Patterns are based on scientific sources.
| Name | Levels of Severity |
| Sources | (Egelman, 2009) |
| Synonyms | None |
| Context | In the case of critical warnings, the description text in the warning should at a minimum explain why the user is seeing the warning and what the possible consequences of ignoring the warning are. |
| Problem | Habituation occurs when similar-looking warnings are used for varying threat levels. |
| Solution | Thresholds should be drawn for threat levels, such that warnings for differing threat levels should be distinguishable from each other. |
| Examples | None |
| Implementation | Warnings should be designed based on their risk level. This risk level should be determined based on the likelihood of the danger, the damage that may be caused to the user by ignoring the warning, and the likelihood that the warning may be triggered in error (due to a false positive). Too many differing discrete risk levels may overwhelm users, causing them to suffer from “warning overload”. |
| Consequences | If a user encounters a particular warning during a relatively low-risk situation, she may choose to disregard this warning due to the low risk level. If she encounters a warning that looks very similar during a high-risk situation, she may disregard this warning because it was confused with the low-risk situation. |
| Dependencies | None |
| Relationships | [Distinguish Security Levels] |
| Principles | [Understandability] |
| Guidelines | None |
| Check lists | None |
| Use cases | None |
| Tags | Levels of Severity, Distinguish Security Levels, Warnings, Critical Warnings, Self-descriptiveness, Fail Safety |
| Log history | [12/21/2015]: Added to repository |
Egelman, S., 2009. Trust me: Design patterns for constructing trustworthy trust indicators. ProQuest.