Biology Reference
In-Depth Information
Player B
Table 12.3
A cooperative
hunting game.; the
pay-off to player A
is shown with
illustrative
numerical values
Hunt (cooperate)
Don't hunt (free ride)
C
4
Share from a hugely
successful hunt
=
S
1
Share of a mildly successful
hunt
=
Hunt
(cooperate)
Player A
F
=
2
Free riding on a mildly
successful hunt
N
=
0
No hunt
Don't hunt
(free ride)
equally, giving each player five units, and so an overall pay-off of 5
4 after the cost
of hunting is subtracted. With these pay-offs, taking part in the hunt (cooperation) is
always the best option irrespective of what the other player does. Consequently, hunt is
the ESS.
Hunt is the ESS, despite the fact that free riding on a hunt (
F
−
1
=
In a cooperative
hunting game,
hunt (cooperate)
is the ESS …
=
2) is better than being the
hunter free ridden on (
S
1). To show why this is so, consider the extreme cases of when
either hunting or not hunting are incredibly rare. If hunting is rare, then hunters will
be with non hunters, and so their average pay-off will be one. Most non hunters will
also be with non hunters and so their average pay-off will be just over zero (just over,
because a small fraction will be with hunters, increasing their pay-off). So, when hunting
is rare, it does better than not hunt and will invade the population. In contrast, if not hunt
is rare, the non hunters will tend to be with hunters and so their average pay-off will be
two, whereas hunters will tend to be with other hunters and so their pay-off will be just
under four (just under, because a small fraction will be with non hunters). So, when not
hunt is rare, it does worse than hunt, and cannot invade the population. This illustrates
the general point that what matters for natural selection is how an individual (or strategy)
does relative to the whole population, not just how it does relative to the partner (or subset
of the population) with which it interacts (Grafen, 2007; see also table 15.1).
=
… even though it
provides a benefit
to others
Cooperative nest founding in ants
A remarkable example of by-product benefit, which comes with a gory ending, is
provided by several ant species where unrelated queens join together to found nests,
cooperating to excavate and build nests (Bernasconi & Strassmann, 1999). This
behaviour occurs in a number of territorial species, where workers from mature colonies
destroy new nests, and workers from newly founded nests steal brood from other newly
founded nests. Cooperative nest founding provides clear benefits, with colonies founded
by more queens growing at a quicker rate, which leads to them being better able to both
raid their nests and defend their own nest from such raids. However, this cooperative
behaviour is unstable, because the advantage of having multiple queens ends when
adult workers emerge. Queens do not forage - instead they seal themselves in the nest
and produce their first brood of workers from their body reserves (fat, protein and
gylycogen obtained by digesting wing muscles). When the workers emerge, this ends
Unrelated queen
ants join together
and cooperatively
form nests …
