Early lexical development in a self-organizing neural network

Ping Li, Igor Farkas, Brian MacWhinney

Research output: Contribution to journalArticle

130 Citations (Scopus)

Abstract

In this paper we present a self-organizing neural network model of early lexical development called DevLex. The network consists of two self-organizing maps (a growing semantic map and a growing phonological map) that are connected via associative links trained by Hebbian learning. The model captures a number of important phenomena that occur in early lexical acquisition by children, as it allows for the representation of a dynamically changing linguistic environment in language learning. In our simulations, DevLex develops topographically organized representations for linguistic categories over time, models lexical confusion as a function of word density and semantic similarity, and shows age-of-acquisition effects in the course of learning a growing lexicon. These results match up with patterns from empirical research on lexical development, and have significant implications for models of language acquisition based on self-organizing neural networks.

Original languageEnglish (US)
Pages (from-to)1345-1362
Number of pages18
JournalNeural Networks
Volume17
Issue number8-9
DOIs
StatePublished - Oct 1 2004

Fingerprint

Learning
Linguistics
Neural networks
Semantics
Language
Confusion
Empirical Research
Neural Networks (Computer)
Self organizing maps

All Science Journal Classification (ASJC) codes

  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Li, Ping ; Farkas, Igor ; MacWhinney, Brian. / Early lexical development in a self-organizing neural network. In: Neural Networks. 2004 ; Vol. 17, No. 8-9. pp. 1345-1362.
@article{e6c47e2138f748d489a2fb55229f0e49,
title = "Early lexical development in a self-organizing neural network",
abstract = "In this paper we present a self-organizing neural network model of early lexical development called DevLex. The network consists of two self-organizing maps (a growing semantic map and a growing phonological map) that are connected via associative links trained by Hebbian learning. The model captures a number of important phenomena that occur in early lexical acquisition by children, as it allows for the representation of a dynamically changing linguistic environment in language learning. In our simulations, DevLex develops topographically organized representations for linguistic categories over time, models lexical confusion as a function of word density and semantic similarity, and shows age-of-acquisition effects in the course of learning a growing lexicon. These results match up with patterns from empirical research on lexical development, and have significant implications for models of language acquisition based on self-organizing neural networks.",
author = "Ping Li and Igor Farkas and Brian MacWhinney",
year = "2004",
month = "10",
day = "1",
doi = "10.1016/j.neunet.2004.07.004",
language = "English (US)",
volume = "17",
pages = "1345--1362",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier Limited",
number = "8-9",

}

Early lexical development in a self-organizing neural network. / Li, Ping; Farkas, Igor; MacWhinney, Brian.

In: Neural Networks, Vol. 17, No. 8-9, 01.10.2004, p. 1345-1362.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Early lexical development in a self-organizing neural network

AU - Li, Ping

AU - Farkas, Igor

AU - MacWhinney, Brian

PY - 2004/10/1

Y1 - 2004/10/1

N2 - In this paper we present a self-organizing neural network model of early lexical development called DevLex. The network consists of two self-organizing maps (a growing semantic map and a growing phonological map) that are connected via associative links trained by Hebbian learning. The model captures a number of important phenomena that occur in early lexical acquisition by children, as it allows for the representation of a dynamically changing linguistic environment in language learning. In our simulations, DevLex develops topographically organized representations for linguistic categories over time, models lexical confusion as a function of word density and semantic similarity, and shows age-of-acquisition effects in the course of learning a growing lexicon. These results match up with patterns from empirical research on lexical development, and have significant implications for models of language acquisition based on self-organizing neural networks.

AB - In this paper we present a self-organizing neural network model of early lexical development called DevLex. The network consists of two self-organizing maps (a growing semantic map and a growing phonological map) that are connected via associative links trained by Hebbian learning. The model captures a number of important phenomena that occur in early lexical acquisition by children, as it allows for the representation of a dynamically changing linguistic environment in language learning. In our simulations, DevLex develops topographically organized representations for linguistic categories over time, models lexical confusion as a function of word density and semantic similarity, and shows age-of-acquisition effects in the course of learning a growing lexicon. These results match up with patterns from empirical research on lexical development, and have significant implications for models of language acquisition based on self-organizing neural networks.

UR - http://www.scopus.com/inward/record.url?scp=9144252205&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=9144252205&partnerID=8YFLogxK

U2 - 10.1016/j.neunet.2004.07.004

DO - 10.1016/j.neunet.2004.07.004

M3 - Article

C2 - 15555870

AN - SCOPUS:9144252205

VL - 17

SP - 1345

EP - 1362

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

IS - 8-9

ER -