> It sounds like they had to torture several mice for days to get this result.

The paper's very explicit on how this is done - they injected the mice either with RTX (https://en.wikipedia.org/wiki/Resiniferatoxin, a functional capcasin analog) or saline (at random) to induce a stress response, and countered it in some mice with buprenorphine to rule out false positives.

So, yeah, some of them got to deal with nothing, some dealt with a burning sensation, some dealt with burning followed by bliss, and some felt nothing but bliss.

But, at the end of the day they are lab mice, and there are plenty that are being subjected to worse treatment for less laudable scientific ends than understanding the physiological responses to stress and pain.