On the role of object knowledge in reference production

Effects of color typicality on content determination

H.G.W. Westerbeek, R.M.F. Koolen, A.A. Maes

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

170 Downloads (Pure)

Abstract

In two language production experiments, we investigated whether stored knowledge of the typical color of objects affects spoken reference. In experiment 1, human speakers referred to objects with colors ranging from very typical (e.g., red tomato) to very atypical (e.g., blue pepper). The probability that speakers redundantly include color in their descriptions was almost linearly predicted by the degree of atypicality. In experiment 2, we extended this finding to references to objects for which color is inherently a less salient property in stored knowledge (i.e., objects with a highly characteristic shape, making color less important for recognition). Following these findings that typicality affects reference production, we conclude that speakers utilize stored knowledge about everyday objects they refer to. We discuss the implications of our findings for artificial agents that generate natural language, arguing that computational models fall short in captur- ing general knowledge about typical properties of objects.
Original languageEnglish
Title of host publicationCogSci 2014
Subtitle of host publicationCognitive Science Meets Artificial Intelligence: Human and Artificial Agents in Interactive Contexts
EditorsPaul Bello, Marcello Guarini, Marjorie McShane, Brian Scassellati
Publication statusPublished - 2014
EventCogSci 2014 - Québec City, Canada
Duration: 23 Jul 201426 Jul 2014

Conference

ConferenceCogSci 2014
CountryCanada
CityQuébec City
Period23/07/1426/07/14

Keywords

  • reference production
  • color typicality
  • content determination
  • visual saliency
  • AI models of reference production

Cite this

Westerbeek, H. G. W., Koolen, R. M. F., & Maes, A. A. (2014). On the role of object knowledge in reference production: Effects of color typicality on content determination. In P. Bello, M. Guarini, M. McShane, & B. Scassellati (Eds.), CogSci 2014: Cognitive Science Meets Artificial Intelligence: Human and Artificial Agents in Interactive Contexts
Westerbeek, H.G.W. ; Koolen, R.M.F. ; Maes, A.A. / On the role of object knowledge in reference production : Effects of color typicality on content determination. CogSci 2014: Cognitive Science Meets Artificial Intelligence: Human and Artificial Agents in Interactive Contexts. editor / Paul Bello ; Marcello Guarini ; Marjorie McShane ; Brian Scassellati. 2014.
@inproceedings{6a651444c3454d9999690346b19459ef,
title = "On the role of object knowledge in reference production: Effects of color typicality on content determination",
abstract = "In two language production experiments, we investigated whether stored knowledge of the typical color of objects affects spoken reference. In experiment 1, human speakers referred to objects with colors ranging from very typical (e.g., red tomato) to very atypical (e.g., blue pepper). The probability that speakers redundantly include color in their descriptions was almost linearly predicted by the degree of atypicality. In experiment 2, we extended this finding to references to objects for which color is inherently a less salient property in stored knowledge (i.e., objects with a highly characteristic shape, making color less important for recognition). Following these findings that typicality affects reference production, we conclude that speakers utilize stored knowledge about everyday objects they refer to. We discuss the implications of our findings for artificial agents that generate natural language, arguing that computational models fall short in captur- ing general knowledge about typical properties of objects.",
keywords = "reference production, color typicality, content determination, visual saliency, AI models of reference production",
author = "H.G.W. Westerbeek and R.M.F. Koolen and A.A. Maes",
year = "2014",
language = "English",
editor = "Paul Bello and Marcello Guarini and Marjorie McShane and Brian Scassellati",
booktitle = "CogSci 2014",

}

Westerbeek, HGW, Koolen, RMF & Maes, AA 2014, On the role of object knowledge in reference production: Effects of color typicality on content determination. in P Bello, M Guarini, M McShane & B Scassellati (eds), CogSci 2014: Cognitive Science Meets Artificial Intelligence: Human and Artificial Agents in Interactive Contexts. CogSci 2014, Québec City, Canada, 23/07/14.

On the role of object knowledge in reference production : Effects of color typicality on content determination. / Westerbeek, H.G.W.; Koolen, R.M.F.; Maes, A.A.

CogSci 2014: Cognitive Science Meets Artificial Intelligence: Human and Artificial Agents in Interactive Contexts. ed. / Paul Bello; Marcello Guarini; Marjorie McShane; Brian Scassellati. 2014.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

TY - GEN

T1 - On the role of object knowledge in reference production

T2 - Effects of color typicality on content determination

AU - Westerbeek, H.G.W.

AU - Koolen, R.M.F.

AU - Maes, A.A.

PY - 2014

Y1 - 2014

N2 - In two language production experiments, we investigated whether stored knowledge of the typical color of objects affects spoken reference. In experiment 1, human speakers referred to objects with colors ranging from very typical (e.g., red tomato) to very atypical (e.g., blue pepper). The probability that speakers redundantly include color in their descriptions was almost linearly predicted by the degree of atypicality. In experiment 2, we extended this finding to references to objects for which color is inherently a less salient property in stored knowledge (i.e., objects with a highly characteristic shape, making color less important for recognition). Following these findings that typicality affects reference production, we conclude that speakers utilize stored knowledge about everyday objects they refer to. We discuss the implications of our findings for artificial agents that generate natural language, arguing that computational models fall short in captur- ing general knowledge about typical properties of objects.

AB - In two language production experiments, we investigated whether stored knowledge of the typical color of objects affects spoken reference. In experiment 1, human speakers referred to objects with colors ranging from very typical (e.g., red tomato) to very atypical (e.g., blue pepper). The probability that speakers redundantly include color in their descriptions was almost linearly predicted by the degree of atypicality. In experiment 2, we extended this finding to references to objects for which color is inherently a less salient property in stored knowledge (i.e., objects with a highly characteristic shape, making color less important for recognition). Following these findings that typicality affects reference production, we conclude that speakers utilize stored knowledge about everyday objects they refer to. We discuss the implications of our findings for artificial agents that generate natural language, arguing that computational models fall short in captur- ing general knowledge about typical properties of objects.

KW - reference production

KW - color typicality

KW - content determination

KW - visual saliency

KW - AI models of reference production

M3 - Conference contribution

BT - CogSci 2014

A2 - Bello, Paul

A2 - Guarini, Marcello

A2 - McShane, Marjorie

A2 - Scassellati, Brian

ER -

Westerbeek HGW, Koolen RMF, Maes AA. On the role of object knowledge in reference production: Effects of color typicality on content determination. In Bello P, Guarini M, McShane M, Scassellati B, editors, CogSci 2014: Cognitive Science Meets Artificial Intelligence: Human and Artificial Agents in Interactive Contexts. 2014