Rethinking Eliminative Connectionism

Humans routinely generalize universal relationships to unfamiliar instances. If we are told “if glork then frum,” and “glork,” we can infer “frum”; any name that serves as the subject of a sentence can appear as the object of a sentence. These universals are pervasive in language and reasoning. One...

Full description

Saved in:
Bibliographic Details
Published in:Cognitive psychology 1998-12, Vol.37 (3), p.243-282
Main Author: Marcus, Gary F.
Format: Article
Language:English
Subjects:
Biological and medical sciences
Cognition & reasoning
Cognition. Intelligence
Cognitive Processes
Cognitive Psychology
Concept Formation
Connectionism
Fundamental and applied biological sciences. Psychology
Generality
Humans
Language
Linguistics
Psychology
Psychology. Psychoanalysis. Psychiatry
Psychology. Psychophysiology
Symbolism
Theories
Training
Verbal Behavior
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Humans routinely generalize universal relationships to unfamiliar instances. If we are told “if glork then frum,” and “glork,” we can infer “frum”; any name that serves as the subject of a sentence can appear as the object of a sentence. These universals are pervasive in language and reasoning. One account of how they are generalized holds that humans possess mechanisms that manipulate symbols and variables; an alternative account holds that symbol-manipulation can be eliminated from scientific theories in favor of descriptions couched in terms of networks of interconnected nodes. Can these “eliminative” connectionist models offer a genuine alternative? This article shows that eliminative connectionist models cannot account for how we extend universals to arbitrary items. The argument runs as follows. First, if these models, as currently conceived, were to extend universals to arbitrary instances, they would have to generalize outside the space of training examples. Next, it is shown that the class of eliminative connectionist models that is currently popular cannot learn to extend universals outside the training space. This limitation might be avoided through the use of an architecture that implements symbol manipulation.
ISSN:0010-0285
1095-5623
DOI:10.1006/cogp.1998.0694