Re: Three roles of "the population".
From: Guy Hoelzer (hoelzer_at_unr.edu)
Date: 02/26/05
- Next message: Guy Hoelzer: "Re: Mathematics Is Not a Science"
- Previous message: Joe Felsenstein: "Re: MRCA question"
- In reply to: Name And Address Supplied: "Re: Three roles of "the population"."
- Next in thread: name_and_address_supplied_at_hotmail.com: "Re: Three roles of "the population"."
- Messages sorted by: [ date ] [ thread ]
Date: Fri, 25 Feb 2005 23:57:08 -0500 (EST)
in article cvnrl7$1vgh$1@darwin.ediacara.org, Name And Address Supplied at
name_and_address_supplied@hotmail.com wrote on 2/25/05 10:43 AM:
> "Perplexed in Peoria" <jimmenegay@sbcglobal.net> wrote in message
> news:<cvm8f1$1fbv$1@darwin.ediacara.org>...
>> It occurs to me that some of the confusion regarding kin selection and
>> group selection occurs because of a failure to distinguish the three
>> different roles that "the population" plays in our models. There
>> are really three different "kinds" of populations.
>>
>> One is the "breeding population". It is natural to assume that
>> mates are selected randomly from this population. By the same
>> token, in a kin selection argument, it is natural to assume that
>> those genes in a recipient that are not IBD (1-r) have been randomly
>> selected according to the overall gene frequency in the breeding
>> population.
>>
>> A second is the "competitive population". Density dependent selection
>> acts to keep the size of this population constant. Group selection
>> arguments sometimes make the implicit assumption that the competitive
>> population is larger than the breeding population. Thus, the "group",
>> which constitutes a breeding population, is permitted to increase in
>> size at the expense of other (less altruistic?) groups. In the
>> kind of simulation model that I advocate on the thread entitled
>> "NS as head to head competition", the competitors in a survival
>> contest must be randomly selected from the "competitive population".
>>
>> The third is the "social population". In a model of random social
>> interactions, this is the population from which the recipient for
>> a donor is randomly selected. Clearly, as was recognized by Hamilton
>> in 1964, forgotten by him in 1970, but remembered by him by the
>> time of "Narrow Roads", the "social population" must be smaller
>> than the "breeding population" if kin selection is to promote
>> indiscriminate altruism within the "social population".
>
> I'd add a fourth form, and infact this is the only one that I would
> refer to as 'the population'. It is the grouping of conspecifics in
> which we are interested in focussing the problem upon. It is with
> respect to this population that we measure allele / phenotype
> frequencies. This is the population whose evolutionary change we are
> ultimately interested in.
I find your proposed fourth category objectionable. In fact, I think that
the notion that the set (not necessarily a real group) of individuals we
carve out to include in our samples constitutes a population at all to be at
the heart of many inferential errors in population biology. We should
always keep in mind that we can only ASSUME that the individuals in our data
constitute a population, or a random sample thereof, to restrain
overreaching in the drawing of conclusions. I think it is the primary
challenge of empiricism to recognize and sample natural (real) populations
effectively, which is a daunting challenge.
Guy Hoelzer
- Next message: Guy Hoelzer: "Re: Mathematics Is Not a Science"
- Previous message: Joe Felsenstein: "Re: MRCA question"
- In reply to: Name And Address Supplied: "Re: Three roles of "the population"."
- Next in thread: name_and_address_supplied_at_hotmail.com: "Re: Three roles of "the population"."
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
|
