Connectionist

David A. Medler

Biological Computation Project

Department of Psychology, University of Alberta

Alberta, Canada, T6G 2E9

Abstract

Connectionist research is firmly established within the scientific

community, especially within the multi-disciplinary field of

cognitive science. This diversity, however, has created an environment

which makes it difficult for connectionist researchers

to remain aware of recent advances in the field, let alone understand

how the field has developed. This paper attempts to

address this problem by providing a brief guide to connectionist

research. The paper begins by defining the basic tenets of connectionism.

Next, the development of connectionist research is

traced, commencing with connectionism’s philosophical predecessors,

moving to early psychological and neuropsychological

influences, followed by the mathematical and computing contributions

to connectionist research. Current research is then

reviewed, focusing specifically on the different types of network

architectures and learning rules in use. The paper concludes by

suggesting that neural network research—at least in cognitive

science—should move towards models that incorporate the relevant

functional principles inherent in neurobiological systems.

1 The Connectionist Revolution

This solution takes the form of a new associationism, or better,

since it differs deeply and

th” ([42], p. 332).

Unfortunately, this revolution has created an environment in which researchers

may find it difficult to keep up with recent advances in neural

network research. Furthermore, the history of connectionist research is

often overlooked, or at least misconstrued [81]. As a result, a view popular

with current researchers is that connectionism really emerged in the

1980’s—there is only brief mention of research before that time (e.g., [8],

[48]).

Connectionism, however, has a very long past. In fact, one can trace

the origin of connectionist ideas to the early Greek philosopher, Aristotle,

and his ideas on mental associations. These ideas were elaborated by the

British empiricists and then naturally extended by the founders of psychology.

Neuropsychologists then contributed to the growth of connectionism

by trying to relate the processes of learning and memory to underlying

properties of the brain. But, this is only half of the picture. The other

half of the picture is filled in by those researchers engaged in mathematical

research and early computing science who contributed to the formal, computationalunderstandingofboththepowerandlimitationsofconnectionist

networks.

Although it might be argued that these past researchers were not true

“connectionists”intoday’sterms,theideastheyputforthinthedisciplines

of philosophy, psychology, neuropsychology, mathematics, and computing

science are fully embodied within today’s connectionism. And, it is only

throughareviewofthecontributionsmadebyeachofthesedisciplinesthat

we can place connectionism in its proper context today.

2 Connectionism and Cognitive Science

Before we begin with our definition of connectionism, a brief digression is

required. As noted earlier, connectionism is used in many different fields

of science. For example, connectionist networks have been used for aiding

astronomical work [106], assisting medical diagnosis [20], regulating investment

management [121], and controlling robotic limb movement [113].

Many of these systems, however, are approached from an engineering perspective;

...