This is a report on an interesting study which, unsurprisingly, confirms previous findings that the use of popular social media (in this case Facebook, Instagram, and Twitter) correlates very closely with negative feelings, reduced well-being, and poor self-esteem: the more you do it, the worse it gets. However, the distinctive thing that these researchers from UBC found is that this applies only to passive scrolling. Direct interaction with others- whatever the medium - has exactly the opposite effect. It ain't what you do, it's the way that you do it.

It occurs to me that this appears to be yet another example of how the structure of general-purpose social networking systems can be inherently harmful, regardless of any further algorithms designed to reinforce the harm. It implies that those who do interact are experiencing all sorts of positive benefits, which are inevitably reflected in their posts, and that are therefore more compelling, and that therefore make those passively scrolling through them feel even more worthless. This is so without needing to take into account the well-established fact that, by and large, what people intentionally reveal of themselves tends to be the stuff they think will show them in a good light. Presumably, less happy people are then less likely to post anything, and will scroll more (there's addiction in this behaviour, to make matters worse), and so further polarization occurs, in a vicious cycle that will rarely end well. When a system that deliberately amplifies such phenomena claims around a third of the world's population as members, it is hard not to be concerned about the large-scale real-world effects of this.

I'd be interested to know whether the same would be true of set-based systems like YouTube or Reddit. I suspect not or, at least, nothing like as much, because the vast majority of visitors are far more interested in the topic than the people making the posts. It would also be interesting to know how it affects users of smaller scale vertical SNSs like, say, Academia.edu or ResearchGate. I'd hypothesize that their (typically) weaker social ties combined with their users' (typically) stronger interests in their topics might make the effect less pronounced, especially because communication is mainly indirect, through papers shared. But it could go either way. Imposter syndrome, for instance, runs strong in academics' blood already, so any feelings of inferiority might get exaggerated.