Problem: if you read that Twitter thread, about the meta-analysis, he talks about A/B testing, that between stimulus A and stimulus B, people chose the one that has a trigger warning, so they don’t work. But nearly everyone who has commented has shown real world effect - they don’t choose between two written pieces of material about something traumatizing, one with or without a trigger warning. They choose C - to use the trigger warning and not read an option at all. I would have to know a lot more about the study design of the studies they used for the meta-analysis. And I can think of a very clear reason why, if only given an A/B choice of reading something traumatizing, you’d read the one with the trigger warning - because you might feel the person giving the trigger warning would treat the material more sensitively.
Maybe I’m way off base and the study design of the studies they used were fine, but it sounds like there’s a shitload of margin for error - how in the world could they possibly have tailored double-blind studies about trigger warnings, without knowing each individual’s fears? So they relied on something less than the gold standard, and then mistakes happen. I mean, they may even have relied on self-reporting, which is basically saying, “Random university volunteers who needed good will from their tutor just told us whatever they were feeling on a random Tuesday morning, and that’s what this study is based on.”