For instance, in regulatory toxicology or when trying to substantiate dose-response dependence for human health risk assessment we seemingly need an animal model of toxic impacts on human populations or subpopulations that are NOT inbred and genetically homogenous, do not breath filtered air and live aseptically and so on. Are we not making our extrapolation from rats to humans even more uncertain and arbitrary than it is intrinsically? (Mind you that I can name a number of toxicological problems making the above standards very useful and even necessary - but is a traditional generalization of such demands justified?)
PS. Honestly, I expected much disagreement with my position, still hoping for receiving some support as well - up to know I've got neither. Is the question really of no interest for my colleagues-toxicologists?
Dear Alex,
I assume that your "No" answers the main question of mine - not the postscriptum. Thank you very mych indeed. However I belive that age and gender homogeneity of animal groups in any comparative experimantel design is obligatory.
Years ago I did experiments on wild-type animals including mallard ducks and mink. Even then one wants to use animals fairly similar, especially age, in order to have a (hopefully) smaller standard deviation etc and thus to (hopefully) develop estimates for other groups of animals or larger groups (ecosystems).
Dear Ross, this way you'll really get "smaller standard deviation etc and thus develop estimates for other groups of animals" - yes, of course, if you want to have these estimates for other groups comprising animals of the same age or whatever. Why and how would it help you to extrapolate your estimates to younger or to older animals, or even to ecosystems is more than I am able to understand.
Honestly, I am rather disappointed to see that my question proved of so little interest for colleagues.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India