Use a multi-view slope summary for task classification

Use a multi-view summary instead of a single slope-only view when the figure is meant to classify or diagnose search tasks. For example, pair mean target-present and target-absent slopes with a target-absent-versus-target-present scatter or a slope-ratio panel rather than showing one average slope alone.

A single slope view hides the fact that task classes overlap strongly on target-present slope while still differing in how target-absent trials behave. Adding a second view of the present-absent relationship exposes those task-specific differences.

Mechanism: A multi-view figure lets readers see both overall efficiency and the structure of unsuccessful-search termination, so they are less likely to mistake one slope value for a full task diagnosis.

Evidence: The paper shows that overall slope distributions are unimodal and overlapping, that different task types have different mean slopes, and that tasks with similar target-present slopes can still differ systematically in target-absent slopes and slope ratios (Wolfe, 1998).

Notes: The paper explicitly argues that slopes and ratios together can support diagnostic comparisons better than slope magnitude alone.

• User Goal: Classify a search task or compare how known task classes differ.
• Task: Distinguish whether a task behaves more like a feature, conjunction, or spatial-configuration search.
• Data: Target-present and target-absent slopes are both available across many observations or conditions.
• Chart Setting: A results figure or summary panel with room for more than one view.
• Audience: Researchers or analysts interpreting visual-search behavior.
• Success Criterion: Readers can see both slope overlap and task differences in target-absent behavior.

Break it when: The figure only needs to show that one search is easier or harder than another, without claiming a task category or a serial-parallel status. Why: The paper states that slope differences still convey search efficiency even when they cannot support categorical classification.

Sacrifice: A multi-view summary uses more space than one slope panel. Risk: The added diagnostic can still mislead if it relies on unstable raw ratios from very small target-present slopes. Mitigation: Transform or restrict ratio views before adding them to the figure.

Mistake: Use one slope panel as the entire argument for task classification. Why it fails: Tasks can overlap in target-present slope yet differ in target-absent slope and slope ratio.

Failure Sign: The figure supports a classification claim even though it only shows slope magnitude. Quick Check: Ask whether the reader can see both target-present and target-absent behavior in the figure. Stronger Test: Compare the single-view conclusion with a two-view version; if the added view changes the interpretation of task class, keep the multi-view design.

• Add a scatter of target-absent slope versus target-present slope next to the main slope summary.
• Add a slope-ratio panel grouped by target-present slope range when task diagnosis depends on unsuccessful-search behavior.
• Keep a distribution view beside any mean-slope summary when you compare multiple task classes.