COPYRIGHT NOTICE:

Camazine, Deneubourg, Franks, et al.: Self-Organization in Biological Systems

is published by Princeton University Press and copyrighted, © 2001, by Princeton
University Press. All rights reserved. No part of this book may be reproduced in any form
by any electronic or mechanical means (including photocopying, recording, or
information storage and retrieval) without permission in writing from the publisher,
except for reading and browsing via the World Wide Web. Users are not permitted to
mount this file on any network servers.

For COURSE PACK and other PERMISSIONS, refer to entry on previous page. For
more information, send e-mail to [email protected]
 1
What Is Self-Organization?

Technological systems become organized by commands

from outside, as when human intentions lead to the building
of structures or machines. But many natural systems
become structured by their own internal processes: these are
the self-organizing systems, and the emergence of order
within them is a complex phenomenon that intrigues
scientists from all disciplines.
—F. E. Yates et al., Self-Organizing Systems:
The Emergence of Order

Self-Organization Defined
Self-organization refers to a broad range of pattern-formation processes in
both physical and biological systems, such as sand grains assembling into
rippled dunes (Figure 1.1), chemical reactants forming swirling spirals (Fig-
ure 1.3a), cells making up highly structured tissues, and fish joining together
in schools. A basic feature of these diverse systems is the means by which they
acquire their order and structure. In self-organizing systems, pattern forma-
tion occurs through interactions internal to the system, without intervention by
external directing influences. Haken (1977, p. 191) illustrated this crucial dis-
tinction with an example based on human activity: “Consider, for example, a
group of workers. We then speak of organization or, more exactly, of organized
behavior if each worker acts in a well-defined way on given external orders,
i.e., by the boss. We would call the same process as being self-organized if
there are no external orders given but the workers work together by some kind
of mutual understanding.” (Because the “boss” does not contribute directly to
the pattern formation, it is considered external to the system that actually builds
the pattern.)
Systems lacking self-organization can have order imposed on them in many
different ways, not only through instructions from a supervisory leader but also
through various directives such as blueprints or recipes, or through pre-existing
patterns in the environment (templates).
To express as clearly as possible what we mean by self-organization in the
context of pattern formation in biological systems, we provide the following
8 – WHAT IS SELF-ORGANIZATION?

definition: Self-organization is a process in which pattern at the global level

of a system emerges solely from numerous interactions among the lower-level
components of the system. Moreover, the rules specifying interactions among
the system’s components are executed using only local information, without
reference to the global pattern. In short, the pattern is an emergent property
of the system, rather than a property imposed on the system by an external or-
dering influence. Emergent properties will be defined in later chapters, but for
now suffice to say that emergent properties are features of a system that arise
unexpectedly from interactions among the system’s components. An emergent
property cannot be understood simply by examining in isolation the properties
of the system’s components, but requires a consideration of the interactions
among the system’s components. It is important to point out that system com-
ponents do not necessarily have to interact directly. As described in Chapter 2,
and Figure 2.4, individuals may interact indirectly if the behavior of one indi-
vidual modifies the environment and thus affects the behavior of other individ-
uals.

Pattern in Group Activities

Critical to understanding our definition of self-organization is the meaning
of the term pattern. As used here, pattern is a particular, organized arrangement
of objects in space or time. Examples of biological pattern include a school of
fish, a raiding column of army ants, the synchronous flashing of fireflies, and
the complex architecture of a termite mound. Examples of other biological
patterns include lichen growth (Figure 1.2a), pigmentation patterns on shells,
fish and mammals (Murray 1988, Meinhardt 1995), (Figures 1.2b,c,d) and the
ocular dominance stripes in the visual cortex of the macaque monkey brain
(Hubel and Wiesel 1977) (Figure 1.2e).
To understand how such patterns are built, it is important to note that in
some cases the building blocks are living units — fish, ants, nerve cells, etc. —
and in others they are inanimate objects such as bits of dirt and fecal cement
that make up the termite mound. In each case, however, a system of living
cells or organisms builds a pattern and succeeds in doing so with no external
directing influence, such as a template in the environment or directions from
a leader. Instead, the system’s components interact to produce the pattern, and
these interactions are based on local, not global, information. In a school of
fish, for instance, each individual bases its behavior on its perception of the
position and velocity of its nearest neighbors, rather than knowledge of the
global behavior of the whole school. Similarly, an army ant within a raiding
column bases its activity on local concentrations of pheromone laid down by
other ants rather than on a global overview of the pattern of the raid.
The literature on nonlinear systems often mentions self-organization, emer-
gent properties, and complexity as well as dissipative structures and chaos
 CHAPTER 1 – 9
Figure 1.2

b(i) b(ii)

Figure 1.2 Self-organized patterns in biological systems include: (a) lichen growth;
(b) pigmentation of a porphyry olive shell (Olivia porphyria) (i) and a marble cone
shell (Conus marmoreus) (ii); (Figure 1.2 continued next page)
10 – W H A T I S S E L F - O R G A N I Z A T I O N ?

Figure 1.2 continued

(c) skin pigmentation on fish (clockwise from top—vermiculated rabbitfish (Siganus

vermiculatus), male boxfish (Ostracion solorensis), and surgeonfish (Acanthurus linea-
tus)); (d) zebra and giraffe coat patterns. (Figure 1.2 continued next page)
 C H A P T E R 1 – 11

Figure 1.2 continued

(e) ocular dominance stripes in the visual cortex

of the macaque monkey (from Hubel and Wiesel
1977). Cortical regions receiving inputs from one
of the monkey’s eyes are shown in black while
regions receiving inputs from the other eye are
represented by white regions between the black
stripes.
e

(Prigogine and Glansdorf 1971; Nicolis and Prigogine 1989). The terms chaos
and dissipative structures have precise scientific meanings that may differ from
popularized definitions, so it is important to discuss these terms at this point.
To begin with, the term complex is a relative one. Individual organisms may
use relatively simple behavioral rules to generate structures and patterns at
the collective level that are relatively more complex than the components and
processes from which they emerge. As discussed in Chapter 6 (see Box 1),
systems are complex not because they involve many behavioral rules and large
numbers of different components but because of the nature of the system’s
global response. Complexity and complex systems, on the other hand, gener-
ally refer to a system of interacting units that displays global properties not
present at the lower level. These systems may show diverse responses that are
often sensitively dependent on both the initial state of the system and nonlinear
interactions among its components. Since these nonlinear interactions involve
amplification or cooperativity, complex behaviors may emerge even though the
system components may be similar and follow simple rules.
Complexity in a system does not require complicated components or numer-
ous complicated rules of interaction.

Self-Organization in Biology
The concept of self-organization in biological systems can be conveyed
through counterexamples. A marching band forming immense letters on a foot-
ball field provides one such example. Here the band’s members are guided in
their behavior by a set of externally imposed instructions for the movements
12 – W H A T I S S E L F - O R G A N I Z A T I O N ?

of each band member that specify in fine detail the final configuration of the
whole band. A particular member of the band may know that the instructions
are to march to the 50-yard line, turn left 90 degrees and march 10 paces. To
the extent that the band member follows this recipe for contributing to the pat-
tern and ignores local information, such as position relative to neighbors, this
pattern formation would not be considered self-organized.
Similarly, a team of carpenters building a house is a pattern-formation pro-
cess that functions without self-organization. Here members of the construc-
tion crew are guided in their collective behavior by predetermined externally
imposed instructions expressed as blueprints, that precisely specify the final
structure of the house. Letter formation by a marching band and house con-
struction by a construction crew both involve pattern building in space.
Let us also consider two counterexamples to self-organization that involve
pattern building over time. One such example is oarsmen in a rowing team
pulling on their oars in perfect synchrony with one another and with appropri-
ate adjustments of their stroke frequency. This pattern arises when each oars-
man responds to the coxswain’s shouted instructions indicating when to begin
each stroke. Clearly, this is an example of a group generating a pattern by fol-
lowing explicit orders from a leader based on the overall state of the group
members. The rhythmic contractions of muscle fibers in the heart are also a
counterexample to self-organization. Here the pattern arises as the component
building blocks (the muscle fibers), follow instructions from special excitable
cells that act as an external pacemaker and send a rhythmic electrical signal to
the fibers.1
We can easily see how a system can form a precise pattern if it receives
instructions from outside—such as a blueprint, recipe, or signals from a
pacemaker—but it is less obvious how a definite pattern can be produced in
the absence of such instructions. A general answer to this puzzle is provided in
the next chapter, while specific answers for particular biological patterns con-
stitute the main body of this book. For now, it is merely asserted that pattern
formation often is achieved by systems without external guidance.
The mechanisms of self-organization in biological systems differ from those
in physical systems in two basic ways. The first is the greater complexity of
the subunits in biological systems. The interacting subunits in physical systems
are inanimate objects such as grains of sand or chemical reactants. In biolog-
ical systems there is greater inherent complexity when the subunits are living
organisms such as fish or ants or neurons.
The second difference concerns the nature of the rules governing interac-
tions among system components. In chemical and physical systems, pattern
is created through interactions based solely on physical laws. For example,
heat applied evenly to the bottom of a tray filled with a thin sheet of viscous
oil transforms the smooth surface of the oil into an array of hexagonal cells
of moving fluid called Bénard convection cells (Figure 1.3) (Velarde and Nor-
 C H A P T E R 1 – 13

a b
Figure 1.3 Further examples of self-organized patterns in physical and chemical sys-
tems: (a) hexagonal Bénard convection cells created when a thin sheet of viscous oil is
heated uniformly from below. Aluminum powder was added to the oil to show the con-
vection pattern; and (b) spiral patterns produced by the Belousov-Zhabotinski reaction.
The chemistry of the reaction is explained by Winfree (1972, 1984). (Image courtesy
of Stefan C. Müller)

mand 1980). The molecules of oil obey physical laws related to surface tension,
viscosity, and other forces governing the motion of molecules in a heated fluid.
Likewise, when wind blows over a uniform expanse of sand a pattern of reg-
ularly spaced ridges is formed (Figure 1.1) through a set of forces attributable
to gravity and wind acting on the sand particles (Anderson 1990; Forrest and
Haff 1992).
Of course, biological systems obey the laws of physics, but in addition to
these laws the physiological and behavioral interactions among the living com-
ponents are influenced by the genetically controlled properties of the com-
ponents. In particular, the subunits in biological systems acquire information
about the local properties of the system and behave according to particular ge-
netic programs that have been subjected to natural selection. This adds an extra
dimension to self-organization in biological systems, because in these systems
selection can finely tune the rules of interaction. By tuning the rules, selection
shapes the patterns that are formed and thus the products of group activity can
be adaptive. What is also intriguing about pattern formation in biological sys-
tems and lends excitement to studies of self-organization in animal groups is
the recent realization that interactions among system components can be sur-
prisingly simple, even when extremely sophisticated patterns are built, such as
the labyrinthine nests of termites, the spatial patterns of army ant raids, and the
coordinated movements of fish in a school.