It's reasonably easy to see any herbivore receiving a fitness advantage for evolving this trait, since they are more likely to use it on close kin with whom they share a fraction of their genes than on any unrelated conspecific.<p>It's interesting, however, to see that the small brain size of mice is still enough to support it. I suspect it has something to do with the hypothalamus/whole brain size ratio.<p>Since rodents have been established to be capable of experiencing entrained trauma responses (stress hormones, hypervigilance, disturbed sleep habits, reduced sociality and panic behavior to audio triggers) and the OP establishes non-parental caregiving behavior, the combination of the two suggest it's likely that they can feel some version of guilt, shame or grief in the instances where their resuscitation reflex is unsuccessful.
There wasn't much, but small discussion (39 points, 5 days ago, 3 comments) <a href="https://news.ycombinator.com/item?id=43261947">https://news.ycombinator.com/item?id=43261947</a><p>Papers:<p><i>A neural basis for prosocial behavior toward unresponsive individuals</i> <a href="https://www.science.org/doi/10.1126/science.adq2679" rel="nofollow">https://www.science.org/doi/10.1126/science.adq2679</a><p><i>Reviving-like prosocial behavior in response to unconscious or dead conspecifics in rodents</i> <a href="https://www.science.org/doi/10.1126/science.adq2677" rel="nofollow">https://www.science.org/doi/10.1126/science.adq2677</a><p><i>An innate drive to save a life</i> <a href="https://www.science.org/doi/10.1126/science.adv3731" rel="nofollow">https://www.science.org/doi/10.1126/science.adv3731</a>