Re: Logic of kin selection

From: Perplexed in Peoria (jimmenegay_at_sbcglobal.net)
Date: 02/09/05 

Date: Wed, 9 Feb 2005 11:42:28 -0500 (EST)

"William Morse" <wdmorse@twcny.rr.com> wrote in message news:cuc7s3$k0$1@darwin.ediacara.org...
> "Perplexed in Peoria" <jimmenegay@sbcglobal.net> wrote in
> news:cu1ot5$2od7$1@darwin.ediacara.org:
>
> >
> > "Joe Felsenstein" <joe@removethispart.gs.washington.edu> wrote in
> > message news:cu08nv$27co$1@darwin.ediacara.org...
> >> In article <ctv50n$1t3l$1@darwin.ediacara.org>,
> >> Perplexed in Peoria <jimmenegay@sbcglobal.net> wrote:
> >> >On second reading, I am even more dismayed by your answer than
> >> >I was on the first reading. Consider an extreme case in which
> >> >recipientship is strongly correlated with genetics, and donorship
> >> >is only weakly correlated with genetics. In this case, we might
> >> >call the situation "manipulation", rather than "altruism". Offhand,
> >> >I would expect that manipulators would want to manipulate non-kin
> >> >rather than kin. Yes, Hamilton's argument can accomodate this,
> >> >but only by reversing the meanings and signs of b and c. To
> >> >accomodate this, you would have to take b to be the fitness effect
> >> >on the *donor*.
> >>
> >> I suppose the real question is why you are "dismayed". As you
> >> correctly see, Hamilton's argument works well, provided b is the
> >> (now negative) effect on the one who is only likely to have the
> >> allele as a result of relatedness, and c is the (now positive)
> >> effect of the allele on the one whose behavior depends on their
> >> having the allele. We previously called these respectively
> >> "recipient" and "donor" but now they are in effect the opposite.
> >> And, as you see clearly, the condition now is that the allele can
> >> increase if the other individuals are not too closely related.
> >>
> >> So everything is fine but one thing. I can't explain why you are
> >> dismayed. That cannot be explained by Hamilton's formula.
> >
> > I claim that, as an assumption vital to Hamilton's rule, it must be
> > the case that the incidence of the behavior is uncorrelated with the
> > genetics of the recipient. That is, I claim that rb>c is only valid
> > for behaviors that are unilateral - that is, the "donor" is "active"
> > in determining whether the behavior eventuates, whereas the
> > "recipient" is "passive".
> >
> > You appear to deny this, and that dismays me.
> >
> > You seem to think that my counterexample involving "manipulation"
> > supports your side. Actually, it does not. If you look carefully,
> > you will see that the only way we were able to apply Hamilton's
> > rule to the situation was to switch the roles of recipient and
> > donor. After this switch of roles, it was again true that the
> > genetics of the (new) recipient was uncorrelated with his genetics.
> > My claim is supported by this switch, not your claim.
> >
>
> You have both lost me here. I will first stoutly maintain that the
> passiveness of the recipient is irrelevant - the only question is whether
> the benefit to the recipient outweighs the cost to the donor. Now perhaps
> you are arguing that the activity of the recipient negates the extent to
> which the donor's behavior can be said to be genetically determined - see
> below.
>
> Let me try to envision your original "manipulator" argument. You are
> positing that there exists a genetically determined behavior that can
> elicit "altruism" from another individual. The question is will this
> elicited altruism follow Hamilton's rule.
>
> There seem to be two cases. The first is one in which there is no allele
> that will give the donor any choice in the matter. In that case the
> presence or absence of an allele in the genotype of the donor has no
> influence on altruism, so there is nothing for Hamilton's rule to work
> on.
>
> The more interesting case is one in which there is a gene with two
> alleles, one of which causes the donor to respond to a request for
> altruism and one of which turns the response off. Does Hamilton's rule
> shed any light on this situation? I think it is still useful. To the
> extent that the donor expends "fitness points" on responding to requests
> from unrelated individuals, the donor may be better off with the "no go"
> allele.But this is adequately covered by the use of r to reduce the
> inclusive fitness benefit.
>
> Now in the case that, for whatever reason (an inbred population, a
> sessile population, etc.) the interactees are likely to be related, an
> allele for "requesting" may actually benefit the donor, if the requests
> are honest - the donor is not wasting "fitness points" on recipients who
> don't need the aid. In other words, the donor is getting more b for its
> c. This means that it can afford a lower average r, which would encourage
> the spread of the "requesting" allele. The system may be subject to
> dishonest requests, but I think only if it were opened up so that
> interactees were unlikely to be related. Otherwise the dishonest
> requester will only be diverting resources from relatives.
>
> I leave the math to you to work out. It should be simple ;-)

But I think that you have made my case for me - especially the part
about "the donor is getting more b for its c." and "it can afford a
lower average r". In other words, you are arguing that Hamilton's
rule no longer applies to the situation - it would need to be modified.

Please understand. I am not claiming that kin selection ceases to work
as a mechanism if the genetics of the recipient are involved in whether
the behavior eventuates. I am simply saying that Hamilton's simple
rule is no longer valid (in general) in defining the threshold between
adaptive and nonadaptive altruism.

But the rule may still work in some special cases. One such special case
was the "Peshawari pig". There the incidence of altruism depended on
the sex of the recipient - obviously a genetic trait in mammals. However,
I'm pretty sure that the reason this is a special case - the reason the
rule still works - is that the sex of the recipient is not a *heritable*
trait.

The rule handles simple unilateral altruism, and by "flipping" it can
be made to handle unilateral directed manipulation. But I doubt that
it can be made to handle a combination of these two, even if r is taken
to be symmetric. (Maybe it can handle the combination of directed
altruism and undirected manipulation - that case is still unclear to
me).

Relevant Pages

  • Re: Hamiltons rule
    ... >> - If the altruism locus is purely dominant, ... That doesn't matter as long as the donor D line never drops ... >> or of how many loci are involved, it remains the case that the recipient ... > favor the altruism allele. ...
    (sci.bio.evolution)
  • Re: An Accounting of the Soul
    ... Dvora wrote: ... Making the recipient a partner or giving him a job. ... Making a donation anonymously to a recipient who is unkown to the ... donor. ...
    (soc.culture.jewish.moderated)
  • Re: An Accounting of the Soul
    ... I was talking of charity. ... Making the recipient a partner or giving him a job. ... Making a donation anonymously to a recipient who is unkown to the ... donor. ...
    (soc.culture.jewish.moderated)
  • Re: Logic of kin selection
    ... >> effect of the allele on the one whose behavior depends on their ... > genetics of the recipient. ... the benefit to the recipient outweighs the cost to the donor. ... extent that the donor expends "fitness points" on responding to requests ...
    (sci.bio.evolution)
  • Re: Hamiltons rule
    ... >> between donor and recipient measured at the base pair level. ... >> % similarity between the donor and a random member of the population. ... gene in the recipient is a copy of the same parental gene as one randomly ...
    (sci.bio.evolution)
Loading