Use position encoding instead of slope encoding for pairwise relation judgments

Encode the value or delta by position from a common baseline instead of by slope when viewers must judge relations across many pairs. For example, replace a row of angled line segments with baseline-aligned dashes for individual values or with a single delta-position mark when readers must find the opposite relation or judge which direction is more prevalent.

Row-aligned slopes make viewers parse mirrored configurations rather than a simpler positional readout. Position encoding reduces that bottleneck when viewers scan many pairwise relations.

Mechanism: Position encoding lets readers compare distance from a baseline directly, while slope encoding requires interpreting angled segments that can group visually into mirrored shapes in a row.

Evidence: In the visual search task and the proportion-judgment task, slope encodings produced significantly worse performance than position encodings. The paper also notes that the row arrangement used here may make slopes appear as grouped /\ and \/ shapes, which likely contributes to the poorer relation reading (Nothelfer & Franconeri, 2020).

Notes: The source presents this result for the specific displays and tasks tested, not as a universal ranking for all possible slope layouts.

• User Goal: Find which pair has a target increase or decrease pattern, or decide which relation direction appears more often.
• Task: Scan many pairwise relations quickly rather than inspect one pair in isolation.
• Data: Quantitative values organized into pairs with directional differences.
• Chart Setting: The current design uses row-aligned slope marks or is considering slope as the main encoding for value or delta.
• Audience: Viewers making quick scans or brief-glance judgments.
• Success Criterion: Faster relation search and higher accuracy than a slope-encoded version.

Break it when: The task is not pairwise relation perception, or the slope display is organized differently from the row-based arrangement tested here. Why: The source limits this finding to the tested tasks and explicitly suggests that the row arrangement may be part of why slope performed poorly.

Sacrifice: You give up a slope-based visual form.
Risk: Changing slope to position alone does not remove the larger cost of inferring relations from two separate values.
Mitigation: When relation perception is central, combine position encoding with direct delta encoding.

Mistake: Using a row of angled line segments for increase-decrease judgments and expecting fast search or reliable majority judgments. Why it fails: In the tested row layout, slope marks were harder to parse than position marks.

Failure Sign: Reviewers hesitate or disagree when asked where the lone opposite relation is or which direction appears more often in a row of slopes.
Quick Check: Redraw the same pairs with baseline-aligned positions and compare performance on the same question.
Stronger Test: Run a brief search or majority-direction task on both versions and keep the version that yields faster or more accurate judgments.

• Replace row-aligned slope marks with marks whose value is encoded by position from a common baseline.
• If the chart must preserve both values in each pair, show the two values as separate baseline-aligned positions rather than as angled segments.
• If the chart can abstract the comparison, show the pairwise difference as a single delta-position mark instead of a slope.