Use hypothetical outcome plots for ambiguous trend judgments

Encode uncertainty as animated hypothetical outcome samples when readers must decide which trend most likely produced noisy observations. For example, replace bar-chart error bars with a bar HOP, or replace a static line ensemble with a line HOP, when the chart asks viewers to choose between competing trends.

Animated hypothetical outcomes let readers experience variability as repeated sampled cases instead of decoding a summary statistic. That makes the uncertainty part of the comparison rather than a separate abstraction that readers may ignore or misread.

Mechanism: Sequential samples expose how noisy observations can still come from different underlying trends, so readers can judge the likelihood of each trend with less evidence.

Evidence: In experiment 1, bar HOPs produced lower JNDs than bar charts with error bars, meaning participants could identify the underlying trend with less evidence. In experiment 2, regular-rate line HOPs also produced lower JNDs than static line ensembles for the same trend-inference task (Kale et al., 2019).

• User Goal: Decide which of a small number of alternative trends is more likely.
• Task: Compare uncertain trends and infer which one produced a noisy sample.
• Data: Quantitative ordered-time samples with visible sampling error or other process noise.
• Chart Setting: The chart can show uncertainty as repeated samples on the same bars or lines used for the main series.
• Audience: Untrained readers with limited statistical background.
• Success Criterion: Readers make correct trend judgments from more ambiguous samples.

Break it when: Viewers cannot or will not watch the display long enough to integrate several frames. Why: HOPs depend on temporal integration, so the uncertainty cue weakens when the motion is not actually read.

Sacrifice: A HOP needs time to play repeated samples instead of showing everything in one static glance. Risk: Readers who ignore the motion may miss the variability cue. Mitigation: Keep the motion focused on the uncertainty comparison and use a trackable playback rate.

Mistake: Adding error bars or another summary-only uncertainty layer for the same ambiguous trend judgment. Why it fails: Summary marks hide the sampled process that readers need to compare, so viewers need stronger evidence before they can make the correct call.

Failure Sign: Readers succeed only when the slope is obvious and fail on borderline samples. Quick Check: Show the same ambiguous sample with the current uncertainty display and with a HOP; if correct choices appear only with the HOP, the current encoding is under-communicating variability. Stronger Test: Run repeated forced-choice judgments across samples with different evidence levels and check whether the HOP lowers the evidence needed for correct decisions.

• Draw repeated plausible outcomes from each model or trend and animate them as the same bars or lines used in the chart.
• Remove summary uncertainty marks such as error bars for this judgment task.
• If the current uncertainty view is a static line ensemble, convert those sampled lines into a sequential HOP.