Re: Mealey

From: HARNAD Stevan (
Date: Thu May 21 1998 - 20:05:43 BST

> From: Chatwin Judy <>
> I am unclear as to the meanings of 'phenotype' and 'genotype'.
> I have referred to The Selfish Gene and have determined that
> 'phenotype' refers to the effects that a gene has on a body which may
> present itself in more than one manifestation i.e. green eyes and curly
> hair, but I am still unclear with regard to 'genotype'. Can you help?

The genotype is the set of genes you have inherited. The phenotype is
the set of traits that developed from the genes. The two are not the
same because two different genotypes can have the same phenotype:
Brown might be dominant over Blond, so someone who has a Brown from his
father and a Blond from his mother would be a Brown/Blond heterozygous
(nonidentical pair of genes at the same locus) genotype, but simply a
Brown phenotype, indistinguishable (phenotypically) from the likewise
Brown phenotype of a Brown/Brown homozygous genotype.

So two different genotypes can have the same phenotype, but two
different phenotypes cannot have the same genotype (except when
the differences are nongenetic, as when one identical twin is trained
as an athlete and the other is not).

> I also feel I need a 'kid sib' explanation for the Cheater's strategy.
> Once again I have looked through Dawkins' explanation but I feel I have
> not quite grasped it. I understand that cheaters mimic the behaviour of
> others in order to reap the benefits without having to invest anything
> themselves but I do not understand 'His average pay-off is better than
> the average for a sucker' (p184).

This is referring to Prisoner's Dilemma (see my reply to the question
about Mealey Commentator Colman). The idea is that if suckers help
everyone, cheats can do better, accepting help but not giving it in
return. If everyone is a sucker, everyone is equally well off; but if
there are cheaters, they are better off than suckers. In this case,
they would do better in survival and reproduction and increase in
number. Soon there would be only cheaters.

Being a sucker (helping regardless of whether it is returned) is not an
ESS (evolutionarily stable strategy.

> Please could you also clarify what is meant by 'frequency-dependent
> selection'?

This came up in connection with a specific question someone asked
recently about Mealey commentator Colman, who suggested that what
Mealey should have picked was not the ordinary Prisoner's Dilemma but
"Chicken," in which the pay-off is in a slightly different order.
(The cost of being a sucker is less than the cost of being a cheater.)
Then you get frequency-dependent specetion: When there are mostly
suckers, cheating is an advantage, so the proportion of cheaters grows.
But at a certain point, the disadvantages of cheating outweigh the
advantages, so the proportion shrinks. The optimal frequency is
proportion is the one on which the process stabilises. Proportion is
relative frequency. So this point depends on frequency.

Professor of Psychology
Director, phone: +44 1703 592582
Cognitive Sciences Centre fax: +44 1703 594597
Department of Psychology
University of Southampton
Highfield, Southampton

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:21 GMT