Hormesis is actually conversial, but is not what I'm referring to. Besides hormesis the other popular models seem to be LNT (linear response, no threshold) and

Hormesis is actually conversial, but is not what I'm referring to. Besides hormesis the other popular models seem to be LNT (linear response, no threshold) and LT (linear response, above a threshold). LNT is used by most right now, mostly because it's the most conservative-possible model that fits the very-high-exposure data that we do have from atomic bomb survivors. It is the model which gives us ALARA and 'Man-REM' planning. LT is more what I would lean towards given available studies of very-low-exposure, but that still leaves open the question of what that would actually mean for public health advice, radiation worksite planning, etc.