Highlight only the class currently in focus
Aliases: focus highlight · salience budget · highlight scope
What it is
Focus-scoped highlighting binds the strongest salient difference to one target set in the current task while keeping the rest readable but recessed. If many classes simultaneously receive equally strong accent color, outline, and enlargement, the figure–ground relation is diluted. “One class” means one current attentional set that may contain several related marks, not a rule that an interface can select only one datum.
Why it happens
Salience is relational. When many items differ from the background and from each other, distractor heterogeneity rises and no single difference defines the target. Area, position, motion, and semantic priority also contribute to hierarchy. Reducing competition can guide search, but excessive de-emphasis can destroy contextual comparison, reduce accessibility, or falsely suggest that unselected data are unimportant.
Studying it
For a real task, compare no highlight, one focused set, and several competing highlight classes. Measure first fixation, location time, errors, context comparison, and comprehension. Vary selection count, density, display size, color-vision condition, and motion, and ask whether highlight is understood as selection, search result, anomaly, alert, or importance. Test residual state after focus changes and synchronization with keyboard and screen reader.
Where it stops holding
Collaboration, multiple alert levels, or direct comparison of several series may require concurrent states. Use structural regions, labels, or an ordered hierarchy instead of making every state pop equally. Instructional annotation may reveal several targets in stages. When nonfocused data provide the comparison context, they must retain sufficient contrast and query access.
Applying it
- Define the highlighted object, reason, and lifecycle. Selection, search results, anomaly, and alert should not share one ambiguous style.
- Reserve strongest contrast for one task set at a time; express other states with labels, position, boundaries, or lower hierarchy and keep legends consistent.
- Support clearing, switching, and keyboard traversal; expose selected state, name, and value to assistive technology rather than changing color alone.
- Verify that users locate and interpret focus while still comparing context. Split hierarchy or views when background vanishes or classes compete.
Related
- Same group: U1.12.1 A single salient difference can be found in parallel · U1.12.2 Layering multiple encodings cancels the parallel advantage
- Adjacent: U1.13.1 Proximity and similarity decide the grouping readers see · U1.10.5 Redundant channels that disagree semantically give contradictory cues
- Search terms:
highlight scope·salience budget·focus context