[halpert]

But what is the likelihood that this result applies to humans vs. doesn't apply to humans? It doesn't sound like you can say. There are also differences even between two different humans. Simply being different is not enough to discount a result using a model that has otherwise proven successful.

[dundarious]

I'm absolutely confident I can discount it in the sense of "reduce its predictive power" compared to a study in humans, but I already said I'm not discounting it in the sense of dismissing it entirely.

> It doesn't mean 0 predictive power.

BTW, I also think that most of the time you shouldn't change your daily behavior based on one single human study.