Ever since starting to learn about the concept of inclusive fitness during my BSc, I've been bugged by this questions: if inclusive fitness can account for many ‘altruistic’ behaviors, why don't the genetic similarities between ‘unrelated’ individuals of the same and different species outweigh the differences? In other words: when we consider (from a gene-centric perspective) that individuals are temporary vehicles for selfish genes, and if we also consider that the fitness effect of different genes should not be measured in the reproductive success of just single individual, should we not expect to see that genes that promote more broad-scale altrustic behaviour since we share most of our genetic heritage with other individuals?

I understand that my confusion arises from the fact that I fail to sufficiently distinguish between behavioral genes that influence altruism and whose distribution in the population is differentially influenced by different game strategies and genes that have no direct influence on this type of behaviour and are thus not under the same type of selection pressure. However, I cannot seem to manage to produce a clear-cut argument from this distinction.

I'm asking because, when I find myself in the position of having to explain inclusive fitness, I feel a bit awkward when I know in the back of my head that I would not be able to answer this basic question if it comes up.


Relatedness measures a specific type of genetic similarity. In particular, it measures the genetic similarity relative to the rest of the population. When $r>0$, the recipient carries the actor's genes at a higher frequency than the population. When $r=0$ it doesn't mean that the actor and the recipient do not share genes; it only says that the recipient carries the genes of the actor at the same frequency as the population.

  • $\begingroup$ I've upvoted your answer, but haven't yet sufficiently thought it through to be able to say whether or not it clears up my confusion. $\endgroup$ – BigSmoke Nov 30 '15 at 17:41
  • $\begingroup$ Saying that we are talking about relative relatedness yield to the question relative to what?. If I am not mistaken, altruism through kin selection does not need population structure to evolve, relative relatedness can only be define relative to other (isolated) lineages. I think I am confused as the OP is. $\endgroup$ – Remi.b Dec 1 '15 at 15:56
  • $\begingroup$ @Remi.b Assuming that relatedness is modeled as a regression coefficient of the genetic values of the recipient and the actor, doesn't it follow that relatedness has to be relative to the mean genetic value of the population? Hope we are not talking pass each other. $\endgroup$ – falsum Dec 2 '15 at 9:40
  • $\begingroup$ Yes, I think so... Though this implies that 1) population structure is necessary for kin selection to evolve and 2) it brings the question of whether you consider a population that is differentiate enough to have partial reproductive isolation as being included or not in the regression. This is where I get confused. $\endgroup$ – Remi.b Dec 2 '15 at 14:17
  • $\begingroup$ @Remi.b Right, I think I understand your question. As to 2, my impression is that the boundary of the population is determined by all and only the individuals affected by the behavior in question. The population doesn't have to be partially reproductively isolated (e.g., migration can happen). Or would like to say that the populations in, say, the infinite island model are partially reproductively isolated? $\endgroup$ – falsum Dec 2 '15 at 16:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.