Group event icons when readers must compare risk magnitude

Arrange event icons as one contiguous group instead of scattering them across the array. For example, place affected icons together at the bottom of each matrix rather than randomly distributing them when readers need to judge which option has the larger or smaller risk.

Grouped icons make the risk magnitude easier to read as a part-whole quantity. Scattered icons may suggest randomness, but they make it harder to judge how many events there are and which option is better.

Mechanism: A contiguous block supports faster magnitude assessment and easier comparison, while random placement increases the work needed to count or estimate the event total.

Evidence: Static grouped icon arrays outperformed static scattered icon arrays on gist knowledge, treatment choice, and graph ratings, and scattered variants generally performed poorly unless they later revealed a grouped arrangement (Zikmund-Fisher et al., 2012).

• User Goal: Identify which option has the lower or higher risk.
• Task: Compare close risk values or recognize equal versus unequal risks.
• Data: Quantitative event rates shown as part-whole counts in an icon array.
• Chart Setting: Risk pictographs used to support understanding and choice.
• Audience: Adults with mixed numeracy.
• Success Criterion: More accurate gist knowledge and more accurate selection of the better option.

Break it when: The main communication goal is to emphasize the randomness of who experiences the event rather than to support a direct magnitude comparison. Why: The paper explicitly treats scattered layouts as a way to convey randomness, even though they hinder magnitude judgment.

Sacrifice: You lose some visual cue that event occurrence is randomly distributed across individuals. Risk: Readers may only see a tidy block and miss the intended randomness message. Mitigation: If the randomness cue matters, show it briefly and then resolve the display into the grouped arrangement before the comparison.

Mistake: Using a scattered event layout for the actual comparison because it feels more realistic. Why it fails: Readers were less accurate and rated those displays worse.

Failure Sign: Readers struggle to tell whether two risks are equal or which one is larger. Quick Check: Show the same risk values once as grouped icons and once as scattered icons; ask which option is riskier and keep the grouped version if accuracy is higher. Stronger Test: Include a graph rating question, since grouped layouts were also preferred.

• Cluster all event icons into one contiguous block.
• Use the grouped arrangement at the point where readers must decide which option is safer.
• Remove random scattering from the final comparison view.
• If you need an initial scattered view, add a transition that ends in the grouped arrangement.