Skip to content
Learnings
AutoXXS (320px)XS (375px)SM (640px)MD (768px)LG (1024px)XL (1280px)XXL (1536px)
SketchMaterialiOSTamagui
DataInjectionKeyPatternsServiceTransactionProcessResearchProductQualityPerformanceSpecDomainFunctionTechnologyArchitectureConfigMiddlewareDataDatabaseDrizzleMigrationModelop-sqliteSchemaSQLState ManagementDraftKeystoneMergePatchPatchesPersistenceReactiveRedoStoreUndoTestingDeviceFactoryIsolationTypeScriptZodTopicsCommunicationBidsNVCDesignDesign ImplicationsEducationPedagogyFoundationsPsychologyAttachmentFloodingRelatingAuthentic RelatingUIEditorReact Native

Learnings

ResearchCommunicationDesignDesign ImplicationsAttachmentFloodingAuthentic Relating
Learning Message Volume Triggers Flooding 2026-05-03 The Science of Couples Communication
AttachmentFlooding

Physiological flooding correlates strongly with message length and frequency rather than content valence alone. A rapid sequence of long messages — even neutral ones — can trigger withdrawal in avoidant-attachment users. Our "typing indicator" design may inadvertently amplify this by signalling incoming volume.

Learning Autonomy-Preserving Framing 2026-05-01 The Science of Couples Communication
CommunicationDesignResearch

Early prototype testing revealed that users reject any intervention they perceive as "telling them how to talk." Framing matters enormously — "Would you like to pause?" works; "You should take a break" does not. Autonomy-preserving language is non-negotiable.

Learning Sentence Stems as Design Pattern Authentic Relating Games
Authentic RelatingDesign Implications

The most powerful design pattern in authentic relating is the sentence stem: a partial structure that channels expression without controlling content. This maps directly to guided conversation interfaces where prompts shape without constraining.