Examining Implicit Discretization in Spectral Schemes

Abstract

Two of the primary reasons rainbow color maps are considered ineffective trace back to the idea that they implicitly discretize encoded data into hue‐based bands, yet no research addresses what this discretization looks like or how consistent it is across individuals. This paper presents an exploratory study designed to empirically investigate the implicit discretization of common spectral schemes and explore whether the phenomenon can be modeled by variations in lightness, chroma, and hue. Our results suggest that three commonly used rainbow color maps are implicitly discretized with consistency across individuals. The results also indicate, however, that this implicit discretization varies across different datasets, in a way that suggests the visualization community's understanding of both rainbow color maps, and more generally effective color usage, remains incomplete.

Rainbow Dash: Intuitiveness, Interpretability and Memorability of the Rainbow Color Scheme in Visualization

“…Also, in a relative color distance judgement task, the RC resulted in higher error rates and longer response times than the SC and a diverging scheme [47]. Discretization appears to be difficult with the RC too, i.e., people were inconsistent in recognizing and placing boundaries between rainbow hue bands [42], further confirming that the RC is not perceptually uniform.…”

Rainbows Revisited: Modeling Effective Colormap Design for Graphical Inference

“…The tendency for rainbows to create boundaries between hues (sometimes referred to as a 'hue banding' effect) is believed to mislead viewers [51]. However, the impact of such banding on data interpretation is still poorly understood [37,39].…”

Color Nameability Predicts Inference Accuracy in Spatial Visualizations

“…A potential explanation starts by recognizing that color names serve to discretize a continuous data domain into a number of color categories that are easy to reference. Though researchers have long argued against this kind of categorization [BI07; QPCM19], we suggest that it can be helpful in the more interpretive tasks like graphical inference. Here, the observer has to identify a plot containing an unusual data distribution.…”