tuational paths of cognitive gene expression defined by the
system of equations (19) are, in fact, information source out-
puts that can, in turn, become entrained - farmed - by exter-
nal signals via the mechanisms of network information theory,
as described in section 14.
12 Multiple models
Recently R.G. Wallace and R. Wallace (2009) have argued
that consciousness may have undergone the characteristic
branching and pruning of evolutionary development, particu-
larly in view of the rapidity of currently surviving conscious
mechanisms. They argue that evolution is littered with poly-
phyletic parallelisms: many roads lead to functional Romes,
and propose that consciousness as a particular form of high
order cognitive process operating in real time, embodies one
such example, represented by an equivalence class structure
that factors the broad realm of necessary conditions informa-
tion theoretic realizations of Baars’ global workspace model.
Many different physiological systems, in their view, can sup-
port rapidly shifting, highly tunable, and even simultane-
ous assemblages of interacting unconscious cognitive mod-
ules. Thus, they conclude, the variety of possibilities suggests
minds today may be only a small surviving fraction of ancient
evolutionary radiations - bush phylogenies of consciousness
pruned by selection and chance extinction.
Even in the realms of rapid global broadcast inherent to
real time cognition, Wallace and Wallace (2009), speculate,
following a long tradition, that ancient backbrain structures
instantiate rapid emotional responses, while the newer fore-
brain harbors rapid ‘reasoned’ responses in animal conscious-
ness. The cooperation and competition of these two rapid
phenomena produces, of course, a plethora of systematic be-
haviors.
Since consciousness is necessarily restricted to realms of a
few hundred milliseconds, evolutionary pruning may well have
resulted in only a small surviving fraction of previous evolu-
tionary radiations. Processes operating on longer timescales
may well be spared such draconian evolutionary selection.
That is, the vast spectrum of mathematical models of cog-
nitive gene expression inherent to our analysis here, in the
context of development times much longer than a few hun-
dred milliseconds, implies current organisms may simultane-
ously harbor several, possibly many, quite different cognitive
gene expression mechanisms.
It seems likely, then, that, with some generality, slow phe-
nomena, ranging from institutional distributed cognition to
cognitive gene expression, permit the operation of very many
quite different cognitive processes simultaneously or in rapid
succession.
One inference is, then, that gene expression and its epige-
netic regulation are, for even very simple organisms, far more
complex than individual human consciousness, currently re-
garded as one of the ‘really big’ unsolved scientific problems.
Neural network models adapted from the cognition studies
of a generation ago are unlikely to cleave the Gordian Knot
of scientific inference surrounding gene expression.
13 Epigenetic focus
The Tuning Theorem analysis of the Appendix permits an
inattentional blindness/concentrated focus perspective on the
famous computational ‘no free lunch’ theorem of Wolpert and
Macready (1995, 1997). Following closely the arguments of
English (1996), Wolpert and Macready have established that
there exists no generally superior function optimizer. There
is no ‘free lunch’ in the sense that an optimizer ‘pays’ for
superior performance on some functions with inferior perfor-
mance on others. if the distribution of functions is uniform,
then gains and losses balance precisely, and all optimizers
have identical average performance. The formal demonstra-
tion depends primarily upon a theorem that describes how
information is conserved in optimization. This Conservation
Lemma states that when an optimizer evaluates points, the
posterior joint distribution of values for those points is ex-
actly the prior joint distribution. Put simply, observing the
values of a randomly selected function does not change the
distribution: An optimizer has to ‘pay’ for its superiority on
one subset of functions with inferiority on the complementary
subset.
As English puts it, anyone slightly familiar with the evo-
lutionary computing literature recognizes the paper template
‘Algorithm X was treated with modification Y to obtain the
best known results for problems P1 and P2.’ Anyone who
has tried to find subsequent reports on ‘promising’ algorithms
knows that they are extremely rare. Why should this be?
A claim that an algorithm is the very best for two functions
is a claim that it is the very worst, on average, for all but two
functions. It is due to the diversity of the benchmark set of
test problems that the ‘promise’ is rarely realized. Boosting
performance for one subset of the problems usually detracts
from performance for the complement.
English concludes that hammers contain information about
the distribution of nail-driving problems. Screwdrivers con-
tain information about the distribution of screw-driving prob-
lems. Swiss army knives contain information about a broad
distribution of survival problems. Swiss army knives do many
jobs, but none particularly well. When the many jobs must be
done under primitive conditions, Swiss army knives are ideal.
Thus, according to English, the tool literally carries infor-
mation about the task optimizers are literally tools-an algo-
rithm implemented by a computing device is a physical entity.
Another way of looking at this is to recognize that a com-
puted solution is simply the product of the information pro-
cessing of a problem, and, by a very famous argument, in-
formation can never be gained simply by processing. Thus
a problem X is transmitted as a message by an information
processing channel, Y , a computing device, and recoded as an
answer. By the Tuning Theorem argument of the Appendix
there will be a channel coding of Y which, when properly
tuned, is most efficiently transmitted by the problem. In
general, then, the most efficient coding of the transmission
13