Avoid hard boundaries when encoding continuous uncertainty

Avoid enclosing continuous or probabilistic ranges with hard boundaries when readers should interpret gradual uncertainty. For example, replace a bounded uncertainty region with a graded fade or an ensemble of possible paths, and avoid interval displays that look like discrete high and low endpoints when the message is probabilistic spread.

A hard outline implies categorical inside-versus-outside status. Readers then treat nearby values across the boundary as more different than they really are, or read the bounded region as a deterministic forecast rather than a probability distribution.

Mechanism: Boundary marks trigger a containment heuristic and make the enclosed area feel semantically distinct, which distorts judgments about continuity and likelihood.

Evidence: The review synthesizes evidence that bounded circles, cones, and interval-like displays can produce containment and deterministic construal errors, while graded or ensemble-style displays can better support probabilistic interpretation (Padilla et al., 2018).

Notes: Summarizing continuous data into bins can simplify a display, but the boundary itself can become the dominant message.

• User Goal: Judge uncertainty, likelihood, or probable location/path.
• Task: Compare nearby positions or values within a continuous or probabilistic range.
• Data: Continuous, uncertain, or distributional information.
• Chart Setting: Maps or interval-like displays that could be drawn with enclosing outlines or bounded regions.
• Audience: Nonexperts or mixed audiences interpreting probability visually.
• Success Criterion: Readers treat the data as gradual or probabilistic rather than categorical.

Break it when: The display intentionally defines discrete bins or a true cutoff where inside and outside should be treated differently. Why: In that case the boundary is not an artifact; it is the intended classification.

Sacrifice: You lose a crisp categorical edge. Risk: Some readers may find graded uncertainty less immediate than a simple enclosed region.

Mistake: Draw a bold enclosing region around continuous uncertainty and expect readers to infer gradual probability from it. Why it fails: The outline encourages an inside/outside reading and hides variation within the region.

Failure Sign: Readers treat points just across a boundary as categorically different, or they describe the display as a sure path or fixed range. Quick Check: Ask whether two nearby positions on opposite sides of the boundary are judged more differently than two equally close positions within the same region. Stronger Test: Compare the bounded version against a graded or ensemble version and see whether judgments become less categorical.

• Remove or soften enclosing outlines around continuous uncertainty.
• Replace a single bounded summary region with a graded fade when the message is gradual likelihood.
• Replace a bounded path region with an ensemble of possible paths when the message is spread across alternatives.
• Redesign interval-like uncertainty displays so they do not read as discrete endpoints.