Whether online connections count as real friendship is a question that turns out to have a more specific answer than most people expect. New research from Oregon State University makes a distinction that changes the practical takeaway: it is not social media use in general that is linked to loneliness, but connection with strangers specifically. Understanding what the research actually found matters more than the usual advice to choose depth over visibility.
Nobody Asked to See Their Follower Count
I spent more than two decades sitting with people near the end of life, and I can tell you something I never once witnessed: a person asking, in those final days, to know how many followers they had. The number of connections on their profile. How many people had liked their posts. What I did see, more times than I can count, was someone asking for a specific name. One person. Not a platform, not a network. A name. That gap, between the number of connections a person carries and the specific human being they actually reach for when things become real, is roughly where this question lives. Are social media friends real friends? The research that came out in 2026 has a more careful answer than the question usually gets.
What the Oregon State Study Actually Found
In May 2026, researchers at Oregon State University published findings in Public Health Reports from a nationally representative study of more than 1,500 adults between ages 30 and 70. The lead researcher, Brian Primack, a professor in OSU’s College of Health, and his colleagues set out to examine something that most prior research had not: how the specific composition of a person’s social media contacts, not just their total use, relates to loneliness.
The findings drew a line that the general coverage mostly flattened out. Connecting online with people you do not know in person was associated with increased loneliness. Connecting online with people you already know was neither protective against loneliness nor a source of it. The effect ran in one direction, and only when strangers were involved. In the study population, roughly 35 percent of social media contacts were people the participants had never met in person.
Primack described the finding this way: people experiencing loneliness may wish to examine their interactions with strangers on social media critically, and to prioritize in-person connections even when some of those social media connections feel close. That last phrase matters. The problem is not simply the number of strangers in someone’s network.
Why Strangers Specifically
The mechanism the researchers point to is idealization. Social media interactions with strangers present a curated surface, and there is no personal knowledge to push back against it. Co-author Jessica Gorman put it directly in the research release: the idealization of other people’s friendships is possibly stronger when those friendships involve people you have never met, because there is no personal experience to counter it. The stranger online is always performing for an audience that cannot verify anything. The person you already know is not.
What follows from idealization is social comparison, and social comparison at a disadvantage. When someone looks at the apparent closeness and ease of relationships among strangers online, they are comparing their interior experience, which they know in full, to an exterior presentation, which has been selected for. The result is a gap that tends to widen the longer it is observed. This is not a flaw in the individual doing the comparing. It is a predictable outcome of the conditions the platform creates.
“What I kept noticing was not that people had too many connections. It was that none of the ones they had accumulated could actually show up.”
The social media friends and loneliness link, in other words, is not about total screen time or overall use. It runs through a specific dynamic: a felt closeness to people who are not close, sustained by content that makes their lives look like something to be measured against. The platform is not the problem. The composition of who is in the network, and what that composition does to a person’s sense of how their own life compares, is where the actual mechanism sits.
What the Brain Does With the Difference
A 2023 brain-imaging study from Sigmund Freud University in Vienna, published in Brain Sciences, adds a layer to this that is worth knowing. Researchers led by Peter Walla used EEG to measure neural responses when participants were shown names of their real friends, names of influencers or celebrities they followed closely, and names of people they felt no connection to at all. The results were unambiguous: the brain processed real friends categorically differently from parasocial connections, even when the subjective feeling of closeness and trust was similar. Real friend names produced the strongest brain activation of the three conditions.
What this means for the question of whether online friends vs real friends are equivalent is that the nervous system does not treat them as equal, even if the person consciously does. A relationship that feels close but lacks the reciprocal, verifiable history of a real friendship registers differently at a level below deliberate thought. This is not a judgment. It is simply what the data showed. The brain appears to be tracking something that the feelings are not always reporting accurately.
The Takeaway Is Not What Most People Suggest
Most of what exists on this question online takes one of two shapes. Either it is a personal essay urging readers to choose depth over visibility, which tends to stop at the level of inspiration without saying anything actionable, or it is a broad argument for deleting accounts, which misreads what the research actually found. The OSU study did not find that social media use in general produces loneliness. It found that the stranger-contact specifically does. The practical question is therefore not whether to quit but what the network actually contains.
A more honest audit of whether social media connections function as real friendship might start with something like this:
- How many of the people you follow or are connected with could you call if something actually went wrong? Not message. Call.
- When you look at someone’s account and feel a drop in your sense of how your own life is going, do you know that person, or are they someone you have only ever seen through their posts?
- For the connections that feel close: has that closeness been tested by anything real? A disagreement, a need, a silence that required someone to show up?
- What percentage of the time you spend on social media involves people you would recognize outside of the platform?
These are not rhetorical. They are the kind of questions the research points toward. The distinction the OSU study makes between known and unknown contacts is a usable lens that most coverage of this topic has not actually translated into anything a person can do. The question of whether online friends count is not a yes or no. It depends almost entirely on who those friends are and whether the connection has any existence outside the platform.
If you are working through a situation where you have realized certain people in your network, online or otherwise, are not what they appeared to be, the patterns that show up in fake friendships and how they tend to operate may be a useful place to look next.
Final Thoughts: The Name Someone Actually Says
I keep returning to that observation from the rooms where I worked. The person who is dying does not think about their network. They think about a name. Sometimes it is someone they have not spoken to in years. Sometimes it is someone they have seen every week. What they are reaching for is not connection in a general sense. They are reaching for a specific person who knows them, who has been present for something real, and whose presence in that final moment would mean something that could not be replicated by anyone else in the network.
Do social media friends count? Some of them do. The ones who would come if you called. The ones you have known outside the platform, through something that was not curated for anyone’s consumption. The ones whose name, if someone asked you to say it, you would say without hesitating. The research is reasonably specific about which ones those are likely to be. And so, I think, is the honest part of what you already know.
The questions around what makes a friendship real sit alongside a lot of what gets collected, at the end of things, about the people who mattered. Some of that is documented in the words people have left about who was worth holding onto.
FAQs
🔬 What did the Oregon State study actually find about social media and loneliness?
The study, published in Public Health Reports in May 2026, found that connecting online with strangers, people never met in person, was associated with increased loneliness. Connecting with people already known in person was neither helpful nor harmful. About 35 percent of participants’ social media contacts were people they had never met.
🧠 Can online friendships become real friendships over time?
They can, if the connection moves off the platform and into something reciprocal and verifiable. The research does not say online friendships are impossible. It says that connections with people never met in person carry a specific risk of idealization and social comparison that known-contact connections do not. Whether an online friendship becomes real depends on whether it eventually develops the same qualities any other friendship requires.
📊 Why does following influencers or strangers increase loneliness?
The mechanism appears to be idealization. Without personal knowledge of someone, there is nothing to push back against the curated version they present. Watching a stranger’s apparently rich social life, without the context that a real relationship would provide, tends to produce social comparison that runs in one direction. Your interior experience is weighed against their exterior performance, and the gap registers as loneliness.
🤝 Does the number of online friends matter, or is it about who they are?
The research points to who they are, not the number. A large network of people you know personally is neutral in terms of loneliness risk. A large network heavily weighted toward strangers is associated with increased loneliness. The count matters far less than the composition.
💬 Should I delete social media to feel less lonely?
The research does not support that conclusion as stated. The finding is specific to stranger-contact, not to social media use in general. A more targeted response would be to audit the composition of your network rather than to leave it entirely. That said, if the platform has become primarily a space for social comparison with people you do not know, then reducing or eliminating it may be a reasonable conclusion to reach on your own terms.
