How Visual Design Can Improve Your Research Figures

Recent Trends

Over the past few publication cycles, scientific journals and preprint servers have seen a marked shift in the quality of figures. Editors note that manuscripts now frequently include figures that mirror the visual standards of data journalism—cleaner grids, intentional color palettes, and guided focal points. Key drivers include:

• Wider adoption of Python’s Matplotlib/seaborn and R’s ggplot2 with custom themes, replacing default Excel and SigmaPlot outputs.
• Rise of dedicated figure-review checklists at major conferences (e.g., AGU, SfN) that explicitly reward accessibility features.
• Increased availability of free, online color-palette generators and accessibility simulators (e.g., Coblis, ColorBrewer).

Background

Research figures have long been undervalued as mere containers for data. But a growing body of evidence from cognitive science shows that humans process visual information orders of magnitude faster than text. Poor design—cluttered axes, rainbow colormaps, illegible fonts—creates unnecessary cognitive load, slowing comprehension and increasing the risk of misinterpretation. Core principles now recognized as essential include:

• Perceptual uniformity: choosing colormaps (e.g., viridis, cividis) that preserve data encoding across grayscale and for color-vision deficiencies.
• Consistent typography: using a single sans-serif family for axis labels, legends, and annotations.
• Hierarchical layout: placing the most critical trend or comparison in the upper-left or center of a figure panel.

User Concerns

Researchers who want to improve their figures routinely express three categories of hesitation. Practical ranges and decision criteria help address each:

• Time investment: Learning a new plotting library or design tool can take 8–20 hours of focused effort. A pragmatic approach is to master one template per figure type (e.g., a journal-specific .Rmd or .ipynb) and reuse it across projects.
• Loss of data fidelity: Design changes should never alter underlying values. A safe workflow is to generate raw data plots first, then apply styling in a separate code block or non-destructive layer. Version control (Git) protects against accidental edits.
• Journal constraints: Most journals require specific dimensions, resolution, and color mode (CMYK for print, RGB for online). Checking author guidelines early and building figures to those specs avoids costly rework.

Likely Impact

Early evidence from field comparisons suggests that applying basic visual design principles can yield measurable benefits—though outcomes depend on context:

• Faster peer review: Well-structured figures reduce reviewer questions about data presentation, potentially shaving weeks off the review cycle.
• Higher citation rates: Anecdotal accounts from systematic reviews indicate that papers with intentionally designed figures receive more citations than those with default-styled graphics, especially across interdisciplinary audiences.
• Reduced misinterpretation: For complex multi-panel figures, clear labeling and consistent color encoding lower the chance that readers misread trends or miss error bars.
• Broader accessibility: Figures designed with colorblind-friendly palettes and sufficient contrast are more likely to be understood by readers with visual impairments, improving equity in science communication.

What to Watch Next

Several developments could reshape how researchers approach figure design in the next two to three years:

• AI-assisted figure generation: Tools that accept natural-language descriptions and raw data, then suggest layouts and colormaps, are emerging. Early versions remain limited, but iterative refinement may soon reduce the design learning curve.
• Interactive and responsive figures: Some journals now accept HTML-embedded figures that allow readers to hover, zoom, or toggle data layers. Adoption is still low but growing in fields like genomics and climate science.
• Cross-publisher design standards: The STM Association and several major publishers are exploring a common figure style guide. If adopted, it would greatly simplify compliance for researchers who submit to multiple journals.
• Integration of alt-text and metadata: Automated generation of descriptive alt-text from figure code is gaining traction, supporting both machine readability and accessibility mandates.