SPREADING ACTIVATION AND CONNECTIONIST MODELS FOR NATURAL LANGUAGE PROCESSING

SPREADING ACTIVATION AND CONNECTIONIST MODELS FOR NATURAL LANGUAGE PROCESSING Introduction. Spreading activation networks and connectionist systems. Feldman & Ballard (1982) introduced the term "connectionist models" to refer to a collection of theories and models in cognitive science (i.e. artificial intelligence, cognitive psychology, cognitive linguistics etc.). These theories share some important features: they are all designed to explain complex cognitive phenomena through the activity of a large number of simple processing elements. The assumption of a central processor operating on a passive memory (the model of the von Neumann machine) or the introduction of a "cognitive executive" in models for cognitive behavior is avoided; all processing is supposed to be done by the interaction of simple processors in parallel. Connectionist models as well as spreading activation theories question fundamental, theoretical assumptions about the nature of the human language processing system. Connectionist models are not based on the traditional "physical symbol system" hypothesis, and language processing is no longer viewed as exclusively rule-driven behavior (see Elman 1988, for example). Instead, language analysis and production is assumed to be the result of the simultaneous consideration of multiple constraints encoded in a network which performs analog computation (e.g. Dell 1985, Diederich 1985, 1988a, b, c, d, and Schnelle 1981). Spreading activation http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Theoretical Linguistics de Gruyter

SPREADING ACTIVATION AND CONNECTIONIST MODELS FOR NATURAL LANGUAGE PROCESSING

Loading next page...
 
/lp/de-gruyter/spreading-activation-and-connectionist-models-for-natural-language-fxEWpx7VTG
Publisher
de Gruyter
Copyright
Copyright © 2009 Walter de Gruyter
ISSN
0301-4428
eISSN
1613-4060
D.O.I.
10.1515/thli.1990.16.1.25
Publisher site
See Article on Publisher Site

Abstract

Introduction. Spreading activation networks and connectionist systems. Feldman & Ballard (1982) introduced the term "connectionist models" to refer to a collection of theories and models in cognitive science (i.e. artificial intelligence, cognitive psychology, cognitive linguistics etc.). These theories share some important features: they are all designed to explain complex cognitive phenomena through the activity of a large number of simple processing elements. The assumption of a central processor operating on a passive memory (the model of the von Neumann machine) or the introduction of a "cognitive executive" in models for cognitive behavior is avoided; all processing is supposed to be done by the interaction of simple processors in parallel. Connectionist models as well as spreading activation theories question fundamental, theoretical assumptions about the nature of the human language processing system. Connectionist models are not based on the traditional "physical symbol system" hypothesis, and language processing is no longer viewed as exclusively rule-driven behavior (see Elman 1988, for example). Instead, language analysis and production is assumed to be the result of the simultaneous consideration of multiple constraints encoded in a network which performs analog computation (e.g. Dell 1985, Diederich 1985, 1988a, b, c, d, and Schnelle 1981). Spreading activation

Journal

Theoretical Linguisticsde Gruyter

Published: Jan 1, 1990

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off