> The problem is: how do you as a researcher become comfortable testing on a voluntary human test subject?

I don't think you and GP are actually saying anything different. I interpreted their statement 'The researcher should first ask "would I feel comfortable testing this in a voluntary human test subject?"' not to mean don't test on animals if you wouldn't on a human, but to use your reaction and feelings to that to guide your judgement, and help judge what type of animal testing you're at, not to be a binary switch of whether any animal testing at all should be done.