Questions of “selfishness” versus “altruism” are recurrent in evolutionary biological and economic philosophy. Ecologists, evolutionary biologists, economists, game theorists and philosophers are apt to shout at one another about these words.
This is an awful framing for exploring animal behaviour, in that it sets up a binary opposition of “selfishness” (or sometimes “rationality”) and “altruism”, which rather presupposes that this is a useful axis along which to measure behaviour. If it were, indeed unequivocally measurable, or a useful independent variable.
My suspicion is that altruism is something more like a dependent variable, (at least when it is not simply a shorthand to suggest that someone else’s theory is naïve because it has an implication that people may act, in some domain, in a manner in which we like to think that a high frequency trader would not act in forex markets.)