I don't think your point is invalid; but as the siblings point out, sample size effectiveness really depends quite heavily on many factors. As an obvious/naive

I don't think your point is invalid; but as the siblings point out, sample size effectiveness really depends quite heavily on many factors. As an obvious/naive example, you wouldn't need a lot of participants to know that an acute poison is, well, poisonous. As you get ones with less of an acute effect, though, the number of participants needed goes up. Similarly, they did controlled and deliberate exposures here. So, I'm very excited to read follow ons to this study. Looks very promising to my naive eyes.