280 CATALYZING INQUIRY
- Division of labor. In order to gather food, maintain the nest, defend against predators, and so on,
a colony has to allocate many different tasks among many different ants simultaneously—again, with-
out the benefit of central planning or individual intelligence. In many cases this is done by a physical
caste system, so that workers do certain jobs, soldiers do others, and so on. Yet ants will often allocate
tasks even within a single caste. A simple mechanism that reproduces this behavior is to give each
individual a response threshold for each task: once the stimuli associated with that task pass the
threshold—imagine the smell of accumulating garbage—the individual gets to work. The result is that
individuals with higher and higher thresholds keep pitching in until the stimuli are under control,
leaving everyone else free to engage in tasks for which they have low thresholds. - Cooperative transport. If a single ant encounters a food item that’s too big for her to carry alone
(e.g., a dead cockroach), she will recruit nest mates via pheromones to help. Now, however, without a
leader or brains, they somehow have to start pulling in the same direction. A simple, two-part rule that
reproduces the observed behavior is (1) if the object is already moving in the direction you’re pulling,
keep pulling, and (2) it’s not moving at all, or is moving in a different direction, reorient yourself at
random and start pulling that way. The result is a sequence in which the ants start out pulling their
burden from every direction at once, to no effect—until suddenly, when enough ants just happen to line
up by accident, a kind of phase transition sets in and the load begins to move. - Cooperative construction. Many species of social insects can build structures of astonishing com-
plexity: witness the vast, hexagonal combs of the honeybee or the multilayered, intricately swirling
nests of the paper wasp. And yet again, they manage to do so without the benefit of central planning or
individual intelligence. One way to account for such behavior in simulated insects is to equip each
individual with a collection of local rules: in situation 1, take action A; in situation 2, take action B; and
so on. For a wasp carrying a load of wood pulp, say, such a rule might be, “If you’re surrounded by
three walls, then deposit the pulp.” In general, each insect will modify the environment encountered by
the others, and the structure will organize itself in much the same way that the proteins comprising a
virus particle assemble themselves inside an infected cell.
Ant algorithms are conceptually similar to the particle swarm optimization algorithm described in
Section 8.2.1. However, at least in the case of the Ant Colony Optimization algorithm, it is known that
ants really use the algorithm described. For this reason, this algorithm was placed in the category of
biologically inspired mechanisms (rather than principles).8.4 BIOLOGY AS PHYSICAL SUBSTRATE FOR COMPUTING
8.4.1 Biomolecular Computing
The idea of constructing computer components from single molecules or atoms is the logical, if
distant, end point of the seemingly inexorable miniaturization of chips and has been foreseen at least
since Richard Feynman’s lecture “There’s Plenty of Room at the Bottom” in 1959.^103 Molecular comput-
ing would have significant advantages, most obviously minuscule size of the resulting component, but
also a potentially low marginal cost per component and extreme energy efficiency. However, the tech-
nology for the precision placing of single atoms or molecules on a large scale is still in its infancy.
However, there is a significant shortcut available: to use biological molecules, including DNA,
RNA, and various enzymes, as instruments to perform computational tasks. The sophisticated func-
tions of DNA and related molecules, coupled with the existing technological infrastructure for synthe-
sizing, manipulating, and analyzing them found in molecular biology laboratories, make it feasible to
employ them as a universal set of computing components. Also, because the code of DNA is essentially(^103) R.P. Feynman, “There’s Plenty of Room at the Bottom,” American Physical Society, December 29, 1959; available at http://
http://www.zyvex.com/nanotech/feynman.html.