Efficient compression in color naming and its evolution

Abstract

We derive a principled information-theoretic account of cross-language semantic variation. Specifically, we argue that languages efficiently compress ideas into words by optimizing the information bottleneck (IB) trade-off between the complexity and accuracy of the lexicon. We test this proposal in the domain of color naming and show that () color-naming systems across languages achieve near-optimal compression; () small changes in a single trade-off parameter account to a large extent for observed cross-language variation; () efficient IB color-naming systems exhibit soft rather than hard category boundaries and often leave large regions of color space inconsistently named, both of which phenomena are found empirically; and () these IB systems evolve through a sequence of structural phase transitions, in a single process that captures key ideas associated with different accounts of color category evolution. These results suggest that a drive for information-theoretic efficiency may shape color-naming systems across languages. This principle is not specific to color, and so it may also apply to cross-language variation in other semantic domains.

Origin of perseveration in the trade-off between reward and complexity

“…Rate distortion theory holds promise as a vehicle for general principles because it unifies two frameworks (information theory and statistical decision theory) that already by themselves have broad explanatory reach. Rate distortion theory has been successfully applied to many different cognitive phenomena, ranging from working memory (Sims et al, 2012;Sims, 2016) and absolute identification (Sims, 2018) to language (Zaslavsky et al, 2018) and motor control (Schach et al, 2018). A complete theory in these domains will eventually use mechanistic models to constrain the rate distortion analysis.…”

Color Naming Reflects Both Perceptual Structure and Communicative Need

“…in CIELAB color space, in which the Euclidean distance between nearby colors corresponds roughly to their perceptual dissimilarity (Brainard, ; but see also Komarova & Jameson, ). This visualization shows that there exist potentially relevant perceptual asymmetries of color—and in fact this perceptual structure has been used to explain patterns of color naming across languages (Jameson & D’Andrade, ; Regier et al., ; Zaslavsky et al., ). We wished to understand whether the structure of perceptual color space could also explain the asymmetry in precision documented by Gibson et al., or that in need, or both—a possibility acknowledged by Gibson et al.…”

“…In each case, following Zaslavsky et al. (), we evaluated the CAP for each language l (real or artificial) with respect to its color naming distribution , and averaged together these language‐specific priors in order to infer a universal need distribution . That is, we defined where L = 111 is the number of languages in the WCS+ dataset.…”

“…The forces that give rise to this constrained variation have long been debated, and it is often held that a major role is played by perceptual structure (e.g., Kay & McDaniel, 1978). A variant of this view emphasizes in addition the importance of communicative forces, and it argues that languages divide perceptual color space into categories in ways that support efficient communication (Baddeley & Attewell, 2009;Jameson & D'Andrade, 1997;Lindsey et al, 2015;Regier et al, 2007;Regier et al, 2015;Zaslavsky et al, 2018).…”

Red, yellow, green, and blue are not particularly colorful

“…. It has been suggested that similarities of color categorization across cultures is optimized to communicate efficiently about the irregularly shaped perceptual color space (Abbott, Griffiths, & Regier, 2016;Zaslavsky, Kemp, Regier, & Tishby, 2018;Zaslavsky, Kemp, Tishby, & Regier, 2019). Evidence for this idea is based on the irregular distribution of maximally saturated Munsell chips and naming data from the World Color Survey (WCS).…”