← CHI 2026 map

CHI '26 · Honorable mention · full-paper review · confidence medium-high

Redundant is Not Redundant: Automating Efficient Categorical Palettes Design Unifying Color & Shape Encodings with CatPAW

Chin Tseng , Arran Zeyu Wang , Ghulam Jilani Quadri , Danielle Albers Szafir

DOI PDF Program page Page JSON

This is a strong CHI paper because it does two things well: it shows that redundant color–shape encoding is genuinely interaction-sensitive, and it turns that result into a usable design tool. The contribution is more than a system demo; it is an empirically grounded design method with clear scope and a credible limitation profile.

Axes Lens

Rare contribution shape, typical evidence profile. The point here is not a score. It is to show what kind of claim the paper makes, and whether the evidence pattern is unusual or baseline in this 268 -review set.

Review Summary

This paper’s main value is that it refuses the common shortcut in categorical visualization design: the idea that redundancy can be built by independently selecting a good color palette and a good shape palette and then simply combining them. The authors show that this intuition is incomplete because the pairing itself changes performance, and that some combinations that look strong in isolation do not remain strong when used redundantly. That is a meaningful departure from common-sense design practice, especially in a CHI context where many tools and guidelines still lean on compositional heuristics. The novelty is also well aligned with the evidence: four crowdsourced experiments are used to establish the interaction effects and to identify where redundancy helps most, with the strongest gains reported for 5–8 categories. The paper then packages those findings into CatPAW, which makes the contribution more than a one-off empirical result. It becomes a practical artifact for designers, grounded in measured behavior rather than intuition. The limitations are appropriately narrow and important: the tested space is restricted to a subset of colors and shapes, a white background, and a correlation-judgment task in multiclass scatterplots. The tool does not yet handle new shapes or multiple encoded dimensions, so it should be read as a design-family contribution rather than a universal palette engine. Overall, the paper is strong because the claim, the experiments, and the tool all line up, and because the authors are explicit about where the result should and should not be generalized.

What Changed

Canon before

Prior CHI visualization guidance treated redundant color+shape encoding as a straightforward robustness tactic: choose good colors and good shapes, then combine them to improve categorical distinction. The paper challenges that simplifying assumption.

Departure from common sense

The paper argues that effective redundant palette design is not just a matter of combining individually strong color and shape choices. Instead, the pairing itself matters, and top-performing colors or shapes do not necessarily yield top-performing redundant encodings when combined.

Actual novelty

CatPAW is a new categorical palette design tool that operationalizes the paper’s empirical findings into a practical system for generating redundant color–shape palettes. Its novelty is not only the tool itself, but the fact that it is grounded in four crowdsourced experiments and a model of channel interaction rather than ad hoc design rules.

Evidence

The paper reports four crowdsourced experiments on redundant color–shape encodings for multiclass scatterplots and a correlation-judgment task. Across these studies, redundancy improves accuracy, especially for 5–8 categories, and the results reveal interaction effects between colors and shapes. The authors then translate these findings into CatPAW, a palette design tool.

Limits

Method limits

The evidence is experimentally grounded but bounded to the tested tasks, category ranges, and palette subsets. The paper’s own limitations note that only a subset of colors and shapes were assessed, only a white background was used, and the work does not cover multiple data dimensions or new shapes.

Deployment limits

CatPAW is positioned as a design aid for categorical visualization, not a general-purpose visualization generator. Its current scope does not extend to adapting to new shapes or simultaneously encoding multiple data dimensions, so deployment is best limited to redundant categorical palettes in similar scatterplot-like settings.

Boundary conditions

The strongest reported benefits occur for 5–8 categories, and the findings are tied to correlation judgments in multiclass scatterplots. The paper also indicates that palette performance depends on specific color-shape pairings, so the model should be used within the tested design family rather than assumed to transfer universally.

Manual check: Verify that the full-text span used for evidence contains the quoted limitation and tool-description language exactly as cited; the stage-A packet’s exact_evidence_spans are not usable for claims.

Position in field

This work sits at the intersection of visualization design guidance, empirical perception studies, and design automation. It contributes a more specific account of when redundant encodings help and turns that account into a practical palette-generation tool, moving beyond generic advice toward empirically constrained automation.

Abstract

Colors and shapes are commonly used to encode categories in multi-class scatterplots. Designers often combine the two channels to create redundant encodings, aiming to enhance class distinctions. However, evidence for the effectiveness of redundancy remains conflicted, and guidelines for constructing effective combinations are limited. This paper presents four crowdsourced experiments evaluating redundant color–shape encodings and identifying high-performing configurations across different category numbers. Results show that redundancy significantly improves accuracy in assessing class-level correlations, with the strongest benefits for 5–8 categories. We also find pronounced interaction effects between colors and shapes, underscoring the need for careful pairing in designing redundant encodings. Drawing on these findings, we introduce a categorical palette design tool that enables designers to construct empirically grounded palettes for effective categorical visualization. Our work advances understanding of categorical perception in data visualization by systematically identifying effective redundant color–shape combinations and embedding these insights into a practical palette design tool.