Over the last two decades or so, several significant advances have been mad
e in two distinct fields: neural networks and fuzzy systems. The theory of
fuzzy systems provides a mathematical framework for capturing the uncertain
ties associated with human cognitive processes, such as thinking and reason
ing, and for emulating corresponding perceptual and control processes. The
paradigms of neural networks offer the complementary attributes of learning
and adaptation, together with the innate efficiency of parallel operation.
In this paper we explore fuzzy neural networks, the product of fusion of ne
ural networks and fuzzy mathematics, which have potential for combining the
se mathematical tools into a single capsule. For their favorable properties
, the fuzzy neural networks could be used in the development of systems wit
h some sort of cognitive abilities. These cognitive systems would have the
potential to recapitulate certain aspects of human cognition such as percep
tion, memory, learning, and decision making.