Tool Attention Compare
Category: Visual Testing Β· Tier: All
Compare attention patterns between two personas on the same page. Shows where they look differently and the Wasserstein divergence between their saliency maps.
When to Use
Visual testing in CBrowser uses Wasserstein optimal transport theory. Smart baselines compute the barycenter (optimal average) of multiple captures. Transport maps show where visual content moved. Attention analysis models persona-specific saliency.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| url | string | Yes | URL to analyze |
| personaA | string | Yes | First persona |
| personaB | string | Yes | Second persona |
Example
Use attention_compare on your target page
Tips
- Visual testing uses Wasserstein optimal transport for robust comparison
- Smart baselines handle non-deterministic rendering automatically
- Transport maps show WHERE content moved, not just THAT it changed
Research
- Klein & Frintrop (DAGM 2012) β Wβ saliency on CIE-Lab distributions
- Agueh & Carlier (SIAM 2011) β Barycenters in Wasserstein space
- Taylor & Fiebach (2025) β OT predicts neural activity at <225ms
Related
- Tools Overview β All 120 tools by category
- Visual & Performance Tools β Full visual testing docs