LINGUIST List 15.1605

Thu May 20 2004

Diss: Neuroling: Koutsomitopoulou: 'A neural...'

Editor for this issue: Tomoko Okuno <>


  1. e_koutso, New Dissertation Abstract added

Message 1: New Dissertation Abstract added

Date: Thu, 20 May 2004 06:34:59 -0400 (EDT)
From: e_koutso <>
Subject: New Dissertation Abstract added

Institution: Georgetown University
Program: Department of Linguistics
Dissertation Status: Completed
Degree Date: 2004

Author: Eleni Koutsomitopoulou

Dissertation Title: 
A neural network model for the representation of Natural Language

Linguistic Field:
Applied Linguistics
Computational Linguistics
Linguistic Theories
Text/Corpus Linguistics
Cognitive Science

Dissertation Director 1: Donald Loritz
Dissertation Director 2: George V. Wilson
Dissertation Director 3: Solomon S.J. Sara
Dissertation Director 4: Allan G. Alderman

Dissertation Abstract:

Current research in natural language processing demonstrates the
importance of analyzing syntactic relationships, such as word order,
topicalization, passivization, dative movement, particle movement,
pronominalization as dynamic resonant patterns of neuronal activation
(Loritz, 1999). Following this line of research this study
demonstrates the importance of also analyzing conceptual
relationships, such as polysemy, homonymy, ambiguity, metaphor,
neologism, coreference, as dynamic resonant patterns represented in
terms of neuronal activation. This view has implications for the
representation of natural language. Alternatively, formal
representation methods abstract away from the actual properties of
real-time natural language input and rule-based systems are of limited
representational power.

Since NL is a human neurocognitive phenomenon we presume that it can
be best represented in a neural network model. This study focuses on a
neural network simulation, the Cognitive Linguistic Adaptive Resonant
Network (CLAR-NET) model of online and real-time associations among
concepts. The CLAR-NET model is a simulated Adaptive Resonance Theory
(ART, Grossberg 1972 et seq.) model. Through a series of experiments,
I address particular linguistic problems such as homonymy, neologism,
polysemy, metaphor, constructional polysemy, contextual coreference,
subject-object control, event-structure metaphor and negation. The
aim of this study is to infer natural language specific mappings of
concepts in the human neurocognitive system on the basis of known
facts and observations provided within the realms of conceptual
metaphor theory (CMT), and adaptive grammar (AG, Loritz 1999),
theories of linguistic analysis, and known variables drawn from the
brain and cognitive sciences as well as previous neural network
systems built for similar purposes. Additionally, this study
investigates the extent to which these linguistic phenomena can be
plausibly analyzed and accounted for within an ART-like neural network

My basic hypothesis is that the association among concepts is
primarily an expression of domain-general cognitive mechanisms that
depend on continuous learning of both previously presented linguistic
input and everyday, direct experiential (i.e. sensory-physical)
behaviors represented in natural language as "common knowledge" (or
"common sense"). According to this hypothesis, complex conceptual
representations are not actually associated with pre-postulated
feature structures, but with time-sensitive dynamic patterns of
activation. These patterns can reinforce previous learning and/or
create new "place-holders" in the conceptual system for future value

This line of investigation holds implications for language learning,
neurol inguistics, metaphor theory, information retrieval, knowledge
engineering, case-based reasoning, knowledge-based machine translation
systems and related ontologies.

This study finds that although STM effects in ART-like networks are
significant, most of the time LTM calculation yields better semantic
discrimination. It is suggested that the internal structure of lexical
frames that correspond to clusters of congenial associations (in fact,
neuronal subnetworks), is maintained as long as it resonates with new
input patterns or learned in long-term memory traces. Different
degrees of similarity (or deviation) from previously acquired
knowledge clusters are computed as activation levels of the
corresponding neuronal nodes and may be measured via differential
equations of neuronal activity.

The overall conclusion is that ART-like networks can model interesting
linguistic phenomena in a neurocognitively plausible way.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue