Visual Search Paradigm Testing

Description

Measures response time to find a target as a function of the number of distractors, revealing whether a design element pops out or requires serial search.

In a visual search experiment, participants find a target among distractors while set size varies. If response time does not increase with set size (slope near 0 ms per item), the target pops out through parallel feature processing; slopes of 20-40 ms per item on target-present trials suggest serial, effortful search. Target-absent trials show how long users take to give up. Applied to UX, these slopes quantify the findability of icons, badges, alerts and status markers under realistic clutter, and test whether a design element truly guides attention as intended. It is more diagnostic than eye-tracking heatmaps because it isolates the attentional mechanism.

Attributes

Loss type
Unmigrated
Stage
Design
Effort
Days

Steps

  1. Define the target element and distractors from the real interface (e.g. an error badge among other badges).
  2. Create displays at several set sizes (e.g. 4, 8, 16, 32 items), with target present on half the trials.
  3. Run participants on many trials (e.g. 200+) with randomised positions, recording response time and accuracy.
  4. Fit linear regressions of RT on set size for present and absent trials; compute slopes and intercepts.
  5. Compare design variants: a good salient design shows near-flat slopes and low error rates.
  6. Confirm in the full interface context with a smaller task-based test.

Sources

  • Treisman & Gelade, A feature-integration theory of attention (1980), Cognitive Psychology
  • Wolfe, Guided Search 2.0 (1994), Psychonomic Bulletin & Review
  • Wolfe & Horowitz, Five factors that guide attention in visual search (2017), Nature Human Behaviour

Also known as

  • Visual search experiment
  • Feature integration testing
  • Set-size slope analysis

Categories

  • User Research
  • Cognitive Psychology & Perception