Hacker News Reader

Comment by pfdietz | Hacker News Reader

The hilarious thing is that hormesis, taken seriously, would make one think the risk curve has a negative second derivative, as this study suggests it does. So a hormesis fan should have been suggesting LNT was underestimating the danger of low dose radiation.

View on HN

No it wouldn't. The hormesis hypothesis proposes that the body's repair mechanism from a low dose of radiation yields benefits that outweigh the damage of the radiation. Eventually those internal repair mechanisms are overwhelmed, leading to a positive second derivative.

There is no specific evidence for hormesis (i.e. evidence lacking good alternative explanations) nor are there reasonable proposed mechanisms by which this would happen. But we should be clear about what it says.

No, what the hormesis idea suggests is that radiation induces an increase in repair mechanisms. So, add radiation, you boost defense mechanisms, and more radiation then has less effect. The curve is bent downward, the second derivative is negative.

All of the graphs in the literature suggest a positive second derivative. Why do you say negative? AIUI the model is that hormesis kicks in even for extremely small amounts of radiation, but has a limited effect and is overwhelmed by the negative effects of larger amounts.

I say negative because that's what this study is suggesting -- the slope of the curve is higher at lower dose than at higher doses.

That doesn't mean negative. The curve could be flat, then steeply rises to match LNT at some point.

It means that somewhere the second derivative must be negative (proof: take the derivative, then apply the mean value theorem of calculus)

You're correct of course. I read that as first derivative for some reason.

Reply on news.ycombinator.com