The People You Have Never Met Are the Ones Making You Lonely, and the Friends Are Not Helping

Sit with that for a moment, because it is the part that gets left out of the headlines. The obvious defence of social media has always been that the problem is the parasocial junk, and the solution is to use it properly: fewer strangers, more actual friends. This study finds the second half of that prescription does nothing measurable.

Which suggests the mechanism is not about who is in the feed. Something about mediated contact is failing to do the thing that unmediated contact does, and having the right people in it does not fix that.

What it cannot tell you

Which way the arrow points, and this is not a small caveat.

The study is cross sectional. It measured people once, so it cannot separate two entirely different stories. Perhaps accumulating contacts you have never met makes a person lonelier. Or perhaps people who are already lonely go looking for company online among strangers, because strangers are who is available at two in the morning. Both fit the data exactly, and the second is at least as plausible as the first.

The survey data was also collected in July and August 2023 and published in 2026, which in the life of a social platform is a long time. And it is American, aged 30 to 70, so it says nothing about teenagers, who are the group most of the public argument is actually about.

What it is reasonable to conclude

Something narrower and more useful than the coverage will suggest.

If you are lonely, adding people to a feed is not a treatment, and adding the right people is not a treatment either. That is worth knowing, because it is exactly the strategy the platforms recommend and exactly the one a lonely person is most likely to try, since it is the cheapest and requires no arrangement with anyone.

It also fits what we found looking at the physical side of the same question in the third places argument: the places where unplanned, in person contact used to happen have thinned out, and the online substitutes have not replaced them. This study is the other half of that observation, arriving from a completely different method.

The practical reading is dull and probably correct. The thing that reduces loneliness is spending time with people, in a room, repeatedly, with no particular agenda. Everything else is an attempt to get the result without the arrangement, and this is one more piece of evidence that the shortcut does not exist. We made a related point about the limits of self optimisation in the wellness reckoning, and about how small the effective daily changes really are in the study of sleep, walking and diet.