coordinating perceptually grounded categories through language: a case study for colour

Abstract

The paper proposes a number of models to examine through what mechanisms a population of autonomous agents could arrive at a repertoire of perceptually grounded categories that is sufficiently shared to allow successful communication. The models are inspired by the main approaches to human categorisation being discussed in the literature: nativism, empiricism, and culturalism. Colour is taken as a case study. Although the paper takes no stance on which position is to be accepted as final truth with respect to human categorisation and naming, it points to theoretical constraints that make each position more or less likely and contains clear suggestions on what the best engineering solution would be. Specifically, it argues that the collective choice of a shared repertoire must integrate multiple constraints, including constraints coming from communication.

Evolutionary Models of Color Categorization Based on Discrimination

“…They also indirectly give an impression how systematic variation in observer-type heterogeneity could influence convergent category solutions (as suggested by Jameson, 2005a)-a topic of recent discussion in the color categorization literature (Steels & Belpaeme 2005). Finally, both results accord with the organizational framework for human color categorization described by Jameson (2005d, pp.…”

“…The modeling methods used here resemble those found in existing research on the evolution of general communication and signaling systems (Grice 1957, 1989, Lewis 1969, Skyrms 1996, 2002, Komarova & Niyogi 2004, Komarova 2004 and from computational modeling specific to perceptually-based color categorization (Zuidema & Westernann 2003, Belpaeme & Bleys 2005, Steels & Belpaeme 2005.…”

Word Meanings across Languages Support Efficient Communication

“…Other researchers have developed formal accounts of linguistic and cultural evolution (e.g. Kirby, 2002;Jäger, 2007;Kalish, Griffiths, & Lewandowsky, 2007), and such evolutionary accounts have helped to explain why color categories (Steels & Belpaeme, 2005;Dowman 2007;Komarova, Jameson, & Narens, 2007;Xu, Dowman, & Griffiths, 2013) and kinship categories (Epling, Kirk, & Boyd, 1973;Jordan, 2011) take the forms that they do. A natural direction for future work is to integrate our account with such an evolutionary perspective.…”

Self-organization in Communicating Groups: The Emergence of Coordination, Shared References and Collective Intelligence

“…Probably the most famous illustration can be found in the work of Steels and his coworkers [Steels, 1998;Steels & Voght, 1997;Steels & Belpaeme, 2005], who made a series of simulations of the self-organization of the fundamental components of language. The simulations start with a group of software agents or robots (hardware agents) that have to learn to communicate without any a priori shared language.…”

“…I will not here go into the further simulation research of Steels and his colleagues which illustrates how also other aspects of language, such as semantic categories [Steels & Belpaeme, 2005], phonetics [de Boer 2000ab] and grammar [Steels, 2005], can self-organize out of random interactions. Although the rules that govern the individual interactions may be more complicated for these cases, the fundamental mechanism of mutual alignment that is amplified and propagated via positive feedback remains the same.…”