Skip to main content
Social Sci LibreTexts

4.1: Chapter Overview

  • Page ID
  • \( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)\(\newcommand{\AA}{\unicode[.8,0]{x212B}}\)

    The previous chapter introduced the elements of classical cognitive science, the school of thought that dominated cognitive science when it arose in the 1950s and which still dominates the discipline today. However, as cognitive science has matured, some researchers have questioned the classical approach. The reason for this is that in the 1950s, the only plausible definition of information processing was that provided by a relatively new invention, the electronic digital computer. Since the 1950s, alternative notions of information processing have arisen, and these new notions have formed the basis for alternative approaches to cognition.

    The purpose of the current chapter is to present the core elements of one of these alternatives, connectionist cognitive science. The chapter begins with several sections (4.1 through 4.4) in which are described the core properties of connectionism and of the artificial neural networks that connectionists use to model cognitive phenomena. These elements are presented as a reaction against the foundational assumptions of classical cognitive science. Many of these elements are inspired by issues related to the implementational level of investigation. That is, connectionists aim to develop biologically plausible or neuronally inspired models of information processing.

    The chapter then proceeds with an examination of connectionism at the remaining three levels of investigation. The computational level of analysis is the focus of Sections 4.5 through 4.7. These sections investigate the kinds of tasks that artificial neural networks can accomplish and relate them to those that can be accomplished by the devices that have inspired the classical approach. The general theme of these sections is that artificial neural networks belong to the class of universal machines.

    Sections 4.8 through 4.13 focus on the algorithmic level of investigation of connectionist theories. Modern artificial neural networks employ several layers of processing units that create interesting representations which are used to mediate input-output relationships. At the algorithmic level, one must explore the internal structure of these representations in an attempt to inform cognitive theory. These sections illustrate a number of different techniques for this investigation.

    Architectural issues are the topics of Sections 4.14 through 4.17. In particular, these sections show that researchers must seek the simplest possible networks for solving tasks of interest, and they point out that some interesting cognitive phenomena can be captured by extremely simple networks.

    The chapter ends with an examination of the properties of connectionist cognitive science, contrasting the various topics introduced in the current chapter with those that were explored in Chapter 3 on classical cognitive science.

    This page titled 4.1: Chapter Overview is shared under a CC BY-NC-ND license and was authored, remixed, and/or curated by Michael R. W. Dawson (Athabasca University Press) .

    • Was this article helpful?