Biology Reference
In-Depth Information
The Principle of Computational Equivalence is of foremost import-
ance in NKS; it states that a pattern corresponds to a computation of
equivalent complexity, in all but the most simple systems. In other words,
the computations necessary to predict the fate of any complex system
require at least as many steps as are contained in the system itself. This
implies that, for most systems, concise predictive laws cannot be found,
i.e., they are irreducible, which implies that there is a fundamental limit to
traditional science. In other words, it is impossible to predict by some
mathematical equation, i.e., a concise ''law'', what the pattern some steps
down will be. This can only be predicted by computations based on the
rule. There is no possibility of short-cutting the process of computation.
Not all scientists agree with Wolfram's claim that cellular automata
modelling has led to a scientific revolution (e.g., Molofsky and Bever
2004
for ecology), but some studies have already used cellular automata
successfully for investigating specialized ecological patterns (see Molofsky
and Bever
2004
for examples). Here, I briefly outline Wolfram's (
2002
)
application of NKS to evolution.
The evolutionary process can be interpreted as random searches for
programs carried out in order to maximise fitness. If rules are simple,
iterative random searches may find the best solutions relatively quickly,
but if rules become even slightly more complex, an astronomical number
of steps are necessary to even approach an optimal solution. Hence, most
species (''programs'') are unlikely to be optimally adapted to their niche,
they are trapped in suboptimal niches that were easy to find (Wolfram
2002
). The numbers of mutations in the course of evolution have been
huge, and because relatively simple rules may lead to complexity
(pp. 185-186), it is inevitable that some mutations have led to complex
patterns early in evolutionary history. There is indeed evidence for
this: many complex organisms are very ancient (see, for example, the
Aspidogastrea discussed in Chapter
10
) and the degree of complexity in
many phylogenetic lines has hardly changed, or not at all, over hundreds of
millions of years (see, for instance, living fossils such as Tridacna and
crossopterygians), or organisms may even have become less complex
(see, for example, the reduction in the number of headbones during the
evolution of higher vertebrates from fishes). Also, complexity of similar
features in closely related species, such as pigmentation patterns in cone
shells, may vary enormously between species, suggesting that single or
few mutations are responsible for the differences, which is possible only in
relatively short programs. Furthermore, such patterns closely correspond
to patterns generated by randomly chosen cellular automata with simple
