Biology Reference
In-Depth Information
(i)
Ants follow pheromone trails, which they detect with the tips of their two antennae,
moving towards the side with the strongest pheromone concentration (Fig. 6.21c).
When the ant is near the middle of the trail, there will be only a slight difference in
concentration between the antennae. As it leaves the middle, the concentration
gradient increases and the ant makes a turn back to the centre again. The result is
that it follows the trail centre in a straight line if there is no error in pheromone
detection, or with a sinusoidal trajectory if there is greater error when the difference
in concentration between the antennae is small (Fig. 6.21d).
(ii)
Individuals have a preferred direction: away from the colony when searching for
food, back home when they have found a prey item.
(iii)
Individuals turn to avoid collisions with other ants and outbound ants have a
higher avoidance turning rate than inbound ants. This may arise simply because
inbound ants are less manoeuvrable because they are burdened with prey.
Computer simulations of ants adopting these three rules, with parameters
estimated from individual army ant behaviour, resulted in the emergence of distinct
spatial structuring on the trail, with returning ants in the centre and outbound ants
on the periphery either side, just as observed in nature. Therefore, local rules can
influence traffic organization on a large spatial scale, reducing the number of
collisions between ants moving in different directions and increasing the efficient
flow of ants and food.
Army ants: traffic
lanes as an
outcome of three
simple rules for
individuals
Local rules in fish shoals
Complex three-dimensional movements in fish shoals can also emerge from local
decision making by individuals (Couzin
et al
., 2002). Imagine an individual fish has
three behavioural zones: a zone of repulsion, a zone of orientation and a zone of
attraction, as explained in Fig. 6.22a. Computer simulations show that groups will form
three distinctive collective organizations simply as a result of changes in these zones,
and that the transitions between the three organizations is sharp. If individuals exhibit
attraction to others but little or no orientation, the group forms into a 'swarm', with
individuals moving in many different directions (Fig. 6.22b). As the size of the zone of
orientation increases, the group forms a 'torus', moving around an empty core
(Fig. 6.22c). As the zone of orientation increases further still, the group begins to
move in a single direction (Fig. 6.22d). All these patterns are seen in real fish shoals
(Fig. 6.22e-g). The key point is that in theory large changes in group behaviour can
emerge simply from minor changes in an individual's local response to neighbours.
In further simulations Couzin
et al
. (2002) added a model predator which moved
towards the highest perceived density of prey. The prey now had a further rule: 'if detect
a predator, move away'. This led to the kind of collective behaviour often seen in real
groups, such as waves of escape spreading through the shoal (the Trafalgar effect),
fragmentation of the shoal or sudden 'flash expansions' in all directions.
Finally, in theory simple changes in an individual's local responses could also result in
changes in group size. Hoare
et al
. (2004) found that group size in the banded killifish
Fundulus diaphanus
varied as might be predicted from our previous discussion of costs
and benefits. When broken skin extract from an injured fish was added to the aquarium,
the fish formed larger shoals; when food odour was added, the fish tended to swim
Large scale
changes in
shoaling as an
outcome of
changes in
individual local
responses …
… and changes in
group size
