On Hamilton's Rule and Inclusive Fitness Theory with Non-additive Payoffs

Hamilton's theory of inclusive fitnesstness is a widely-used framework for study-ing the evolution of social behaviour, but controversy surrounds its status. Hamilton originally derived his famous rb > c rule for the spread of a social gene by assuming additivity of costs and benefits. However it has recently been argued that the additivity assumption can be dispensed with, so long as the -c and b terms are suitably dened, as partial regression coecients. I argue that this way of generalizing Hamilton's rule to the non-additive case, while formally correct, faces conceptual problems.