Mechanization of Cognition - Biomimetics: Biologically Inspired Technologies - page 98

— for starting and ending the sentence), the language module composes one or more grammatical
sentences that describe the object and its attributes.
The number of sentences is determined by a meaning content critic subsystem (not shown in
Figure 3.12) which stops sentence generation when all of the distinctive, excited, sentence summary
lexicon symbols have been ''used'' in one or more of the generated sentences.
This sketch illustrates the monkey-see/monkey-do principle of cognition: there is never any
complicated algorithm or software; no deeply principled system of rules or mathematical con-
straints; just confabulation and consensus building. It is a lot like that famous cartoon where
scientists are working at a blackboard, attempting, unsuccessfully, to connect up a set of facts on
the left with a desired conclusion on the right via a complicated scientific argument spanning the
gap between them. In frustration, one of the scientists erases a band in the middle of the argument
and puts in a box (equipped with input and output arrows) labeled ''And Then a Miracle Occurs.''
THAT is the nature of cognition.

3.6 DISCUSSION

This chapter has reviewed a ''unified theory of cognition'' which purports to explain all aspects of
this vast subject with one type of knowledge and one information processing operation. The hope is
that this discussion has convinced you that this approach to cognition is worthy of more extensive
investigation. Only after language, sound, and vision systems such as those described here have
been built, and widely evaluated and criticized, will a sense begin to emerge that the mechanization
of cognition is truly possible. I am hopeful that the arguments and discussion presented here are
sufficiently compelling to make such a research program sensible.

REFERENCES

Ackley D.H., G.E. Hinton and T.J. Sejnowski, A learning algorithm for Boltzmann machines, Cognitive
Science , 9 (1985), pp. 147-169.
Caid W.R. and R. Hecht-Nielsen (Inventors), Representation and retrieval of images using context vectors
derived from image information elements, U.S. Patent No. 6,173,275 (January 9, 2001).
Caid W.R. and R. Hecht-Nielsen (Inventors), Representation and retrieval of images using context vectors
derived from image information elements, U.S. Patent No. 6,760,714 (July 6, 2004).
Chomsky N., Rules and Representations , Columbia University Press, New York, New York (1980).
Daugman J.G., Uncertainty relation for resolution in space, spatial frequency, and orientation optimized
by two-dimensional visual cortical filters, Journal of Optical Society of America A , 2(7) (1985),
pp. 1160-1169.
Daugman J.G., Image analysis and compact coding by oriented 2-D Gabor primitives, in: Eamon B. Barrett and
James J. Pearson (eds), Image Understanding and the Man-Machine Interface, SPIE Proceedings ,
vol. 758, Society for Photo-Optical Instrumentation Engineers, Bellingham, Washington (1987),
pp. 19-30.
Daugman J.G., Complete discrete 2-D Gabor transforms by neural networks for image analysis and compres-
sion, IEEE Transactions of Acoustics, Speech Signal Processing , 36(7) (1988a), pp. 1169-1179.
Daugman J.G., Relaxation neural network for non-orthogonal image transforms, Proceedings of the Inter-
national Conference on Neural Networks , vol. 1, IEEE Press, New York, New York (1988b),
pp. 547-560.
Daugman J.G. and D.M. Kammen, Image statistics, gases, and visual neural primitives, Proceedings of the
International Conference on Neural Networks , vol. 4, San Diego, California (1987), pp. 163-175.
Gregory R., The blind leading the sighted, Nature , 430 (2004), p. 836.
Hecht-Nielsen R., Neurocomputing , Addison-Wesley, Reading, Massachusetts (1990); Corrected Printing,
1991; Japanese Edition, Addison-Wesley, Toppan, Tokyo (1993).