The algorithm decides what you read and how you engage, even if it’s negative content or something bad for your mental health.
This may be the wrong place to post this, but it's something I've been thinking about for a while. "Algorithm" isn't a dirty word. And in fact, IMHO Mastodon could benefit from a few alternatives to its most-recent-first algorithm.
For instance, I might want to see posts by emergency services in my area first, followed by posts by friends, and posts by a bot that posts a cat picture every minute further down. Or someone might be going off on a rant, and I'd like to turn their firehose of posts down to a trickle for a few hours. Or maybe I'd like Mastodon to just stop showing me anything after a few hours of activity, to encourage me to take a break.
The reason Twitter's, Facebook's, algorithms are evil is that they encourage you to do things you wouldn't want to do, and because they show you content you don't want. Not because they're algorithms.
In a perfect world, every user on every instance would be able to choose how posts are presented. But that may be too computationally expensive, especially for large instances, especially when you start trying to figure out things like the mood of a post. But maybe each instance could decide which algorithm it wants to use, and user can migrate from one instance to another, depending whether they like how things are presented.