Evidence Policy
We use research, lived experience, and product feedback for different purposes, and explain which supports each claim.
Evidence grades
- Standard
- Authoritative accessibility, security, privacy, or platform requirements.
- Strong support
- Multiple high-quality, applicable sources with reasonably consistent findings.
- Limited support
- Early, indirect, small, mixed, or context-dependent evidence.
- Contested framework
- Material scientific disagreement or limited validation.
- Product hypothesis
- A testable design or business assumption, not a scientific fact.
- Brand choice
- An aesthetic, narrative, or positioning decision.
Visual styling and autonomic regulation
Our approach treats color, light, contrast, spacing, texture, imagery, and motion as sensory conditions that can shape an experience. Responses vary by person and context. We design for readability, choice, and changing capacity; visual preferences are not measurements of physiology or safety.
How evidence is selected
Reviews consider population, setting, intervention or concept, outcome, time horizon, limitations, conflicting findings, and relevance to the specific claim. We do not convert a general finding into proof that a particular product feature works.
Claims and uncertainty
We avoid unqualified labels such as science-backed. Claims must match the strength and applicability of the evidence. Contested concepts may be discussed as context, but they are not presented as settled biological fact or used to diagnose a user.
Review and change
Evidence records include a reviewer, review date, permitted claim, limitations, and next review date. Material new evidence can result in revised language, changed product logic, correction, or withdrawal.