, late on a Tuesday, phone screen throwing cool light across someone’s fingers as they scroll through hundreds of contacts.
It feels like connection. The data says something different. All those contacts are busy, present, and mostly strangers.
What is actually happening inside that feed, and why does the type of person on the other end matter so much?
How a stranger on your feed triggers a different response than a friend does
When you interact with someone you already know in real life, your brain draws on a rich file of shared context: their tone of voice, the way they laugh, a decade of dinners. A message lands inside that framework. Misreading their words is possible, but the existing relationship corrects for it quickly.
Strangers online carry none of that scaffolding. Researchers suspect that one reason interacting with strangers on social media is associated with loneliness is because of social media’s high potential for facilitating misinterpretation. Each ambiguous exchange can also feed idealization of other people’s friendships, which exacerbates social comparison, and, as co-author Jessica Gorman notes, “this idealization is possibly stronger when those friendships involve people you’ve never met because there is no personal experience to counter that idealization.”
Those costs accumulate fast when roughly one in three of your contacts is someone you have never met. That compounding effect leaves people feeling more isolated, not less.
What the average American’s social media contact list actually looks like
Most people picture their friends list as mostly people they know. The reality is messier. Around 35 percent of contacts were people who had never been met in person, roughly one in three who exists only as a profile picture, a handle and a string of text exchanges.
Ten platforms were covered: Facebook, X, Reddit, YouTube, LinkedIn, Instagram, TikTok, Snapchat, Pinterest and WhatsApp. That span captures everything from close-knit family group chats to the open-comment battlegrounds of large public feeds, all counted together as social media use.
The 35 percent figure is almost certainly an average that conceals wide variation. Someone deep in a Discord community could be running a stranger ratio well above 60 percent, while a user who mainly trades photos with family in a private group sits near zero. The gap between those two people’s loneliness risk, the research suggests, may be larger than almost anyone assumed.
The numbers behind the study and where the data came from
Oregon State University scientists published the findings in Public Health Reports, describing it as the first study of its kind to focus on how stranger contacts specifically affect loneliness. The sample covered more than 1,500 adults aged 30 to 70, a group largely ignored in earlier research, which focused almost entirely on teens and young adults, people who aren’t teens or young adults comprise 75 percent of the US population.
Having more interactions with relative strangers was linked to increased loneliness, and the relationship was linear: the higher the proportion of never-met contacts, the worse the outcome. Those with the highest stranger proportion were more than twice as likely to report loneliness compared to those with the lowest.
The flip side is just as striking. Connecting online with people you actually do know was not linked with greater loneliness, but neither was it associated with decreased loneliness. Only the strangers moved the needle, and they moved it the wrong way. As GCN has previously reported, only 22 percent of US adults have a close friend 15 or more years their senior or junior, which means the pool of genuinely close contacts most people can draw on is already surprisingly narrow.
Where the finding gets complicated and what the data does not settle
The study calls for future research to examine the directionality of the relationship between stranger contacts and loneliness, and to explore nuanced factors such as platform type, reasons for connection and expectations of emotional support. Future research is needed because lonely people may actively seek out stranger-heavy spaces, which would reverse the arrow of cause and effect entirely.
Platform design may matter enormously. A subreddit where someone builds a reputation over months is a fundamentally different environment from an algorithmic short-video feed where the same clip reaches a million strangers simultaneously. Both count as social media, and the study does not separate them.
There is also a generational wrinkle. Adults in their 30s and 40s typically built contact lists in an era when friending someone implied some prior real-world connection. Adults who came of age on TikTok built theirs differently, treating parasocial bonds with strangers as normal and meaningful. Whether the stranger penalty holds equally across both groups remains open.
What the pattern means for how Americans try to build social lives going forward
The practical implication is not to leave social media. It is to treat contact list composition as seriously as the volume of activity. A feed packed with strangers, even warm and generous ones, appears to work against the very feeling it seems to promise.
Study leader Brian Primack said that “people experiencing loneliness may wish to examine critically their interactions with strangers on social media and to prioritize in-person connections over social media ones, even when those social media connections are considered close.” That last phrase is the sharpest part: even contacts users label close can function as strangers if the two have never actually met.
The research adds a structural dimension to a trend GCN has tracked across several surveys: digital contact timing and medium shift its social meaning in ways people rarely account for consciously. Knowing the stranger ratio carries a measurable association with loneliness gives users, and potentially designers, a concrete number to weigh against every new follow.
