Eye-Tracking Study Reveals What Makes an Atlas Cover Beautiful

Atlas covers occupy a curious place in the world of design. They are the first thing a reader sees, the visual handshake between a cartographic institution and its audience, and yet their creation has long been governed by intuition, tradition, and the accumulated habits of publishing houses rather than by empirical evidence. A new eye-tracking study published in PLOS One by Pizhao Hu, Hao Fang, Yuhui Gui, Nai Yang, Gangqiang Yu, and Zixi Huang set out to change that, asking a deceptively simple question: which visual features of an atlas cover actually make people find it attractive, and do those same features shape where their eyes actually go?

The answer, it turns out, is more complicated and more interesting than a straightforward yes. The researchers found that three design variables—color harmony, graphic style, and layout configuration—do influence how attractive viewers perceive a cover to be. But when they compared those subjective judgments against objective measurements of gaze behavior, the two layers of evidence only partially overlapped. Beauty, at least on the cover of an atlas, and attention do not march in lockstep.

To reach this conclusion, the team constructed a rigorously controlled experimental framework. They generated eight atlas-cover stimuli according to a 2×2×2 within-subject factorial design, systematically varying color harmony (contrast versus analogous schemes), graphic style (concrete landscape imagery versus abstract map symbols), and layout configuration (centralized versus decentralized arrangements). Forty valid participants each viewed the full set of covers across two counterbalanced viewing cycles, producing sixteen trials per participant. The two repeated observations for each participant–cover combination were averaged before analysis, a precaution that reduced the influence of moment-to-moment fluctuations in attention and mood.

The subjective side of the experiment yielded clear and statistically robust results. Participants rated covers featuring contrast harmony—palettes built on deliberate chromatic opposition—as significantly more attractive than their analogous counterparts. Concrete landscape imagery proved even more powerful: covers depicting recognizable scenery outperformed those built from map symbols with a mean difference of 0.66 on the rating scale, a difference that survived Holm correction for multiple comparisons with a p-value of 0.003. Color harmony contributed a smaller but highly reliable boost (mean difference 0.29, p less than 0.001), and centralized layouts added a further modest advantage (mean difference 0.14, p equal to 0.031). In short, what people said they liked most was a cover that showed them a beautiful, recognizable place in a striking color scheme, arranged with a clear visual center.

The eye-tracking data told a different and partially independent story. When the researchers examined total fixation duration and fixation count—the two most widely used indicators of sustained visual attention—they found that layout configuration was the dominant factor. Centralized layouts held viewers’ eyes longer, adding an average of 0.34 seconds of total fixation duration and 1.13 additional fixations compared with decentralized arrangements, both differences highly significant after correction. This makes intuitive sense once one pictures the stimuli: a cover with a strong central focal point gives the eye an obvious anchor, encouraging repeated returns and longer dwell times, whereas a decentralized composition scatters attention across competing regions and invites quicker exits.

Graphic style, by contrast, exerted its influence mainly at the very beginning of viewing, shaping initial attentional orienting rather than sustained engagement. Landscape imagery captured the eye early, presumably because recognizable scenes trigger rapid semantic processing and emotional resonance, but this early advantage did not necessarily translate into longer overall inspection. Color harmony showed only small effects on visual exploration in either direction, suggesting that while palette strongly colors our aesthetic verdict, it does relatively little to choreograph the movement of our eyes across the page.

The most provocative finding emerged when the researchers formally tested whether perceived attractiveness and gaze behavior were associated at all. Using linear mixed-effects association models—the same statistical machinery that handled the factorial comparisons—they found no robust relationship between subjective attractiveness ratings and any of the four gaze-based indicators once Holm–Bonferroni correction was applied, with all corrected p-values at or above 0.110. In other words, a cover that people loved was not reliably a cover they stared at, and a cover that held their gaze was not reliably a cover they admired.

This partial dissociation between aesthetic evaluation and visual attention carries real theoretical weight. It suggests that the two responses are governed by partly separable mechanisms: gaze behavior reflects the allocation of attentional resources, driven by salience, layout structure, and the search for information, while aesthetic preference reflects an affective judgment integrating color emotion, semantic content, and compositional harmony. The authors frame this as a dual-layer logic of cover design, in which layout serves as the scaffolding that organizes visual processing, while color harmony and graphic style operate on a second layer, enhancing emotional appeal and thematic expressiveness without necessarily prolonging looking.

For designers and publishers, the practical implications are tangible. If the goal is to make a cover feel beautiful, the evidence points squarely toward concrete landscape imagery, supported by a contrast-harmonious palette and a reasonably centralized composition. If the goal is to maximize the time a browser spends visually engaged with the cover—for instance, on a crowded bookshop shelf or a dense online catalog—then layout configuration becomes the primary lever, with a strong central focal point doing the heavy lifting. The two goals can be pursued simultaneously, but they should not be conflated, because a design that wins one contest does not automatically win the other.

The study also demonstrates the value of bringing experimental rigor to a domain that has rarely received it. Eye-tracking research has long illuminated how people read maps, scan packaging, and navigate digital interfaces, but atlas covers—a hybrid genre blending cartographic symbolism with book-cover aesthetics—had been largely left to heuristic experience. By combining a factorial stimulus design with mixed-effects modeling and rigorous correction for multiple comparisons, this work offers a template for evidence-based design research in adjacent fields, from textbook covers to data-visualization dashboards. It also issues a caution to the growing industry of attention metrics: what the eye does and what the mind prefers are related but distinct phenomena, and measuring only one risks misunderstanding the other. For the humble atlas cover, the path to a reader’s heart apparently runs through the landscape painting and the color wheel, while the path to a reader’s eyes runs through the geometry of the layout—and the two paths, though they share a starting point, lead to different destinations.

Subject of Research: Eye-tracking investigation of how color harmony, graphic style, and layout shape aesthetic preference and visual attention in atlas-cover design

Article Title: Aesthetic preference and gaze-based visual attention in atlas-cover design: Evidence from an eye-tracking study

Article References: Hu, P., Fang, H., Gui, Y., Yang, N., Yu, G., & Huang, Z. (2026). Aesthetic preference and gaze-based visual attention in atlas-cover design: Evidence from an eye-tracking study. PLOS One, 21(10), e0359864. https://doi.org/10.1371/journal.pone.0359864

Image Credits: AI Generated

DOI: 10.1371/journal.pone.0359864

Keywords: eye-tracking, atlas cover design, aesthetic preference, visual attention, color harmony, layout configuration, graphic style, fixation duration, cartography, PLOS One, mixed-effects models, design research