240 POTENTIAL BETWEEN BRAINSindividual ants. As Mehdi Mossaid and colleagues put it, “Th e contrast
between the limited information owned by single individuals and the
‘global knowledge’ that would be required to coordinate the group’s
activity is oft en remarkable.”^3
No one is yet very sure. Individuals may appear to be responding on the
basis of simple if- then rules. Information picked up by individuals is re-
sponded to in ste reo typed ways that then become information for others
to respond to. It all suggests little in the way of cognitive resources. Com-
puter modeling— so- called individual- based modeling— has suggested
that much can be achieved from individuals using simple response rules
reciprocally. Indeed, the models have been copied in artifi cial intelli-
gence programs for commercial delivery systems, telephone companies,
and even air traffi c control systems.
A closer examination suggests that the cognitive resources of ants may
be a little more complex than that. Even the following of a pheromone
trail requires integration of positive and negative feedback involving
a number of variables. Th ese variables include the freshness and density
of the volatile pheromones, the quality of the food source, crowding at
feeding sites, direct bodily contact with other nest members, and changes
in the environment itself (including the original perception and then
the depletion of the food source).
Ants also communicate in several other ways: vibrations through
antennae and legs, food or liquid exchange (trophallaxis), mandibular
contact, and direct visual contact. Research has also revealed how the
diff er ent volatility of diff er ent pheromones, together with these other
forms of communication, and used in diff er ent combinations (multi-
modal communication), achieves remarkable sophistication of commu-
nication in all key tasks.
Such a multiplicity of variables suggests dynamical princi ples rather
than cue- response (if- then) rules. In the group, the fl ow of communication
results in streams of positive and negative feedback loops with nonlinear
relations. Th ese self- organize dynamically, as attractor basins, refl ecting
the statistical structure of those interactions (the relational par ameters
being the “rules”). It is the outputs of these attractors that are then re-
fracted through individuals. As Claire Detrain and Jean- Louis Deneu-
bourg, put it, “Fluctuations, colony size and environmental par ameters act
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:54:53 UTChttp://www.ebook3000.com