Game theory in evolution

"Jarek Hirny" <[email hidden]> wrote in message

Quoted message said:

I'd be very grateful if you could point me to game theory
applications in evolution.

The problem is that games are theoretical constructs that
don't map obviously to real organisms. One factor is that
games suppose a "currency", but we don't generally know how
valuable a grain of feed is to a pigeon, for example. (If
it's a lab pigeon then it is guaranteed not to be starved to
death anyway, so the survival value of food is zero. However
pigeons probably don't know this.)

Guppies will inspect predators, to assess whether they are
hungry and thus a threat. If two guppies are in a group,
then this creates a game similar to "prisoner's dilemma".
The theoretical best strategy is "[censored] for tat", though
there's some argument about this. However attempts to match
the actual behaviour of guppies up to "[censored] for tat" haven't
proved too successful.

Jarek Hirny said:

Hi,

I'd be very grateful if you could point me to game theory
applications in evolution. I'm aware of a hawk-and-pigeon
model, described by Dawkins in "Selfish Gene", and of its
more slightly more sophisticated versions, but, because
of my interest in GT, I'd like to find something more
advanced/less known/et cetera. Thanks. Should I turn to
classics, like Hamilton, and hiss evolutionary-stable-
strategy idea? How difficult his writings are to
understand for somebody with interest in GT, but with
just a little knowledge about biology and evolution?

(I'm sorry if the exact names of models or books vary
from original ones; I'm not a native english speaker and
I may be mistranslating them. Hope you get the point,
tho. 🙂 )

The classic is Maynard Smith's "Evolution and the Theory of
Games", from 1982. IIRC, it doesn't need too deep a
knowledge of biology. Of course, it's now about 20 years out
of date, so you should be able to find a more recent
textbook. However, the more recent stuff builds on this
work, so it's still useful.

Bob

--
Bob O'Hara

Dept. of Mathematics and Statistics
P.O. Box 4 (Yliopistonkatu 5) FIN-00014 University of
Helsinki Finland Telephone: +358-9-191 23743 Mobile:
+358 50 599 0540 Fax: +358-9-191 22 779 WWW:
rni.helsinki.fi~bohOpen ↗ Journal of Negative
Results - EEB: jnr-eeb.orgjnr-eeb.orgOpen ↗

Quoted message said:

The problem is that games are theoretical constructs that
don't map obviously to real organisms. One factor is that
games suppose a "currency", but we don't generally know
how valuable a grain of feed is to a pigeon, for example.

Well, it's not important how valuable it is, by which I mean
that you don't have to measure it in any units. You rather
need to know only whether action A is more valuable to
pigeon than action B to hawk. Which is not simple at all to
estimate, but, I think, not impossible to do too. It's just
the same thing as with "Utility" idea in economic sciences.

Thanks for answering.

--
_-(_)- _-(_)- _-(_)- _-("😉- _-(_)- _-(_)- _-(_)-
`(___) `(___) `(___) `%%%%% `(___) `(___) `(___) jgs
// \\ // \\ // \\ // \\ // \\ // \\ // \\ Who's the
black sheep in the family?

Jarek Hirny said:

Quoted message said:

The problem is that games are theoretical constructs that
don't map obviously to real organisms. One factor is that
games suppose a "currency", but we don't generally know
how valuable a grain of feed is to a pigeon, for example.

Well, it's not important how valuable it is, by which I
mean that you don't have to measure it in any units. You
rather need to know only whether action A is more valuable
to pigeon than action B to hawk. Which is not simple at
all to estimate, but, I think, not impossible to do too.
It's just the same thing as with "Utility" idea in
economic sciences.

Actually, there is a clear utility in the case of
evolution: fitness. This is one of the reasons why game
theory works so well.

Of course, estimating the fitness is another matter, but yes
there are ways of doing that too - if you can fin a copy of
Manly's book "The Statistics of Natural Selection", then
you'll see a host of methods (and there have been more
developments in the last 20 years, too).

Bob

--
Bob O'Hara Department of Mathematics and Statistics
P.O. Box 4 (Yliopistonkatu 5) FIN-00014 University of
Helsinki Finland Telephone: +358-9-191 23743 Mobile:
+358 50 599 0540 Fax: +358-9-191 22 779 WWW:
rni.helsinki.fi~bohOpen ↗ Journal of Negative
Results - EEB: www.jnr-eeb.org

In article <[email hidden]>,

Anon. said:

The classic is Maynard Smith's "Evolution and the Theory of
Games", from 1982. IIRC, it doesn't need too deep a
knowledge of biology. Of course, it's now about 20 years
out of date, so you should be able to find a more recent
textbook. However, the more recent stuff builds on this
work, so it's still useful.

Maynard Smith's book is still in print, but he also has a
new book: "Animal Signals" by him and David Harper, (Oxford
University Press 2003,
24.95 UK pounds paperback) which he says is in effect a new
edition of the game theory book.

--
Joe Felsenstein [email hidden]
Department of Genome Sciences and Department of
Biology, University of Washington, Box 357730, Seattle,
WA 98195-7730 USA

"Anon." <[email hidden]> wrote in

Quoted message said:

Of course, estimating the fitness is another matter, but
yes there are ways of doing that too - if you can fin a
copy of Manly's book "The >

Statistics of Natural Selection", then you'll see a host
of methods

Quoted message said:

(and there have been more developments in the last 20
years, too).

The problem is your estimates have got to be quite good. For
instance if a chick is going to leave a worm to a weaker
nestmate, rather than eat it itself, then the worm should be
exactly twice as valuable to the nestmate than to itself,
assuming the same paternity, and adjusted if there are cases
of the clutch being fathered by more than one male. The
complication is that, to the mother, both chicks are of
equal genetic value, but the stronger one is probably a
better bet, because survival is usually quite low. Mother
can impose costs on a chick grabbing more than its fair
share of food, but we've got see whether she actually does
so, and then see if we can measure these as well. All in all
its very difficult to disentagle what is going on with the
chicks' feeding strategies.

Quoted message said:

Actually, there is a clear utility in the case of
evolution: fitness. This is one of the reasons why game
theory works so well.

I have just finished reading "Extended Phenotype" by
Dawkins, so I'd like to ask, which meaning of fitness you
are talking about -- since he presented five different
intepretations of this word. I guess it's not the strict
genetic meaning (How much, relatively, of different alleles
are passed to next generations), since fitness in this
definition would be quite easy to measure (I think).

--
_-(_)- _-(_)- _-(_)- _-("😉- _-(_)- _-(_)- _-(_)-
`(___) `(___) `(___) `%%%%% `(___) `(___) `(___) jgs
// \\ // \\ // \\ // \\ // \\ // \\ // \\ Who's the
black sheep in the family?