Biology Reference
In-Depth Information
BOX 14.1
SEQUENTIAL ASSESSMENT
Magnus Enquist and Olof Leimar (1983, 1987, 1990) suggest that information
accumulates during a contest in a way similar to statistical sampling. The
outcome of a single bout contains a random error and to get a more accurate
estimate contestants must 'increase the sample size' by repeating the behaviour.
On this view we would expect the least costly displays to be used first of all,
followed by the more costly but more accurate means of assessment. An
individual will give up when it assesses its fighting ability to be less than that of
its opponent. With more closely matched opponents, fights should last longer
and escalate further simply because it will take longer to assess which is the
strongest, just as a statistician needs to sample more to detect a small difference
than a large one.
This idea makes good sense of contests involving sequences of displays, such
as those of the South American cichlid fish
Nannacara anomala
. In nature,
males compete for females who are about to spawn and contests can be readily
staged in laboratory tanks. The order of displays in a contest is very consistent,
with constant rates of behaviour within each phase (Fig. B14.1.1). (a) Initially
(a)
(b)
(c)
(d)
(e)
(f)
Fig. B14.1.1
Fighting sequence of male cichlid fish,
Nannacara anomala
. (a) Lateral
orientation. (b) Tail beating. (c) Frontal orientation. (d) Biting. (e) Mouth wrestling.
(f) The loser (right) gives up. Jakobsson
et al
. (1979). Drawing by Bibbi Mayrhofer.
